WO2024208167A1 - 信息处理方法、信息处理装置、终端及网络侧设备 - Google Patents
信息处理方法、信息处理装置、终端及网络侧设备 Download PDFInfo
- Publication number
- WO2024208167A1 WO2024208167A1 PCT/CN2024/085422 CN2024085422W WO2024208167A1 WO 2024208167 A1 WO2024208167 A1 WO 2024208167A1 CN 2024085422 W CN2024085422 W CN 2024085422W WO 2024208167 A1 WO2024208167 A1 WO 2024208167A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- target
- model
- distance
- measurement information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/18—Network design, e.g. design based on topological or interconnect aspects of utility systems, piping, heating ventilation air conditioning [HVAC] or cabling
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Definitions
- the present application belongs to the field of communication technology, and specifically relates to an information processing method, an information processing device, a terminal and a network side device.
- a machine learning (ML) model can be used to locate the terminal.
- factors such as the displacement of the terminal and changes in the communication environment in which the terminal is located will reduce the accuracy of the positioning results obtained by the ML positioning model. Therefore, in the related art, the positioning method based on the ML model has the defect of low positioning reliability.
- the embodiments of the present application provide an information processing method, an information processing device, a terminal and a network-side device, which can monitor the accuracy of the output information of a first model, so as to promptly determine that the first model is invalid when the accuracy of the output information of the first model is low, thereby reducing the risk of low positioning reliability caused by continuing to position the terminal using the invalid first model.
- an information processing method which is executed by a terminal, and the method includes:
- the terminal inputs the target measurement information into a first model to obtain first target information output by the first model, where the first target information is related to the location of the terminal;
- the terminal obtains a supervision result of the first model, wherein the supervision result of the first model is determined based on an association relationship between the first target information and second target information, the second target information is information output by the second model after the target measurement information is input into the second model, and the second target information is related to the location of the terminal.
- an information processing method which is performed by a first node, and the method includes:
- the first node receives first information from the terminal, wherein the first information includes target measurement information, or the first information includes first target information and target measurement information, the first target information is information output by the first model after the target measurement information is input into the first model, and the first target information is related to the location of the terminal;
- the first node inputs the target measurement information into a second model to obtain second target information output by the second model, where the second target information is related to the location of the terminal;
- the first node sends at least one of the second information and the second target information to the terminal, wherein the supervision result of the first model is determined based on the association relationship between the first target information and the second target information,
- the second information is related to the supervision result or effectiveness of the first model.
- an information processing device including:
- a first processing module configured to input target measurement information into a first model to obtain first target information output by the first model, where the first target information is related to a location of a terminal;
- An acquisition module is used to obtain the supervision result of the first model, wherein the supervision result of the first model is determined based on the association relationship between the first target information and second target information, the second target information is information output by the second model after the target measurement information is input into the second model, and the second target information is related to the location of the terminal.
- an information processing device comprising:
- a first receiving module configured to receive first information from a terminal, wherein the first information includes target measurement information, or the first information includes first target information and target measurement information, the first target information is information output by the first model after the target measurement information is input into a first model, and the first target information is related to a location of the terminal;
- a second processing module configured to input the target measurement information into a second model to obtain second target information output by the second model, where the second target information is related to the location of the terminal;
- a first sending module is used to send at least one of the second information and the second target information to the terminal, wherein the supervision result of the first model is determined based on the association relationship between the first target information and the second target information, and the second information is related to the supervision result or validity of the first model.
- a terminal comprising a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented.
- a terminal comprising a processor and a communication interface, wherein the processor is used for the terminal to input target measurement information into a first model to obtain first target information output by the first model, wherein the first target information is related to the location of the terminal; the processor or the communication interface is used to obtain a supervision result of the first model, wherein the supervision result of the first model is determined based on an association relationship between the first target information and second target information, the second target information is information output by the second model after the target measurement information is input into the second model, and the second target information is related to the location of the terminal.
- a first node comprising a processor and a memory, wherein the memory stores a program or instruction that can be executed on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the second aspect are implemented.
- a first node comprising a processor and a communication interface, wherein the communication interface is used to receive first information from a terminal, wherein the first information includes target measurement information, or the first information includes first target information and target measurement information, the first target information is information output by the first model after the target measurement information is input into a first model, and the first target information is related to the location of the terminal; the processor is used to input the target measurement information into a second model to obtain second target information output by the second model, and the first target information is related to the location of the terminal; The second target information is related to the location of the terminal; the communication interface is also used to send the second information and at least one of the second target information to the terminal, wherein the supervision result of the first model is determined based on the association relationship between the first target information and the second target information, and the second information is related to the supervision result or validity of the first model.
- a readable storage medium on which a program or instruction is stored.
- the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the second aspect are implemented.
- a wireless communication system including: a terminal and a first node, the terminal can be used to execute the steps of the method described in the first aspect, and the first node can be used to execute the steps of the method described in the second aspect.
- a chip comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the method described in the first aspect, or to implement the method described in the second aspect.
- a computer program/program product is provided, wherein the computer program/program product is stored in a storage medium, and the program/program product is executed by at least one processor to implement the steps of the information processing method as described in the first aspect, or to implement the steps of the information processing method as described in the second aspect.
- the same target measurement information can be input into the first model and the second model respectively to obtain the first target information output by the first model and the second target information output by the second model.
- the first model when the first model is valid, there is a certain correlation between the first target information and the second target information.
- the supervision result of the first model can be determined. For example: if the first target information and the second target information correspond to the same terminal location information, the first model is determined to be valid. If the first target information and the second target information correspond to different terminal location information, the first model is determined to be invalid.
- FIG1 is a schematic diagram of the structure of a wireless communication system to which an embodiment of the present application can be applied;
- FIG. 2 is a schematic diagram of a structure of an information processing method provided in an embodiment of the present application
- FIG3 is a schematic diagram of the architecture of a neural network model
- Fig. 4 is a schematic diagram of a neuron
- FIG5a is a flow chart of a model supervision scheme 1 in an embodiment of the present application.
- FIG5 b is a flow chart of a second model supervision scheme in an embodiment of the present application.
- FIG5c is a flow chart of model supervision scheme 3 in an embodiment of the present application.
- FIG5d is a graph of the empirical cumulative distribution (Empirical CDF) function of the third distance in an embodiment of the present application.
- FIG6 is a flow chart of an information processing method provided in an embodiment of the present application.
- FIG7 is a schematic diagram of the structure of an information processing device provided in an embodiment of the present application.
- FIG8 is a schematic diagram of the structure of an information processing device provided in an embodiment of the present application.
- FIG9 is a schematic diagram of the structure of a communication device provided in an embodiment of the present application.
- FIG10 is a schematic diagram of the structure of a terminal provided in an embodiment of the present application.
- FIG11 is a schematic diagram of the structure of a network side device provided in an embodiment of the present application.
- FIG. 12 is a schematic diagram of the structure of another network-side device provided in an embodiment of the present application.
- first, second, etc. of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by “first” and “second” are generally of one type, and the number of objects is not limited, for example, the first object can be one or more.
- “or” in the present application represents at least one of the connected objects.
- “A or B” covers three schemes, namely, Scheme 1: including A but not including B; Scheme 2: including B but not including A; Scheme 3: including both A and B.
- the character "/" generally indicates that the objects associated with each other are in an "or” relationship.
- indication in this application can be either a direct indication (or explicit indication) or an indirect indication (or implicit indication).
- a direct indication can be understood as the sender explicitly informing the receiver of specific information, operations to be performed, or request results in the sent indication;
- an indirect indication can be understood as the receiver determining the corresponding information according to the indication sent by the sender, or making a judgment and determining the operations to be performed or request results according to the judgment result.
- LTE Long Term Evolution
- LTE-A Long Term Evolution
- CDMA Code Division Multiple Access
- TDMA Time Division Multiple Access
- FDMA Frequency Division Multiple Access
- OFDMA Orthogonal Frequency Division Multiple Access
- SC-FDMA Single-carrier Frequency Division Multiple Access
- NR New Radio
- 6G 6th Generation
- FIG1 is a block diagram of a wireless communication system applicable to the embodiments of the present application.
- the wireless communication system includes a terminal 11 and a network side device 12.
- the terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a palm computer, a netbook, an ultra-mobile personal computer (Ultra-mobile Personal Computer, UMPC), a mobile Mobile Internet Device (MID), Augmented Reality (AR), Virtual Reality (VR) equipment, robots, wearable devices (Wearable Device), flight vehicles (flight vehicles), vehicle-mounted equipment (VUE), ship-mounted equipment, pedestrian terminals (PUE), smart homes (home appliances with wireless communication functions, such as refrigerators, televisions, washing machines or furniture, etc.), game consoles, personal computers (PC), ATMs or self-service machines and other terminal-side devices.
- MID mobile Mobile Internet Device
- AR Augmented Reality
- VR Virtual Reality
- VA
- Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc.
- the vehicle-mounted equipment can also be called a vehicle-mounted terminal, a vehicle-mounted controller, a vehicle-mounted module, a vehicle-mounted component, a vehicle-mounted chip or a vehicle-mounted unit, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiment of the present application.
- the network side device 12 may include an access network device or a core network device, wherein the access network device may also be referred to as a radio access network (RAN) device, a radio access network function or a radio access network unit.
- the access network device may include a base station, a wireless local area network (WLAN) access point (AS) or a wireless fidelity (WiFi) node, etc.
- WLAN wireless local area network
- AS wireless local area network
- WiFi wireless fidelity
- the base station may be referred to as a Node B (NB), an evolved Node B (eNB), a next generation Node B (gNB), a New Radio Node B (NR Node B), an access point, a Relay Base Station (RBS), a Serving Base Station (SBS), a Base Transceiver Station (BTS), a radio base station, a radio transceiver, a Basic Service Set (BSS), an Extended Service Set (ESS), a Home Node B (HNB), a Home Evolved Node B, a Transmission Reception Point (TRP) or other appropriate terms in the field.
- NB Node B
- eNB evolved Node B
- gNB next generation Node B
- NR Node B New Radio Node B
- an access point a Relay Base Station
- SBS Serving Base Station
- BTS Base Transceiver Station
- a radio base station a radio transceiver
- BSS Basic Service Set
- ESS Extended Service Set
- HNB Home No
- the core network device may include but is not limited to at least one of the following: core network node, core network function, location management function (Location Management Function, LMF), mobility management entity (Mobility Management Entity, MME), access mobility management function (Access and Mobility Management Function, AMF), session management function (Session Management Function, SMF), user plane function (User Plane Function, UPF), policy control function (Policy Control Function, PCF), policy and charging rules function unit (Policy and Charging Rules Function, PCRF), edge application service discovery function (Edge Application Server Discovery Function, EASDF), unified data management (Unified Data Management, UDM), unified data storage (Unified Data Repository, UDR), home user server (Home Subscriber Server, HSS), centralized network configuration (CNC), network storage function (Network Repository Function, NRF), network exposure function (Network Exposure Function, NEF), local NEF (Local NEF, or L-NEF), Binding Support Function (BSF), Application Function (AF), etc.
- an information processing method provided in an embodiment of the present application, the execution subject of which may be a terminal, and the terminal may be a terminal 11 of various types listed in FIG1 , or other terminals other than the terminal types listed in the embodiment shown in FIG1 , which are not specifically limited here.
- the information processing method may include the following steps:
- Step 201 The terminal inputs target measurement information into a first model to obtain first target information output by the first model, where the first target information is related to the location of the terminal.
- Step 202 The terminal obtains a supervision result of the first model, wherein the supervision result of the first model is determined based on an association relationship between the first target information and second target information, the second target information is information output by the second model after the target measurement information is input into the second model, and the second target information is related to the location of the terminal.
- the target measurement information may be information obtained by the terminal measuring a reference signal.
- the reference signal may include at least one of the following:
- CSI-RS CSI Reference Signal
- SRS Sounding Reference Signal
- SSB Synchronization Signal and PBCH block
- PRS Positioning Reference Signal
- TRS Tracking Reference Signal
- PTRS Phase-Tracking Reference Signal
- the above-mentioned target measurement information may include at least one of time domain channel impulse response (Channel Impulse Response, CIR), delay power spectrum (Power Delay Profile, PDP), channel frequency response, channel energy response, reference signal received power (Reference Signal Received Power, RSRP), reference signal received path power (Reference Signal Received Path Power, RSRPP), reference signal received quality (Reference Signal Received Quality, RSRQ), signal to interference plus noise ratio (Signal to Interference plus Noise Ratio, SINR), and delay Doppler domain channel.
- CIR Channel Impulse Response
- PDP Power Delay Profile
- RSRP Reference Signal Received Power
- RSRPP reference signal received path power
- RSRQ Reference Signal Received Quality
- SINR Signal to interference plus Noise Ratio
- the above-mentioned measurement information such as channel frequency response, channel energy response, RSRP, RSRPP, RSRQ and SINR can be layer 1 (Layer 1, L1) measurement information, that is, real-time measurement information; or these measurement information can be layer 3 (Layer 3, L3) measurement information, that is, measurement information after smoothing and filtering the real-time measurement information and historical measurement information.
- the target measurement information includes T TRPs or cell-associated measurement information, where T is an integer greater than or equal to 1.
- the reference signal used to measure and obtain the target measurement information may be sent by one or at least two TRPs or cells.
- the target reference signal used to estimate and obtain the target measurement information includes M TRPs or cell-associated reference signals, where M is an integer greater than or equal to 1.
- the target measurement information may be preset, such as configured on the network side.
- the first target information is related to the location of the terminal, and the first target information may include the location information of the terminal, or the first target information includes feature information related to the location of the terminal.
- the location information of the terminal includes at least one of the following:
- the relative position information of the terminal is the relative position information of the terminal.
- the absolute position information may include the coordinate position information of the terminal, such as Global Positioning System (GPS) positioning information.
- GPS Global Positioning System
- the relative position information may be the position information of the terminal relative to a reference point, such as the position information of the terminal relative to a base station.
- the target model may be used to output feature information related to the terminal location.
- the characteristic information related to the location of the terminal may include at least one of the following:
- LOS Line of Sight
- TOA path arrival time
- LOS Line of Sight
- AOA Angle of Arrival
- LOS Line of Sight
- RSTD path reference signal time difference
- PMI Precoding Matrix Indicator
- the above-mentioned LOS path can be the actual LOS path between the terminal and the network side device, or it can be the hypothetical LOS path between the terminal and the network side device. That is to say, no matter whether there is a LOS path between the user equipment (User Equipment, UE) and the base station in actual circumstances, the TOA, AOA, AOD, RSTD, etc. of the LOS path between the UE and the base station can be determined based on the embodiments of the present application, and then the UE's location information can be determined accordingly.
- User Equipment User Equipment
- the TOA, AOA, AOD, RSTD of the above-mentioned LOS path, as well as beam quality, channel compression indication, PMI, time domain channel information, frequency domain channel information, spatial domain channel information, cell communication quality, cell switching judgment results, etc. are all information closely related to the position of the terminal. That is, when the position of the terminal changes, the TOA, AOA, AOD, RSTD of the above-mentioned LOS path, as well as beam quality, channel compression indication, PMI, time domain channel information, frequency domain channel information, spatial domain channel information, cell communication quality, cell switching judgment results, etc. will also change accordingly. For example: the farther the terminal is from the base station, the worse the beam quality will be, the transmission delay of the reference signal will increase, the cell communication quality will deteriorate, and even trigger cell switching.
- the second model may be a supervisory model of the first model, used to supervise the accuracy of the first model or to supervise whether the first model is effective.
- the second model may be a model corresponding to a different space than the first model, such as the first model corresponding to The physical space, the second model corresponds to the latent space, such as the feature space.
- the second model can be used to map CIR to another latent space, and the relative position relationship d2 of the UE in the latent space is consistent with the relative position relationship d1 of the UE in the physical space.
- the first model if the first model is valid, the relative positions of the CIRs of the two UEs in the latent space and the physical space should be consistent; if the first model fails, the relative positions of the CIRs of the two UEs in the latent space and the physical space are no longer consistent.
- the physical space may be understood as the space corresponding to the position coordinates of the UE, such as the position coordinates of the UE may be regarded as a point in the physical space.
- g is the first model to be supervised, that is, the mapping from CIR to physical space
- f is the second model, that is, the mapping from CIR to latent space
- d1 represents the first distance
- d2 represents the second distance
- represents the third distance
- ⁇ represents the third threshold.
- the second target information is related to the location of the terminal.
- the second target information may include the location information of the terminal, or the second target information may include distance information between at least two terminal locations, for example, distance information between the locations of the same terminal at two different times, or distance information between the locations of different terminals.
- the second target information may be relative position information of the terminal, and a reference position of the second target information is different from a reference position of the first target information.
- the second target information includes characteristic information related to the location of the terminal, or the second target information includes the distance or difference between characteristic information related to the locations of at least two terminals of the terminal.
- first target information and the second target information are different types of information, such as the first target information is the location information of the terminal, and the second target information is the distance information between the locations of the terminal at different times, or even if the first target information and the second target information correspond to different spaces, such as the first target information is the location information of the terminal in the physical space, and the second target information is the location information of the terminal in the latent space, given that the first target information and the second target information are respectively related to the location of the terminal, there is still a certain correlation between the first target information and the second target information, such as the first distance between the locations of the terminal at different times in the physical space determined based on the first target information, and the second distance between the locations of the terminal at different times in the latent space determined based on the second target information can be equal, or linearly related.
- the model supervision results such as the accuracy of the first target information output by the first target or whether the first model is effective can be determined.
- the first target information is usually the terminal position in the physical space
- the second target information is the terminal position in the latent space
- the terminal distance is used as an example for illustration, which does not constitute a specific limitation here.
- first model and the second model in the implementation of the present application can be an artificial intelligence (AI) model or a machine learning (ML) model.
- AI artificial intelligence
- ML machine learning
- the first model and the second model are AI models as an example in the embodiments of the present application.
- AI modules can be implemented in many ways, such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc. This application uses neural networks as an example for illustration, but does not limit the specific type of AI modules.
- the neural network includes an input layer, a hidden layer, and an output layer, which can predict possible output results (Y) based on the input and output information (X 1 ⁇ X n ) obtained from the input layer.
- K the total number of input parameters.
- the parameters of the neural network are optimized through optimization algorithms.
- An optimization algorithm is a type of algorithm that can help us minimize or maximize an objective function (sometimes called a loss function).
- the objective function is often a mathematical combination of model parameters and data. For example, given data X and its corresponding label Y, we build a neural network f(.). With the model neural network, we can get the predicted output f(x) based on the input x, and we can calculate the difference between the predicted value and the true value (f(x)-Y), which is the loss function. Our goal is to find the right W and b to minimize the value of the above loss function. The smaller the loss value, the closer our neural network is to the actual situation.
- the common optimization algorithms are basically based on the error back propagation (BP) algorithm.
- the basic idea of the error back propagation algorithm is that the learning process consists of two processes: the forward propagation of the signal and the back propagation of the error.
- the input sample is transmitted from the input layer, processed by each hidden layer layer by layer, and then transmitted to the output layer. If the actual output of the output layer does not match the expected output, it will enter the error back propagation stage.
- Error back propagation is to propagate the output error layer by layer through the hidden layer to the input layer in some form, and distribute the error to all units in each layer, so as to obtain the error signal of each layer unit, and this error signal is used as the basis for correcting the weights of each unit.
- This process of adjusting the weights of each layer of the signal forward propagation and error back propagation is repeated.
- the process of continuous adjustment of weights is the learning and training process of the network. This process continues until the error of the network output is reduced to an acceptable level, or until the pre-set number of learning times is reached.
- Common optimization algorithms include gradient descent, stochastic gradient descent (SGD), mini-batch gradient descent, momentum method (Momentum), Nesterov (which means stochastic gradient descent with momentum), adaptive gradient descent (Adaptive gradient descent, Adagrad), adaptive learning rate adjustment (Adadelta), root mean square error speed reduction (root mean square prop, RMSprop), adaptive momentum estimation (Adaptive Moment Estimation, Adam), etc.
- the above-mentioned second model can be deployed in the terminal, or it may be deployed in the first node, and the first node may be a network-side device, such as an access network device or a core network device, or the first node may be any device with a target model, such as an application server.
- the first node is usually an access network device, such as a base station, for example, which is not specifically limited here.
- the above-mentioned second target information needs to input the target measurement information into the second model to obtain, and the supervision result of the above-mentioned first model needs to be determined based on the association relationship between the first target information and the second target information.
- the terminal may obtain the supervision result of the first model by processing the target measurement information based on its own second model to obtain the second target information, and compare the first target information and the second target information obtained based on the same target measurement information to determine the supervision result of the first model based on the correlation between the two.
- the terminal may obtain the supervision result of the first model in the following manner: the terminal sends target measurement information to a first node having a second model, and the first node processes the target measurement information using the second model to obtain second target information, and feeds back the second target information to the terminal. Thereafter, the terminal may compare the first target information and the second target information obtained based on the same target measurement information to determine the supervision result of the first model based on the correlation between the two.
- the terminal may obtain the supervision result of the first model in such a manner that: the terminal sends target measurement information and first target information to a first node having a second model, and the first node processes the target measurement information using the second model to obtain the second target information, and after comparing the first target information and the second target information obtained based on the same target measurement information to determine the supervision result of the first model according to a correlation between the two, the second information is fed back to the terminal to indicate the supervision result of the first model through the second information, such as the second information indicating that the first model is valid or invalid, or indicating that the first model is valid in some scenarios and invalid in other scenarios, etc.
- the first target information and the second target information in the embodiment of the present application are obtained based on the same target measurement information, that is, the same target measurement information is input into the first model and the second model respectively, then the information output by the first model is the first target information, and the information output by the second model is the second target information.
- the terminal location information or the feature information related to the location corresponding to the same target measurement information should be the same or can be related. Based on this, the supervision result of the first model can be determined based on the association relationship between the first target information and the second target information, such as whether the first model is valid or invalid.
- the supervision result of the first model determined based on the association relationship between the first target information and the second target information in the embodiment of the present application may be determined based on the association relationship between the first target information and the second target information corresponding to the same target measurement information.
- the terminal when the terminal has the first model and the first node has the second model, the terminal obtains the supervision result of the first model, including:
- the terminal sends first information to the first node, where the first information includes the target measurement information, or the first information includes the first target information and the target measurement information;
- the terminal receives second target information from the first node and determines the supervision result of the first model based on the association relationship between the first target information and the second target information, or the terminal receives second information from the first node, and the second information is related to the supervision result or validity of the first model.
- the terminal sends first target information and the target measurement information to the first node, and the first node inputs the target measurement information into a second model to obtain second target information. Then, the first node determines the supervision result of the first model based on the association relationship between the first target information and the second target information, that is, the first node determines the supervision result of the first model and indicates the supervision result or validity-related information of the first model to the terminal through the second information.
- the terminal sends the target measurement information to the first node, the first node inputs the target measurement information into the second model to obtain second target information, and then the first node sends the second target information to the terminal. Finally, the terminal determines the supervision result of the first model based on the association relationship between the first target information and the second target information, that is, the supervision result of the first model is determined by the terminal.
- the terminal when the terminal has the first model and the second model, the terminal obtains the supervision result of the first model, including:
- the terminal inputs the target measurement information into the second model to obtain the second target information output by the second model;
- the terminal determines a supervision result of the first model according to an association relationship between the first target information and the second target information.
- the terminal has both the first model and the second model. Therefore, the terminal can use the first model to obtain the first target information, use the second model to obtain the second target information, and compare the first target information and the second target information corresponding to the same target measurement information to determine the supervision result of the first model based on the correlation between the first target information and the second target information.
- the method further includes:
- the terminal sends third information to the first node, where the third information is related to the supervision result or validity of the first model.
- the terminal reports the supervision results or validity-related information of the first model to the first node, so that the first node can decide whether to use the data output by the first model as positioning information.
- the first node uses other methods (such as non-AI methods) to determine the location information of the terminal; or the first node decides whether to update the first model.
- the first model fails, the first node trains and updates the first model, and sends the updated first model to the terminal; or, the first node decides whether to deactivate the current first model and activate or select a new first model.
- the first node deactivates the currently used first model and activates or selects a new first model.
- the terminal determines the supervision result of the first model according to the association relationship between the first target information and the second target information, including:
- the terminal obtains a first distance between a group of first target information and obtains a second distance between a group of second target information, wherein the group of first target information and the group of second target information are obtained by processing the same group of target measurement information based on the first model and the second model respectively;
- the terminal determines a supervision result of the first model according to a difference or correlation between the first distance and the second distance.
- a set of target measurement information may include at least two pieces of target measurement information.
- a set of target measurement information includes two pieces of target measurement information.
- a set of target measurement information may include at least two pieces of target measurement information from the same terminal at different times.
- the time may include an Orthogonal Frequency Division Multiplex (OFDM) symbol, a subframe, a frame, a nanosecond, a microsecond, a millisecond, a second, a minute, an hour, a day, a month, etc.
- a set of target measurement information may include target measurement information obtained by the terminal by measuring a reference signal at time 1, and target measurement information obtained by the terminal by measuring a reference signal at time 2.
- a set of target measurement information may include target measurement information from at least two different terminals.
- a set of target measurement information may include target measurement information of terminal A and target measurement information of terminal B.
- a set of target measurement information may include at least two target measurement information from different terminals
- the terminal needs to obtain the target measurement information of other terminals so as to utilize the target measurement information of the terminal and the target measurement information of other terminals to form a set of target measurement information.
- the first node when the first node has the second model, different terminals can respectively send their respective target measurement information to the first node, such as a base station, so that the base station uses the target measurement information of different terminals to form a group of target measurement information.
- the first node in the present application may also be a core network device.
- the first node in the embodiment of the present application is taken as an example that the base station is an example, which does not constitute a specific limitation here.
- the terminal may input target measurement information in a set of target measurement information into the first model respectively to obtain a set of first target information respectively output by the first model, and then calculate the first distance between any two first target information in the set of first target information.
- a set of target measurement information includes target measurement information A of UE 1 and target measurement information B of UE 2
- the terminal inputs the target measurement information A and the target measurement information B into the first model respectively, obtains the location information of UE 1 and the location information of UE 2 respectively output by the first model, and then calculates the first distance between UE 1 and UE 2 based on this.
- a set of target measurement information includes CIRs collected by the same terminal in time slots (slot) 1 and slot 2, respectively
- the terminal inputs the CIR collected in slot 1 and the CIR collected in slot 2 into the first model respectively, obtains the position information of the terminal in slot 1 and the position information of the terminal in slot 2 respectively output by the first model, and then calculates the first distance between the positions of the terminals in slot 1 and slot 2 respectively.
- the terminal may input a set of target measurement information into the second model in parallel to obtain second target information output by the second model, where the second target information is the second distance.
- the terminal inputs the target measurement information into the second model to obtain second target information output by the second model, including:
- the terminal inputs a set of target measurement information into a second model, and obtains second target information output by the second model, wherein the second target information includes:
- the distance between positions corresponding to at least two pieces of target measurement information in the set of target measurement information is a distance between positions corresponding to at least two pieces of target measurement information in the set of target measurement information.
- a set of target measurement information includes target measurement information A of UE 1 and target measurement information B of UE 2
- the terminal inputs target measurement information A and target measurement information B into the second model in parallel to obtain the second distance between UE 1 and UE 2 output by the second model.
- a set of target measurement information includes CIRs collected by the same terminal in slot 1 and slot 2 respectively
- the terminal inputs the CIR collected in slot 1 and the CIR collected in slot 2 into the second model in parallel, and obtains the second distance between the positions of the terminal in slot 1 and slot 2 respectively output by the second model.
- the terminal may input a set of target measurement information into the second model respectively to obtain a set of second target information respectively output by the second model, and then calculate the second distance between any two second target information in the set of second target information.
- a parallel Siamese network may be used to train the second model.
- the twin network includes two positioning models with consistent structures and parameters.
- the distance between the predicted positions output by the two positioning models is obtained as the output result of the twin network, and the output result of the twin network is compared with the distance in the label to obtain the loss function of the twin network, such as the mean absolute error (MAE).
- MAE mean absolute error
- the two target measurement information in the group of target measurement information are respectively input into the first model, and two first target information will be obtained.
- the distance between the two first target information is the first distance; the two target measurement information in the group of target measurement information are input into the second model in parallel, and the second distance will be obtained.
- the first distance includes at least one of the following: Euclidean distance, Manhattan distance, and cosine distance;
- the second distance includes at least one of the following: Euclidean distance, Manhattan distance, and cosine distance.
- Minkowski distance is defined as: given a sample space X, X is a set of points in an m-dimensional real vector space, where x i ,x j ⁇ X, the Minkowski distance between the two is:
- d ij the Euclidean distance
- d ij represents the Chebyshev distance, which takes the maximum absolute value of the difference between the coordinate values, that is,
- the difference between the first distance and the second distance may include: a difference between the first distance and the second distance, or a spatial distance.
- the correlation between the first distance and the second distance may include: the similarity between the first distance and the second distance, such as whether the first distance and the second distance are equal, and whether the first distance can be linearly transformed to obtain the second distance.
- the correlation between the first distance and the second distance may be measured by calculating a correlation coefficient between the first distance and the second distance.
- X is a set of points in an m-dimensional real vector space, where x i ,x j ⁇ X, the correlation coefficient of samples x i ,x j can be expressed as the following formula:
- rij represents the correlation coefficient between xi and xj .
- X is a set of points in an m-dimensional real vector space, where x i ,x j ⁇ X, then the correlation coefficient of samples x i ,x j can be the cosine of the angle between samples x i ,x j , which can be expressed as the following formula:
- s ij represents the cosine of the angle between xi and xj .
- the correlation and distance between the first distance and the second distance in the embodiment of the present application are two different concepts, wherein the distance describes the spatial similarity, while the correlation describes the consistency of the change trend.
- the terminal determines the supervision result of the first model according to the difference or correlation between the first distance and the second distance, which may be: if the first distance is equal to the second distance, or the first distance and the second distance satisfy a preset linear conversion relationship, or the difference between the first distance and the second distance is less than or equal to a certain threshold, or after the first distance and the second distance are converted to the space of the same dimension, the difference between the two is less than or equal to a certain threshold, then the first model can be judged to be valid.
- the difference or correlation between the first distance and the second distance may be: if the first distance is equal to the second distance, or the first distance and the second distance satisfy a preset linear conversion relationship, or the difference between the first distance and the second distance is less than or equal to a certain threshold, or after the first distance and the second distance are converted to the space of the same dimension, the difference between the two is less than or equal to a certain threshold, then the first model can be judged to be valid.
- the first model can be judged to be invalid.
- the terminal determines the supervision result of the first model according to the difference between the first distance and the second distance, including:
- the terminal determines that the first model is invalid, and the first condition includes at least one of the following:
- the difference between the first distance and the second distance is greater than a first threshold
- F groups of first target measurement information wherein a difference between a first distance and a second distance corresponding to the same group of first target measurement information is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is greater than a second threshold
- the third distance between the first distance distribution and the second distance distribution is greater than or equal to the third threshold value, and the first distance distribution is the distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the difference between the first distance and the second distance may be a difference value obtained by directly subtracting the first distance from the second distance, or a distance value obtained by performing a distance comparison after transforming the first distance and the second distance into the same spatial coordinates.
- the first target measurement information represents target measurement information that satisfies the following condition: a difference between a first distance and a second distance corresponding to the same group of target measurement information is greater than a first threshold.
- the difference value between the first distance and the second distance corresponding to the same set of target measurement information is greater than the first threshold, which can reflect that the output results of the first model and the second model are inconsistent.
- the first model may be judged to be invalid if the number of times the output results of the first model and the second model are inconsistent reaches a certain threshold F, or the probability of the output results of the first model and the second model being inconsistent reaches a second threshold.
- the first model and the second model may be used to obtain the first distance and the second distance corresponding to each of all groups of target measurement information or preset R groups of target measurement information measured within the target time range, respectively, and then, based on the third distance between the distribution of the first distance and the distribution of the second distance, it is determined that the first model has failed.
- the third distance includes at least one of the following:
- the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution is the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution.
- line X as shown in FIG5d represents the Empirical CDF function curve of the third distance when the target measurement information is the measurement information measured in the same environment as the first training set;
- line Y as shown in FIG5d represents the Empirical CDF function curve of the third distance when the target measurement information is the measurement information measured in a different environment from the first training set.
- first threshold, second threshold, third threshold, F and other thresholds in the embodiment of the present application can be determined by the terminal and reported to the network side device, or the above thresholds can be indicated by the network side device or by protocol agreement, which will not be repeated here.
- the terminal determines the supervision result of the first model according to the difference between the first distance and the second distance, including:
- the terminal determines that the first model is valid, and the second condition includes at least one of the following:
- the difference between the first distance and the second distance is less than or equal to a first threshold
- the number of groups of first target measurement information is less than or equal to F, wherein the difference between the first distance and the second distance corresponding to the same group of first target measurement information is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is less than or equal to a second threshold
- the third distance between the first distance distribution and the second distance distribution is less than the third threshold value, and the first distance distribution is the distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the present embodiment is similar to the previous embodiment in how to determine the failure of the first model, except that: In this implementation manner, it is used to determine whether the first model is valid.
- the specific explanation can refer to the explanation in the previous implementation manner, which will not be repeated here.
- the supervision result of the first model includes at least one of the following:
- the number of groups of target measurement information that meets the first condition such as the number of groups of first target measurement information
- the proportion of target measurement information that meets the first condition such as the proportion of the first target measurement information in all target measurement information within a specified time range, or the proportion of the first target measurement information in a specified R group of target measurement information;
- Timestamp information such as timestamp information of obtaining the supervision result of the first model, or timestamp information of target measurement information based on which the supervision result of the first model is obtained.
- the method further includes at least one of the following:
- the terminal performs a first preprocessing on the target measurement information, wherein the data input by the first model includes the target measurement information after the first preprocessing;
- the terminal performs a second preprocessing on the target measurement information, wherein the data input by the second model includes the target measurement information after the second preprocessing.
- the first preprocessing and the second preprocessing may be the same preprocessing, such as compression, quantization, truncation, filtering, normalization, etc.
- the first preprocessing and the second preprocessing may reduce the data volume of the target measurement information, such as reducing the number of bits of the target measurement information, or the first preprocessing and the second preprocessing may simplify or change the format of the target measurement information to facilitate input into the first model and the second model.
- the second pre-processed target measurement information may be sent to the first node, thereby reducing air interface resources consumed by transmitting the target measurement information.
- the method further includes at least one of the following:
- the terminal performs a first post-processing on the output information of the first model, wherein the first target information includes information obtained by the first post-processing;
- the terminal performs second post-processing on the output information of the second model, wherein the second target information includes information obtained by the second post-processing.
- the first post-processing and the second post-processing may be the same or similar processing, and the information after the first post-processing and the second post-processing are in the same dimension or in the same space, so that it is easy to obtain the correlation relationship between the first target information and the second target information.
- the first post-processing and the second post-processing are used to convert the position information output by the first model and the position information output by the second model into the global coordinate position. In this way, by comparing whether the two global coordinate positions are consistent, the first Whether the output information of the model is consistent with the output information of the second model.
- the first post-processing and the second post-processing are used to convert the time domain information output by the first model and the time domain information output by the second model into frequency domain information. In this way, by comparing whether the two frequency domain information are consistent, it can be determined whether the output information of the first model is consistent with the output information of the second model.
- the same target measurement information can be input into the first model and the second model respectively to obtain the first target information output by the first model and the second target information output by the second model.
- the first model when the first model is valid, there is a certain correlation between the first target information and the second target information.
- the supervision result of the first model can be determined. For example: if the first target information and the second target information correspond to the same terminal location information, the first model is determined to be valid. If the first target information and the second target information correspond to different terminal location information, the first model is determined to be invalid.
- the first node may be a network side device or a server, wherein the network side device may be various types of network side devices 12 listed in Figure 1, such as an access network device or a core network device, or other network side devices other than the types of network side devices listed in the embodiment shown in Figure 1.
- the first node is usually taken as a base station in the embodiment of the present application for example. This does not constitute a specific limitation.
- the signal processing method may include the following steps:
- Step 601 the first section receives first information from a terminal, wherein the first information includes target measurement information, or the first information includes first target information and target measurement information, the first target information is information output by the first model after the target measurement information is input into a first model, and the first target information is related to the location of the terminal.
- Step 602 The first node inputs the target measurement information into a second model to obtain second target information output by the second model, where the second target information is related to the location of the terminal.
- Step 603 The first node sends at least one of the second information and the second target information to the terminal, wherein the supervision result of the first model is determined based on the association relationship between the first target information and the second target information, and the second information is related to the supervision result or validity of the first model.
- the supervision result of the first model is determined by the first node.
- the method before the first node sends the second information to the terminal, the method further includes:
- the first node determines the supervision result of the first model according to the association relationship between the first target information and the second target information.
- the supervision result of the first model is determined by the terminal.
- the first node interacts with the terminal to jointly implement the model supervision function of the first model.
- the target measurement information includes at least one of the following:
- Time domain channel impulse response channel frequency response, delay power spectrum, channel energy response, reference signal received power RSRP, reference signal received path power RSRPP, reference signal received quality RSRQ, signal to interference plus noise ratio SINR, delay Doppler domain channel.
- the first node determines the supervision result of the first model according to the association relationship between the first target information and the second target information, including:
- the first node obtains a first distance between a group of first target information and obtains a first distance between a group of second target information
- the first target information and the second target information are obtained by processing the same set of target measurement information based on the first model and the second model respectively;
- the first node determines the supervision result of the first model according to the difference between the first distance and the second distance, including:
- the first node determines that the first model fails, and the first condition includes at least one of the following:
- the difference between the first distance and the second distance is greater than a first threshold
- F groups of first target measurement information wherein a difference between a first distance and a second distance corresponding to the same group of first target measurement information is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- the third distance between the first distance distribution and the second distance distribution is greater than or equal to the third threshold value, and the first distance distribution is the distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the first node determines the supervision result of the first model according to the difference between the first distance and the second distance, including:
- the first node determines that the first model is valid, and the second condition includes at least one of the following:
- the difference between the first distance and the second distance is less than or equal to a first threshold
- the number of groups of first target measurement information is less than or equal to F, wherein the difference between the first distance and the second distance corresponding to the same group of first target measurement information is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is less than or equal to a second threshold
- the third distance between the first distance distribution and the second distance distribution is less than the third threshold value, and the first distance distribution is the distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the set of target measurement information includes at least one of the following:
- the first node inputs the target measurement information into the second model to obtain second target information output by the second model, including:
- the first node inputs a set of target measurement information into the second model, and obtains second target information output by the second model, wherein the second target information includes:
- the distance between positions corresponding to at least two pieces of target measurement information in the set of target measurement information is a distance between positions corresponding to at least two pieces of target measurement information in the set of target measurement information.
- the third distance includes at least one of the following:
- the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution is the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution.
- the first distance includes at least one of the following: Euclidean distance, Manhattan distance, and cosine distance;
- the second distance includes at least one of the following: Euclidean distance, Manhattan distance, and cosine distance.
- the supervision result of the first model includes at least one of the following:
- the target measurement information includes T TRPs or cell-associated measurement information, where T is an integer greater than or equal to 1.
- the target reference signal used to estimate the target measurement information includes M TRPs or cell-associated reference signals, where M is an integer greater than or equal to 1.
- the first node in the case where the second model is located at the first node, the first node interacts with the terminal to utilize the second model of the first node to monitor the effectiveness of the first model used by the terminal, which can achieve similar beneficial effects as the method embodiment shown in Figure 2. To avoid repetition, it will not be repeated here.
- the information processing method provided in the embodiment of the present application can be executed by an information processing device.
- the information processing device provided in the embodiment of the present application is described by taking the information processing device executing the information processing method as an example.
- the information processing device 700 provided in the embodiment of the present application may be a device in a terminal.
- the information processing device 700 may include the following modules:
- a first processing module 701 is used to input target measurement information into a first model to obtain first target information output by the first model, where the first target information is related to the location of the terminal;
- the acquisition module 702 is used to obtain the supervision result of the first model, wherein the supervision result of the first model is determined based on the association relationship between the first target information and second target information, the second target information is information output by the second model after the target measurement information is input into the second model, and the second target information is related to the location of the terminal.
- the terminal has the first model and the first node has the second model:
- the acquisition module 702 includes: a first sending unit, a first receiving unit and a first determining unit;
- a first sending unit configured to send first information to the first node, where the first information includes the target measurement information, or the first information includes the first target information and the target measurement information;
- a first receiving unit configured to receive second target information from the first node
- a first determining unit configured to determine a supervision result of the first model according to an association relationship between the first target information and the second target information
- the acquisition module 702 includes: the first sending unit and the second receiving unit;
- the second receiving unit is used to receive second information from the first node, where the second information is related to the supervision result or validity of the first model.
- the acquiring module 702 includes:
- a first processing unit configured to input the target measurement information into the second model to obtain the second target information output by the second model
- the second determining unit is used to determine the supervision result of the first model according to the association relationship between the first target information and the second target information.
- the information processing device 700 further includes:
- the second sending module is used to send third information to the first node, where the third information is related to the supervision result or validity of the first model.
- the target measurement information includes at least one of the following:
- Time domain channel impulse response channel frequency response, delay power spectrum, channel energy response, reference signal received power RSRP, reference signal received path power RSRPP, reference signal received quality RSRQ, signal to interference plus noise ratio SINR, delay Doppler domain channel.
- the first target information includes at least one of the following:
- the first determining unit is specifically configured to:
- a supervision result of the first model is determined according to a difference or correlation between the first distance and the second distance.
- the first determining unit is specifically configured to:
- the first condition includes at least one of the following:
- the difference between the first distance and the second distance is greater than a first threshold
- F groups of first target measurement information wherein a difference between a first distance and a second distance corresponding to the same group of first target measurement information is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is greater than a second threshold
- the third distance between the first distance distribution and the second distance distribution is greater than or equal to the third threshold value, and the first distance distribution is the distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the first determining unit is specifically configured to:
- the first model is determined to be valid, and the second condition includes at least one of the following:
- the difference between the first distance and the second distance is less than or equal to a first threshold
- the number of groups of first target measurement information is less than or equal to F, wherein the difference between the first distance and the second distance corresponding to the same group of first target measurement information is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is less than or equal to a second threshold
- a third distance between the first distance distribution and the second distance distribution is less than a third threshold value, and the first distance distribution is The distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the set of target measurement information includes at least one of the following:
- the first processing unit is specifically configured to:
- a set of target measurement information is input into a second model, and second target information output by the second model is obtained, wherein the second target information includes:
- the distance between positions corresponding to at least two pieces of target measurement information in the set of target measurement information is a distance between positions corresponding to at least two pieces of target measurement information in the set of target measurement information.
- the third distance includes at least one of the following:
- the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution is the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution.
- the supervision result of the first model includes at least one of the following:
- the information processing device 700 further includes at least one of the following:
- a third processing module configured to perform a first preprocessing on the target measurement information, wherein the data input by the first model includes the target measurement information after the first preprocessing;
- the fourth processing module is used to perform a second preprocessing on the target measurement information, wherein the data input by the second model includes the target measurement information after the second preprocessing.
- the information processing device 700 further includes at least one of the following:
- a fifth processing module configured to perform a first post-processing on the output information of the first model, wherein the first target information includes information obtained through the first post-processing;
- a sixth processing module is used to perform a second post-processing on the output information of the second model, wherein the second target information includes information obtained by the second post-processing.
- the target measurement information includes T TRPs or cell-associated measurement information, where T is an integer greater than or equal to 1.
- the target reference signal used to estimate the target measurement information includes M TRPs or cell-associated reference signals, where M is an integer greater than or equal to 1.
- the information processing device 700 in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip.
- the electronic device may be a terminal, or may be other devices other than a terminal.
- the terminal may include but is not limited to the types of terminal 11 listed above, and other devices may be servers, network attached storage (NAS), etc., which are not specifically limited in the embodiment of the present application.
- the information processing device 700 provided in the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 2 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
- the information processing device 800 provided in the embodiment of the present application may be a device within a first node.
- the first node may be a network-side device or other devices other than the network-side device, such as a server.
- the information processing device 800 may include the following modules:
- a first receiving module 801 is configured to receive first information from a terminal, wherein the first information includes target measurement information, or the first information includes first target information and target measurement information, the first target information is information output by the first model after the target measurement information is input into a first model, and the first target information is related to the location of the terminal;
- a second processing module 802 is used to input the target measurement information into a second model to obtain second target information output by the second model, where the second target information is related to the location of the terminal;
- the first sending module 803 is used to send at least one of the second information and the second target information to the terminal, wherein the supervision result of the first model is determined based on the association relationship between the first target information and the second target information, and the second information is related to the supervision result or validity of the first model.
- the information processing device 800 further includes:
- the first determination module is used to determine the supervision result of the first model according to the association relationship between the first target information and the second target information.
- the information processing device 800 further includes:
- the second receiving module is used to receive third information from the terminal, where the third information is related to the supervision result or effectiveness of the first model.
- the target measurement information includes at least one of the following:
- Time domain channel impulse response channel frequency response, delay power spectrum, channel energy response, reference signal received power RSRP, reference signal received path power RSRPP, reference signal received quality RSRQ, signal to interference plus noise ratio SINR, delay Doppler domain channel.
- the first target information includes at least one of the following:
- the first determining module is specifically configured to:
- a supervision result of the first model is determined according to a difference or correlation between the first distance and the second distance.
- the first determining module is specifically configured to:
- the first condition includes at least one of the following:
- the difference between the first distance and the second distance is greater than a first threshold
- F groups of first target measurement information wherein the difference between the first distance and the second distance corresponding to the same group of first target measurement information is greater than the first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is greater than a second threshold
- the third distance between the first distance distribution and the second distance distribution is greater than or equal to the third threshold value, and the first distance distribution is the distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the first determining module is specifically configured to:
- the first model is determined to be valid, and the second condition includes at least one of the following:
- the difference between the first distance and the second distance is less than or equal to a first threshold
- the number of groups of first target measurement information is less than or equal to F, wherein the first distance and the second distance corresponding to the same group of first target measurement information are less than or equal to F.
- the difference value of the distance is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is less than or equal to a second threshold
- the third distance between the first distance distribution and the second distance distribution is less than the third threshold value, and the first distance distribution is the distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the set of target measurement information includes at least one of the following:
- the second processing module 802 is specifically configured to:
- the first node inputs a set of target measurement information into the second model, and obtains second target information output by the second model, wherein the second target information includes:
- the distance between positions corresponding to at least two pieces of target measurement information in the set of target measurement information is a distance between positions corresponding to at least two pieces of target measurement information in the set of target measurement information.
- the third distance includes at least one of the following:
- the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution is the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution.
- the supervision result of the first model includes at least one of the following:
- the target measurement information includes T TRPs or cell-associated measurement information, where T is an integer greater than or equal to 1.
- the target reference signal used to estimate the target measurement information includes M TRPs or cell-associated reference signals, where M is an integer greater than or equal to 1.
- the information processing device 800 provided in the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 6 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
- the embodiment of the present application further provides a communication device 900, including a processor 901 and a memory 902, wherein the memory 902 stores a program or instruction that can be run on the processor 901.
- the communication device 900 is
- the program or instruction is executed by the processor 901 to implement the various steps of the above-mentioned information transmission method embodiment, and can achieve the same technical effect.
- the communication device 900 is a network side device
- the program or instruction is executed by the processor 901 to implement the various steps of the above-mentioned information processing method embodiment, and can achieve the same technical effect. To avoid repetition, it is not repeated here.
- the embodiment of the present application also provides a terminal, including a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the steps in the method embodiment shown in Figure 2.
- This terminal embodiment corresponds to the above-mentioned terminal side method embodiment, and each implementation process and implementation method of the above-mentioned method embodiment can be applied to the terminal embodiment and can achieve the same technical effect.
- Figure 10 is a schematic diagram of the hardware structure of a terminal implementing an embodiment of the present application.
- the terminal 1000 includes but is not limited to: a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009 and at least some of the components of a processor 1010.
- the terminal 1000 may also include a power source (such as a battery) for supplying power to each component, and the power source may be logically connected to the processor 10 10 through a power management system, so as to implement functions such as managing charging, discharging, and power consumption management through the power management system.
- a power source such as a battery
- the terminal structure shown in FIG10 does not constitute a limitation on the terminal, and the terminal may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.
- the input unit 1004 may include a graphics processing unit (GPU) 10041 and a microphone 10042, and the graphics processor 10041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode.
- the display unit 1006 may include a display panel 10061, and the display panel 10061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc.
- the user input unit 1007 includes a touch panel 10071 and at least one of other input devices 10072.
- the touch panel 10071 is also called a touch screen.
- the touch panel 10071 may include two parts: a touch detection device and a touch controller.
- Other input devices 10072 may include, but are not limited to, a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.
- the RF unit 1001 can transmit the data to the processor 1010 for processing; in addition, the RF unit 1001 can send uplink data to the network side device.
- the RF unit 1001 includes but is not limited to an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.
- the memory 1009 can be used to store software programs or instructions and various data.
- the memory 1009 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instruction required for at least one function (such as a sound playback function, an image playback function, etc.), etc.
- the memory 1009 may include a volatile memory or a non-volatile memory.
- the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.
- the volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (SRAM), or a volatile memory.
- RAM random access memory
- SRAM static random access memory
- SRAM dynamic random access memory
- volatile memory volatile memory.
- the memory 1009 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.
- the processor 1010 may include one or more processing units; optionally, the processor 1010 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 1010.
- the processor 1010 is configured to input the target measurement information into a first model to obtain first target information output by the first model, where the first target information is related to the location of the terminal;
- At least one of the processor 1010 and the radio frequency unit 1001 is used to obtain the supervision result of the first model, wherein the supervision result of the first model is determined based on the association relationship between the first target information and the second target information, the second target information is information output by the second model after the target measurement information is input into the second model, and the second target information is related to the location of the terminal.
- the obtaining of the supervision result of the first model performed by the processor 1010 and the radio frequency unit 1001 includes:
- the radio frequency unit 1001 is configured to send first information to the first node, where the first information includes the target measurement information, or the first information includes the first target information and the target measurement information;
- the radio frequency unit 1001 is also used to receive second target information from the first node, and the processor 1010 is used to determine the supervision result of the first model based on the association relationship between the first target information and the second target information.
- the radio frequency unit 1001 is also used to receive second information from the first node, and the second information is related to the supervision result or validity of the first model.
- the obtaining of the supervision result of the first model performed by the processor 1010 and the radio frequency unit 1001 includes:
- Processor 1010 configured to input the target measurement information into the second model to obtain the second target information output by the second model;
- Processor 1010 is further used to determine the supervision result of the first model according to the association relationship between the first target information and the second target information.
- the radio frequency unit 1001 is further used to send third information to the first node, where the third information is related to the supervision result or validity of the first model.
- the target measurement information includes at least one of the following:
- Time domain channel impulse response channel frequency response, delay power spectrum, channel energy response, reference signal received power RSRP, reference signal received path power RSRPP, reference signal received quality RSRQ, signal to interference plus noise ratio SINR, delay Doppler domain channel.
- the first target information includes at least one of the following:
- the determining, by the processor 1010, a supervision result of the first model according to an association relationship between the first target information and the second target information includes:
- a supervision result of the first model is determined according to a difference or correlation between the first distance and the second distance.
- the determining of the supervision result of the first model according to the difference between the first distance and the second distance performed by the processor 1010 includes:
- the first condition includes at least one of the following:
- the difference between the first distance and the second distance is greater than a first threshold
- F groups of first target measurement information wherein a difference between a first distance and a second distance corresponding to the same group of first target measurement information is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is greater than a second threshold
- the third distance between the first distance distribution and the second distance distribution is greater than or equal to the third threshold value, and the first distance distribution is the distribution of the first distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range; the second distance distribution is the distribution of the second distances corresponding to all groups of target measurement information or the preset R groups of target measurement information measured within the target time range.
- the determining of the supervision result of the first model according to the difference between the first distance and the second distance performed by the processor 1010 includes:
- the first model is determined to be valid, and the second condition includes at least one of the following:
- the number of groups of first target measurement information is less than or equal to F, wherein the difference between the first distance and the second distance corresponding to the same group of first target measurement information is greater than a first threshold, F is an integer greater than or equal to 1, and R is an integer greater than or equal to F;
- a proportion of the number of groups of the first target measurement information is less than or equal to a second threshold
- the set of target measurement information includes at least one of the following:
- the step of inputting the target measurement information into the second model to obtain second target information output by the second model, performed by the processor 1010 includes:
- a set of target measurement information is input into a second model, and second target information output by the second model is obtained, wherein the second target information includes:
- the third distance includes at least one of the following:
- the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution is the maximum vertical distance between the cumulative probability density curves of the first distance distribution and the second distance distribution.
- the supervision result of the first model includes at least one of the following:
- processor 1010 is further configured to perform at least one of the following:
- the target measurement information is subjected to a second preprocessing, wherein the data input by the second model includes the target measurement information after the second preprocessing.
- processor 1010 is further configured to perform at least one of the following:
- the output information of the second model is subjected to a second post-processing, wherein the second target information includes information obtained through the second post-processing.
- the target measurement information includes T TRPs or cell-associated measurement information, where T is an integer greater than or equal to 1.
- the target reference signal used to estimate the target measurement information includes M TRPs or cell-associated reference signals, where M is an integer greater than or equal to 1.
- the embodiment of the present application also provides a network side device, including a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the steps of the method embodiment shown in Figure 6.
- the network side device embodiment corresponds to the above network side device method embodiment, and each implementation process and implementation method of the above method embodiment can be applied to the network side device embodiment, and can achieve the same technical effect.
- the embodiment of the present application also provides a network side device.
- the network side device 1100 includes: an antenna 1101, a radio frequency device 1102, a baseband device 1103, a processor 1104 and a memory 1105.
- the antenna 1101 is connected to the radio frequency device 1102.
- the radio frequency device 1102 receives information through the antenna 1101 and sends the received information to the baseband device 1103 for processing.
- the baseband device 1103 processes the information to be sent and sends it to the radio frequency device 1102.
- the radio frequency device 1102 processes the received information and sends it out through the antenna 1101.
- the method executed by the network-side device in the above embodiment may be implemented in the baseband device 1103, which includes a baseband processor.
- the baseband device 1103 may include, for example, at least one baseband board, on which multiple chips are arranged, as shown in Figure 11, one of which is, for example, a baseband processor, which is connected to the memory 1105 through a bus interface to call the program in the memory 1105 and execute the network device operations shown in the above method embodiment.
- the network side device may also include a network interface 1106, which is, for example, a Common Public Radio Interface (CPRI).
- CPRI Common Public Radio Interface
- the network side device 1100 of the embodiment of the present application further includes: instructions or programs stored in the memory 1105 and executable on the processor 1104, and the processor 1104 calls the instructions or programs in the memory 1105 to execute as shown in FIG. 8
- the methods executed by each module achieve the same technical effect, so they will not be described here to avoid repetition.
- the embodiment of the present application also provides a network side device.
- the network side device 1200 includes: a processor 1201, a network interface 1202, and a memory 1203.
- the network interface 1202 is, for example, a Common Public Radio Interface (CPRI).
- CPRI Common Public Radio Interface
- the network side device 1200 of the embodiment of the present application also includes: instructions or programs stored in the memory 1203 and executable on the processor 1201.
- the processor 1201 calls the instructions or programs in the memory 1203 to execute the method executed by each module as shown in Figure 8 and achieves the same technical effect. To avoid repetition, it will not be repeated here.
- An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored.
- a program or instruction is stored.
- the various processes of the method embodiment shown in Figure 2 or Figure 6 are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
- the processor is the processor in the terminal described in the above embodiment.
- the readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.
- the readable storage medium may be a non-transient readable storage medium.
- An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the method embodiment shown in Figure 2 or Figure 6, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
- the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.
- the embodiments of the present application further provide a computer program/program product, which is stored in a storage medium, and is executed by at least one processor to implement the various processes of the method embodiment shown in Figure 2 or Figure 6, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.
- An embodiment of the present application also provides a communication system, including: a terminal and a first node, wherein the terminal can be used to execute the steps of the information processing method shown in FIG. 2 , and the first node can be used to execute the steps of the information processing method shown in FIG. 6 .
- the above embodiment method can be implemented by means of a computer software product plus a necessary general hardware platform, and of course, it can also be implemented by hardware.
- the machine software product is stored in a storage medium (such as ROM, RAM, disk, CD, etc.), including a number of instructions for enabling the terminal or network side device to execute the methods described in each embodiment of the present application.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Human Computer Interaction (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
本申请公开了一种信息处理方法、信息处理装置、终端及网络侧设备,属于通信技术领域,本申请实施例的信息处理方法包括:终端将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与所述终端的位置相关;所述终端获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
Description
相关申请的交叉引用
本申请主张在2023年4月7日在中国提交的中国专利申请No.202310371177.1的优先权,其全部内容通过引用包含于此。
本申请属于通信技术领域,具体涉及一种信息处理方法、信息处理装置、终端及网络侧设备。
在相关技术中,可以采用机器学习(Machine Learning,ML)模型进行终端定位。但是,基于终端位移、终端所处通信环境改变等因素影响,会降低ML定位模型得出的定位结果的精确度,由此可见,相关技术中,基于ML模型的定位方法存在定位可靠性低的缺陷。
发明内容
本申请实施例提供一种信息处理方法、信息处理装置、终端及网络侧设备,能够监督第一模型的输出信息的精度,以在该第一模型的输出信息的精度低时,及时判断该第一模型失效,从而可以降低因采用失效的第一模型继续对终端进行定位所造成的定位可靠性低的风险。
第一方面,提供了一种信息处理方法,由终端执行,该方法包括:
终端将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与所述终端的位置相关;
所述终端获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
第二方面,提供了一种信息处理方法,由第一节点执行,该方法包括:
第一节点接收来自终端的第一信息,其中,所述第一信息包括目标测量信息,或者所述第一信息包括第一目标信息和目标测量信息,所述第一目标信息为将所述目标测量信息输入第一模型后,由所述第一模型输出的信息,所述第一目标信息与所述终端的位置相关;
所述第一节点将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,所述第二目标信息与所述终端的位置相关;
所述第一节点向所述终端发送第二信息和所述第二目标信息中的至少一项,其中,所述第一模型的监督结果基于所述第一目标信息和所述第二目标信息之间的关联关系确定,
所述第二信息与所述第一模型的监督结果或有效性相关。
第三方面,提供了一种信息处理装置,包括:
第一处理模块,用于将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与终端的位置相关;
获取模块,用于获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
第四方面,提供了一种信息处理装置,包括:
第一接收模块,用于接收来自终端的第一信息,其中,所述第一信息包括目标测量信息,或者所述第一信息包括第一目标信息和目标测量信息,所述第一目标信息为将所述目标测量信息输入第一模型后,由所述第一模型输出的信息,所述第一目标信息与所述终端的位置相关;
第二处理模块,用于将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,所述第二目标信息与所述终端的位置相关;
第一发送模块,用于向所述终端发送第二信息和所述第二目标信息中的至少一项,其中,所述第一模型的监督结果基于所述第一目标信息和所述第二目标信息之间的关联关系确定,所述第二信息与所述第一模型的监督结果或有效性相关。
第五方面,提供了一种终端,该终端包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第六方面,提供了一种终端,包括处理器及通信接口,其中,所述处理器用于终端将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与所述终端的位置相关;所述处理器或所述通信接口用于获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
第七方面,提供了一种第一节点,该第一节点包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第二方面所述的方法的步骤。
第八方面,提供了一种第一节点,包括处理器及通信接口,其中,所述通信接口用于接收来自终端的第一信息,其中,所述第一信息包括目标测量信息,或者所述第一信息包括第一目标信息和目标测量信息,所述第一目标信息为将所述目标测量信息输入第一模型后,由所述第一模型输出的信息,所述第一目标信息与所述终端的位置相关;所述处理器用于将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,所述第
二目标信息与所述终端的位置相关;所述通信接口还用于向所述终端发送第二信息和所述第二目标信息中的至少一项,其中,所述第一模型的监督结果基于所述第一目标信息和所述第二目标信息之间的关联关系确定,所述第二信息与所述第一模型的监督结果或有效性相关。
第九方面,提供了一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤,或者实现如第二方面所述的方法的步骤。
第十方面,提供了一种无线通信系统,包括:终端及第一节点,所述终端可用于执行如第一方面所述的方法的步骤,所述第一节点可用于执行如第二方面所述的方法的步骤。
第十一方面,提供了一种芯片,所述芯片包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现如第一方面所述的方法,或实现如第二方面所述的方法。
第十二方面,提供了一种计算机程序/程序产品,所述计算机程序/程序产品被存储在存储介质中,所述程序/程序产品被至少一个处理器执行以实现如第一方面所述的信息处理方法的步骤,或者实现如第二方面所述的信息处理方法的步骤。
在本申请实施例中,可以分别向第一模型和第二模型输入相同的目标测量信息,以获得第一模型输出的第一目标信息和第二模型输出的第二目标信息,鉴于第一目标信息和第二目标信息对应相同的目标测量信息,在第一模型有效的情况下,第一目标信息和第二目标信息之间存在一定的关联关系,基于该规则,根据所述第一目标信息和所述第二目标信息之间的关联关系,便可以确定所述第一模型的监督结果。例如:若第一目标信息和第二目标信息对应相同的终端位置信息,则确定第一模型有效,若第一目标信息和第二目标信息对应不相同的终端位置信息,则确定第一模型失效。
图1是本申请实施例能够应用的无线通信系统的结构示意图;
图2是本申请实施例提供的一种信息处理方法的结构示意图
图3是神经网络模型的架构示意图;
图4是神经元的示意图;
图5a是本申请实施例中模型监督方案一的流程图;
图5b是本申请实施例中模型监督方案二的流程图;
图5c是本申请实施例中模型监督方案三的流程图;
图5d是本申请实施例中第三距离的经验累积分布(Empirical CDF)函数曲线图;
图6是本申请实施例提供的一种信息处理方法的流程图;
图7是本申请实施例提供的一种信息处理装置的结构示意图;
图8是本申请实施例提供的一种信息处理装置的结构示意图;
图9是本申请实施例提供的一种通信设备的结构示意图;
图10是本申请实施例提供的一种终端的结构示意图;
图11是本申请实施例提供的一种网络侧设备的结构示意图;
图12是本申请实施例提供的另一种网络侧设备的结构示意图。
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚描述,显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员所获得的所有其他实施例,都属于本申请保护的范围。
本申请的术语“第一”、“第二”等是用于区别类似的对象,而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换,以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施,且“第一”、“第二”所区别的对象通常为一类,并不限定对象的个数,例如第一对象可以是一个,也可以是多个。此外,本申请中的“或”表示所连接对象的至少其中之一。例如“A或B”涵盖三种方案,即,方案一:包括A且不包括B;方案二:包括B且不包括A;方案三:既包括A又包括B。字符“/”一般表示前后关联对象是一种“或”的关系。
本申请的术语“指示”既可以是一个直接的指示(或者说显式的指示),也可以是一个间接的指示(或者说隐含的指示)。其中,直接的指示可以理解为,发送方在发送的指示中明确告知了接收方具体的信息、需要执行的操作或请求结果等内容;间接的指示可以理解为,接收方根据发送方发送的指示确定对应的信息,或者进行判断并根据判断结果确定需要执行的操作或请求结果等。
值得指出的是,本申请实施例所描述的技术不限于长期演进型(Long Term Evolution,LTE)/LTE的演进(LTE-Advanced,LTE-A)系统,还可用于其他无线通信系统,诸如码分多址(Code Division Multiple Access,CDMA)、时分多址(Time Division Multiple Access,TDMA)、频分多址(Frequency Division Multiple Access,FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access,OFDMA)、单载波频分多址(Single-carrier Frequency-Division Multiple Access,SC-FDMA)或其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用,所描述的技术既可用于以上提及的系统和无线电技术,也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio,NR)系统,并且在以下大部分描述中使用NR术语,但是这些技术也可应用于NR系统以外的系统,如第6代(6th Generation,6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端11和网络侧设备12。其中,终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)、笔记本电脑、个人数字助理(Personal Digital Assistant,PDA)、掌上电脑、上网本、超级移动个人计算机(Ultra-mobile Personal Computer,UMPC)、移动
上网装置(Mobile Internet Device,MID)、增强现实(Augmented Reality,AR)、虚拟现实(Virtual Reality,VR)设备、机器人、可穿戴式设备(Wearable Device)、飞行器(flight vehicle)、车载设备(Vehicle User Equipment,VUE)、船载设备、行人终端(Pedestrian User Equipment,PUE)、智能家居(具有无线通信功能的家居设备,如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(Personal Computer,PC)、柜员机或者自助机等终端侧设备。可穿戴式设备包括:智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。其中,车载设备也可以称为车载终端、车载控制器、车载模块、车载部件、车载芯片或车载单元等。需要说明的是,在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备,其中,接入网设备也可以称为无线接入网(Radio Access Network,RAN)设备、无线接入网功能或无线接入网单元。接入网设备可以包括基站、无线局域网(Wireless Local Area Network,WLAN)接入点(Access Point,AS)或无线保真(Wireless Fidelity,WiFi)节点等。其中,基站可被称为节点B(Node B,NB)、演进节点B(Evolved Node B,eNB)、下一代节点B(the next generation Node B,gNB)、新空口节点B(New Radio Node B,NR Node B)、接入点、中继站(Relay Base Station,RBS)、服务基站(Serving Base Station,SBS)、基收发机站(Base Transceiver Station,BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set,BSS)、扩展服务集(Extended Service Set,ESS)、家用B节点(home Node B,HNB)、家用演进型B节点(home evolved Node B)、发送接收点(Transmission Reception Point,TRP)或所述领域中其他某个合适的术语,只要达到相同的技术效果,所述基站不限于特定技术词汇,需要说明的是,在本申请实施例中仅以NR系统中的基站为例进行介绍,并不限定基站的具体类型。
核心网设备可以包含核心网设备可以包含但不限于如下至少一项:核心网节点、核心网功能、定位管理功能(Location Management Function,LMF)、移动管理实体(Mobility Management Entity,MME)、接入移动管理功能(Access and Mobility Management Function,AMF)、会话管理功能(Session Management Function,SMF)、用户平面功能(User Plane Function,UPF)、策略控制功能(Policy Control Function,PCF)、策略与计费规则功能单元(Policy and Charging Rules Function,PCRF)、边缘应用服务发现功能(Edge Application Server Discovery Function,EASDF)、统一数据管理(Unified Data Management,UDM)、统一数据仓储(Unified Data Repository,UDR)、归属用户服务器(Home Subscriber Server,HSS)、集中式网络配置(Centralized network configuration,CNC)、网络存储功能(Network Repository Function,NRF)、网络开放功能(Network Exposure Function,NEF)、本地NEF(Local NEF,或L-NEF)、绑定支持功能(Binding Support Function,BSF)、应用功能(Application Function,AF)等。需要说明的是,在本申请实施例中仅以NR系统中的核心网设备为例进行介绍,并不限定核心网设备的具体类型。需要说明的是,在本申请实施例中仅以NR系统中的核心网设备为例进行介绍,并不限定核心网设备的具体类型。
下面结合附图,通过一些实施例及其应用场景对本申请实施例提供的信息处理方法、信息处理装置以及相关设备进行详细地说明。
请参阅图2,本申请实施例提供的一种信息处理方法,其执行主体可以是终端,该终端可以是如图1中列举的各种类型的终端11,或者是除了如图1所示实施例中列举的终端类型之外的其他终端,在此不作具体限定。如图2所示,该信息处理方法可以包括以下步骤:
步骤201、终端将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与所述终端的位置相关。
步骤202、所述终端获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
一种实施方式中,上述目标测量信息可以是终端对参考信号进行测量所得到的信息。
可选地,上述参考信号可以包括以下至少一项:
CSI参考信号(CSI Reference Signal,CSI-RS)、探测参考信号(Sounding Reference Signal,SRS)、同步信号/物理广播信道信号块(或同步信号块)(Synchronization Signal and PBCH block,SSB)、定位参考信号(Positioning Reference Signal,PRS)、跟踪参考信号(Tracking Reference Signal,TRS)、相位跟踪参考信号(Phase-Tracking Reference Signal,PTRS)中的至少一项,其中,CSI为信道状态信息(Channel State Information)的简称。
可选地,上述目标测量信息可以包括时域信道脉冲响应(Channel Impulse Response,CIR)、和时延功率谱(Power Delay Profile,PDP)、信道频率响应、信道能量响应、参考信号接收功率(Reference Signal Received Power,RSRP)、参考信号接收路径功率(Reference Signal Received Path Power,RSRPP)、参考信号接收质量(Reference Signal Received Quality,RSRQ)、信号与干扰加噪声比(Signal to Interference plus Noise Ratio,SINR)、时延多普勒域信道中的至少一项。
值得说明的是,上述信道频率响应、信道能量响应、RSRP、RSRPP、RSRQ和SINR等测量信息,可以是层1(Layer 1,L1)的测量信息,也就是即时测量信息;或者这些测量信息可以是层3(Layer 3,L3)的测量信息,即对即时测量信息以及历史测量信息进行平滑滤波处理后的测量信息。
可选地,所述目标测量信息包括T个TRP或小区关联的测量信息,T为大于或等于1的整数。换而言之,用于测量得到所述目标测量信息的参考信号可以由一个或至少两个TRP或小区发送,具体地,用于估计得到所述目标测量信息的目标参考信号包括M个TRP或小区关联的参考信号,M为大于或等于1的整数。
另一种实施方式中,目标测量信息可以是预设的,如网络侧配置的。
一种实施方式中,第一目标信息与所述终端的位置相关,可以是第一目标信息包括所述终端的位置信息,或者,第一目标信息包括与所述终端的位置相关的特征信息。
可选地,所述终端的位置信息包括以下至少一项:
所述终端的绝对位置信息;
所述终端的相对位置信息。
其中,绝对位置信息可以包括终端的坐标位置信息,如全球定位系统(Global Positioning System,GPS)定位信息。
其中,相对位置信息,可以是终端相对参考点的位置信息,如终端相对基站的位置信息。
另一种实施方式中,上述目标模型可以用于输出与终端位置相关的特征信息。
可选地,与所述终端的位置相关的特征信息可以包括的以下至少一项:
视距(Line of Sight,LOS)径到达时间(Time of Arrival,TOA);
视距(Line of Sight,LOS)径到达角(Angle of Arrival,AOA);
视距(Line of Sight,LOS)径离开角(Angle of Departure,AOD);
视距(Line of Sight,LOS)径参考信号时间差(Reference Signal Time Different,RSTD);
波束质量;
信道压缩指示;
预编码矩阵指示(Precoding Matrix Indicator,PMI);
时域信道信息;
频域信道信息;
空域信道信息;
小区通信质量;
小区切换判断结果。
值得提出的是,上述LOS径可以是终端与网络侧设备之间实际存在的LOS径,也可以能是假设的终端与网络侧设备之间的LOS径,也就是说,无论实际情况下,用户设备(User Equipment,UE)与基站之间是否存在LOS径,都可以基于本申请实施例确定UE与基站之间的LOS径的TOA、AOA、AOD、RSTD等,进而可以据此确定UE的位置信息。
此外,上述LOS径的TOA、AOA、AOD、RSTD,以及波束质量、信道压缩指示、PMI、时域信道信息、频域信道信息、空域信道信息、小区通信质量、小区切换判断结果等都是与终端的位置密切相关的信息,即终端位置发生改变时,上述LOS径的TOA、AOA、AOD、RSTD,以及波束质量、信道压缩指示、PMI、时域信道信息、频域信道信息、空域信道信息、小区通信质量、小区切换判断结果等也会发生相应的改变,例如:终端的位置与基站距离越远,则波束质量会越差、参考信号的传输时延会增大、小区通信质量会变差甚至触发小区切换。
上述第二模型可以是第一模型的监督模型,用于监督第一模型的精度或监督第一模型是否有效。
一种实施方式中,第二模型可以是与第一模型对应不同空间的模型,如第一模型对应
物理空间,第二模型对应潜在的(latent)空间,如特征空间。
例如:假设目标测量信息包括CIR,通过第二模型可以用于将CIR映射到另一个latent空间,该latent空间内UE的相对位置关系d2与物理空间中UE的相对位置关系d1是一致的。这样,对于两个UE的CIR,如果第一模型有效,那么两个UE的CIR在latent空间和物理空间的相对位置应该一致;如果第一模型失效,那么两个UE的CIR在latent空间和物理空间的相对位置则不再一致。
其中,物理空间可以理解为UE的位置坐标所对应的空间,如UE的位置坐标可以视为这个物理空间的一个点。
即若满足以下公式,则两个UE的CIR在latent空间和物理空间的相对位置一致,即第一模型有效:
d1=||g(cir1)-g(cir2)||
d2=||f(cir1)-f(cir2)||
|d1-d2|<ζ
d1=||g(cir1)-g(cir2)||
d2=||f(cir1)-f(cir2)||
|d1-d2|<ζ
其中,g为待监督的第一模型,即CIR到物理空间的映射;f为第二模型即CIR到latent空间的映射;d1表示第一距离;d2表示第二距离;|d1-d2|表示第三距离;ζ表示第三阈值。
一种实施方式中,第二目标信息与终端的位置相关,可以是第二目标信息包括所述终端的位置信息,或者,第二目标信息包括至少两个终端位置之间的距离信息,例如:同一终端在两个不同时刻的位置之间的距离信息,或者是不同终端的位置之间的距离信息。
可选地,第二目标信息可以是终端的相对位置信息,且第二目标信息与第一目标信息的参考位置不同。
另一种实施方式中,第二目标信息包括所述终端的位置相关的特征信息,或者,第二目标信息包括所述终端的至少两个终端位置相关的特征信息之间的距离或差异。
值得注意的是,即使第一目标信息和第二目标信息为不同种类的信息,如第一目标信息为终端的位置信息,第二目标信息为终端在不同时刻的位置之间的距离信息,或者,即使第一目标信息和第二目标信息对应不同的空间,如第一目标信息为终端在物理空间内的位置信息,第二目标信息为终端在latent空间内的位置信息,鉴于第一目标信息和第二目标信息分别与终端的位置相关,这样,第一目标信息和第二目标信息之间仍然存在一定的关联关系,如基于第一目标信息确定的终端在物理空间中的不同时刻的位置之间的第一距离,与基于第二目标信息确定的终端在latent空间中的不同时刻的位置之间的第二距离可以是相等的,或者是线性相关的。基于以上规则,本申请实施例中通过比较与相同目标测量信息对应的第一目标信息和第二目标信息之间的关联关系,便可以确定第一目标输出的第一目标信息的准确度或第一模型是否有效等模型监督结果。
为了便于说明,本申请实施例中,通常以第一目标信息是物理空间内的终端位置,第二目标信息是latent空间内的终端位置,或终端距离为例进行举例说明,在此不构成具体限定。
值得说明的是,本申请实施中的第一模型和第二模型可以是人工智能(Artificial Intelligence,AI)模型,也可以是机器学习(Machine Learning,ML)模型,为了便于说明,本申请实施例中以第一模型和第二模型是AI模型为例进行举例说明。
其中,将AI模块融入无线通信网络,显著提升吞吐量、时延以及用户容量等技术指标是未来的无线通信网络的重要任务。AI模块有多种实现方式,例如:神经网络、决策树、支持向量机、贝叶斯分类器等。本申请以神经网络为例进行说明,但是并不限定AI模块的具体类型。
如图3所示,神经网络包括输入层、隐层和输出层,其可以根据输入层获取的出入信息(X1~Xn)预测可能的输出结果(Y)。神经网络由大量的神经元组成,如图4所示,神经元的参数包括:输入参数a1~aK、权值w、偏置b以及激活函数σ(z),以及与这些参数获取输出值a,其中,常见的激活函数包括S型生长曲线(Sigmoid)函数、双曲正切(tanh)函数、线性整流函数(Rectified Linear Unit,ReLU,其也称之为修正线性单元)函数等等,且上述函数σ(z)中的z可以通过以下公式计算得到:
z=a1w1+…+akwk+aKwK+b
z=a1w1+…+akwk+aKwK+b
其中,K表示输入参数的总数。
神经网络的参数通过优化算法进行优化。优化算法就是一种能够帮我们最小化或者最大化目标函数(有时候也叫损失函数)的一类算法。而目标函数往往是模型参数和数据的数学组合。例如给定数据X和其对应的标签Y,我们构建一个神经网络f(.),有了模神经网络后,根据输入x就可以得到预测输出f(x),并且可以计算出预测值和真实值之间的差距(f(x)-Y),这个就是损失函数。我们的目的是找到合适的W和b,使上述的损失函数的值达到最小,损失值越小,则说明我们的神经网络越接近于真实情况。
目前常见的优化算法,基本都是基于误差反向传播(error Back Propagation,BP)算法。误差反向传播算法的基本思想是,学习过程由信号的正向传播与误差的反向传播两个过程组成。正向传播时,输入样本从输入层传入,经各隐层逐层处理后,传向输出层。若输出层的实际输出与期望的输出不符,则转入误差的反向传播阶段。误差反传是将输出误差以某种形式通过隐层向输入层逐层反传,并将误差分摊给各层的所有单元,从而获得各层单元的误差信号,此误差信号即作为修正各单元权值的依据。这种信号正向传播与误差反向传播的各层权值调整过程,是周而复始地进行的。权值不断调整的过程,也就是网络的学习训练过程。此过程一直进行到网络输出的误差减少到可接受的程度,或进行到预先设定的学习次数为止。
常见的优化算法有梯度下降(Gradient Descent)、随机梯度下降(Stochastic Gradient Descent,SGD)、小批量梯度下降(mini-batch gradient descent)、动量法(Momentum)、Nesterov(其表示带动量的随机梯度下降)、自适应梯度下降(Adaptive gradient descent,Adagrad)、自适应学习率调整(Adadelta)、均方根误差降速(root mean square prop,RMSprop)、自适应动量估计(Adaptive Moment Estimation,Adam)等。
这些优化算法在误差反向传播时,都是根据损失函数得到的误差/损失,对当前神经元求导数/偏导,加上学习速率、之前的梯度/导数/偏导等影响,得到梯度,将梯度传给上一层。
值得注意的是,上述第二模型可以部署在终端,也可能部署在第一节点,该第一节点可以是网络侧设备,如接入网设备或核心网设备,或者是第一节点可以是应用服务器等任意具有目标模型的设备。为了便于说明,本申请实施例中,通常以第一节点是接入网设备,如基站为例进行举例说明,在此不构成具体限定。此时,上述第二目标信息需要将目标测量信息输入第二模型得到,上述第一模型的监督结果需要基于所述第一目标信息和所述第二目标信息之间的关联关系确定。
基于此,上述终端获取第一模型的监督结果的方式可以是:基于自身具有的第二模型对目标测量信息进行处理,得出第二目标信息,并将基于相同的目标测量信息得出的第一目标信息和第二目标信息进行比较,以根据两者之间的关联关系确定第一模型的监督结果。
或者,上述终端获取第一模型的监督结果的方式可以是:终端向具有第二模型的第一节点发送目标测量信息,由第一节点采用第二模型对目标测量信息进行处理,得出第二目标信息,并向终端反馈第二目标信息,此后,终端可以将基于相同的目标测量信息得出的第一目标信息和第二目标信息进行比较,以根据两者之间的关联关系确定第一模型的监督结果。
或者,上述终端获取第一模型的监督结果的方式可以是:终端向具有第二模型的第一节点发送目标测量信息和第一目标信息,由第一节点采用第二模型对目标测量信息进行处理,得出第二目标信息,并在将基于相同的目标测量信息得出的第一目标信息和第二目标信息进行比较,以根据两者之间的关联关系确定第一模型的监督结果之后,向所述终端反馈第二信息,以通过第二信息指示第一模型的监督结果,如第二信息指示第一模型有效或无效,或指示第一模型在某些场景下有效,在另一些场景下失效等。
值得注意的是,本申请实施例中的第一目标信息与第二目标信息是基于相同的目标测量信息得到的,即将相同的目标测量信息分别输入第一模型和第二模型,则第一模型输出的信息为第一目标信息,第二模型输出的信息为第二目标信息。基于相同的目标测量信息所对应的终端位置信息或与位置相关的特征信息应该相同或者可以相关,基于此,可以基于所述第一目标信息和所述第二目标信息之间的关联关系确定第一模型的监督结果,如第一模型有效或失效等。
也就是说,本申请实施例中的基于所述第一目标信息和所述第二目标信息之间的关联关系确定第一模型的监督结果,可以是基于与相同目标测量信息对应的第一目标信息和第二目标信息之间的关联关系确定第一模型的监督结果。
在一种可选的实施方式中,在所述终端具有所述第一模型,且第一节点具有所述第二模型的情况下,所述终端获取所述第一模型的监督结果,包括:
所述终端向所述第一节点发送第一信息,所述第一信息包括所述目标测量信息,或者,所述第一信息包括所述第一目标信息和所述目标测量信息;
所述终端接收来自所述第一节点的第二目标信息,并根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,或者,所述终端接收来自所述第一节点的第二信息,所述第二信息与所述第一模型的监督结果或有效性相关。
一种实施方式中,终端向所述第一节点发送第一目标信息和所述目标测量信息,第一节点将所述目标测量信息输入第二模型,以得到第二目标信息,然后,第一节点基于第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,即由第一节点来确定所述第一模型的监督结果,并通过第二信息向终端指示该第一模型的监督结果或有效性相关的信息。
另一种实施方式中,终端向所述第一节点发送所述目标测量信息,第一节点将所述目标测量信息输入第二模型,以得到第二目标信息,然后,第一节点将第二目标信息发送给终端,最后,终端基于第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,即由终端来确定所述第一模型的监督结果。
在另一种可选的实施方式中,在所述终端具有所述第一模型和所述第二模型的情况下,所述终端获取所述第一模型的监督结果,包括:
所述终端将所述目标测量信息输入所述第二模型,获得所述第二模型输出的所述第二目标信息;
所述终端根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果。
本实施方式中,终端既有第一模型,也有第二模型,因此,终端可以利用第一模型得出第一目标信息,利用第二模型得出第二目标信息,并将与相同的目标测量信息对应的第一目标信息和第二目标信息进行比较,以根据第一目标信息和第二目标信息之间的关联关系,确定第一模型的监督结果。
可选地,在终端根据所述第一目标信息和所述第二目标信息之间的关联关系之后,所述方法还包括:
所述终端向第一节点发送第三信息,所述第三信息与所述第一模型的监督结果或有效性相关。
这样,终端通过向第一节点上报第一模型的监督结果或有效性相关的信息,可以使第一节点据此决定是否采用第一模型输出的数据作为定位信息,如在第一模型失效时,第一节点采用其他方式(如非AI的方式)确定终端的位置信息;或者使第一节点据此决定是否更新第一模型,如在第一模型失效时,第一节点训练更新第一模型,并向终端下发更新后的第一模型;或者,使第一节点据此决定是否去激活当前的第一模型,并且激活或选择新的第一模型,如在第一模型失效时,第一节点去激活当前使用的第一模型,并激活或选择新的第一模型。
作为一种可选的实施方式,所述终端根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,包括:
所述终端获取一组第一目标信息之间的第一距离,以及获取一组第二目标信息之间的第二距离,其中,所述一组第一目标信息与所述一组第二目标信息分别基于所述第一模型和所述第二模型对同一组目标测量信息进行处理得到;
所述终端根据所述第一距离和所述第二距离之间的差异或相关性,确定所述第一模型的监督结果。
其中,一组目标测量信息可以包括至少两个目标测量信息。为了便于说明,本申请实施例中,以一组目标测量信息包括两个目标测量信息为例进行举例说明。
一种实施方式中,一组目标测量信息可以包括至少两个来自同一终端的不同时刻的目标测量信息。
其中,本申请实施例中的时刻可以包括正交频分复用(Orthogonal Frequency Division Multiplex,OFDM)符号、子帧、帧、纳秒、微妙、毫秒、秒、分钟、小时、天、月等。例如:一组目标测量信息可以包括所述终端通过对时刻1的参考信号进行测量得到的目标测量信息,以及所述终端通过时刻2的参考信号进行测量得到的目标测量信息。
另一种实施方式中,一组目标测量信息可以包括至少两个来自不同终端的目标测量信息。例如:一组目标测量信息可以包括终端A的目标测量信息和终端B的目标测量信息。
值得说明的是,在一组目标测量信息可以包括至少两个来自不同终端的目标测量信息的情况下,若终端具有第二模型,则终端需要获取其他终端的目标测量信息,以利用本终端的目标测量信息与其他终端的目标测量信息,构成一组目标测量信息。
此外,在第一节点具有第二模型的情况下,不同终端可以分别将各自的目标测量信息发送至第一节点,如基站,以使基站利用不同终端的目标测量信息构成一组目标测量信息,值得说明的是,本申请中的第一节点也可能是核心网设备,为了便于说明,本申请实施例中以第一节点是基站为例进行举例说明,在此不构成具体限定。
一种实施方式中,终端可以将一组目标测量信息中的目标测量信息分别输入第一模型,以获取第一模型分别输出的一组第一目标信息,然后计算一组第一目标信息中的任意两个第一目标信息之间的第一距离。
例如:如图5a所示场景中,假设一组目标测量信息包括UE 1的目标测量信息A和UE 2的目标测量信息B,则终端将目标测量信息A和目标测量信息B分别输入第一模型,得到第一模型分别输出的UE 1的位置信息和UE 2的位置信息,然后据此计算UE 1和UE 2之间的第一距离。
再例如:如图5b所示场景中,假设一组目标测量信息包括同一终端在时隙(slot)1和slot 2分别采集的CIR,则终端将slot 1采集的CIR和slot 2采集的CIR分别输入第一模型,得到第一模型分别输出的终端在slot 1的位置信息和终端在slot 2的位置信息,然后据此计算终端分别在slot 1和slot 2的位置之间的第一距离。
一种实施方式中,终端可以将一组目标测量信息并行输入第二模型,以获取第二模型输出的第二目标信息,该第二目标信息即为第二距离。
可选地,所述终端将所述目标测量信息输入所述第二模型,获得所述第二模型输出的第二目标信息,包括:
所述终端将一组目标测量信息输入第二模型,并获得所述第二模型输出的第二目标信息,其中,所述第二目标信息包括:
所述一组目标测量信息中的至少两个目标测量信息各自对应的位置之间的距离。
例如:如图5a所示场景中,假设一组目标测量信息包括UE 1的目标测量信息A和UE 2的目标测量信息B,则终端将目标测量信息A和目标测量信息B并行输入第二模型,得到第二模型输出的UE 1和UE 2之间的第二距离。
再例如:如图5b所示场景中,假设一组目标测量信息包括同一终端在slot 1和slot 2分别采集的CIR,则终端将slot 1采集的CIR和slot 2采集的CIR并行输入第二模型,得到第二模型分别输出的终端分别在slot 1和slot 2的位置之间的第二距离。
另一种实施方式中,终端可以将一组目标测量信息分别输入第二模型,以获取第二模型分别输出的一组第二目标信息,然后计算一组第二目标信息中的任意两个第二目标信息之间的第二距离。
在一种实施方式中,可以采用并行的孪生网络来训练第二模型。
例如:如图5c所示,该孪生网络包括两个结构和参数一致的定位模型,此时,通过向两个定位模型分别输入不同的测量信息,得到两个定位模型输出的预测位置之间的距离作为该孪生网络的输出结果,并将该孪生网络的输出结果与标签中的距离进行比较,得到孪生网络的损失函数,如平均绝对误差(Mean Absolute Error,MAE)。最后,基于该损失函数来判断训练的第二模型是否满足精度需求,其中,标签中的距离表示输入的两个不同的测量信息分别对应的真实位置之间的距离。
以一组目标测量信息包括两个目标测量信息为例,此时,将一组目标测量信息中的两个目标测量信息分别输入第一模型,会得出两个第一目标信息,这两个第一目标信息之间的距离便是第一距离;将一组目标测量信息中的两个目标测量信息并行输入第二模型,会得出第二距离。
可选地,所述第一距离包括以下至少一项:欧式距离、曼哈顿距离、余弦距离;
或,
所述第二距离包括以下至少一项:欧式距离、曼哈顿距离、余弦距离。
例如:闵可夫斯基距离定义为:给定样本空间X,X是m维实数向量空间中的点集合,其中xi,xj∈X,二者之间的闵可夫斯基距离为:
当p=2时,dij表示欧氏距离,即
当p=1时,dij表示曼哈顿距离,即
当p=∞时,dij表示切比雪夫距离,取各个坐标数值差的绝对值的最大值,即
本申请实施例中的距离的定义与矩阵论中向量范数的定义一致,在此不再赘述。
一种实施方式中,上述第一距离和第二距离之间的差异,可以包括:第一距离和第二距离之间的差值,或者是空间距离。
一种实施方式中,上述第一距离和第二距离之间的相关性,可以包括:第一距离和第二距离的相似程度,如第一距离和第二距离是否相等,以及是否可以对第一距离进行线性变换得到第二距离等。
可选地,可以通过计算第一距离和第二距离之间的相关系数,来度量第一距离和第二距离之间的相关性。
例如:给定样本空间X,X是m维实数向量空间中的点集合,其中xi,xj∈X,此时,样本xi,xj的相关系数可以表示为以下公式:
其中,rij表示xi,xj之间的相关性系数,该相关系数的绝对值越接近于1,则xi,xj越相似;该相关系数的绝对值越接近于0,则xi,xj越不相似。
再例如:给定样本空间X,X是m维实数向量空间中的点集合,其中xi,xj∈X,此时,样本xi,xj的相关系数可以是样本xi,xj的夹角余弦,该夹角余弦可以表示为以下公式:
其中,sij表示xi,xj之间的夹角余弦。
值得提出的是,本申请实施例中第一距离和第二距离之间的相关性和距离是两个不同的概念,其中,距离描述的是空间上的相似度,而相关性描述的是变化趋势的一致性。
本实施方式中,所述终端根据所述第一距离和所述第二距离之间的差异或相关性,确定所述第一模型的监督结果,可以是:若所述第一距离等于所述第二距离,或者所述第一距离与所述第二距离之间满足预设的线性转换关系,或第一距离与第二距离之间的差值小于或等于一定阈值,或者将第一距离和第二距离转换至同一维度的空间内之后,两者的差值小于或等于一定阈值,则可以判断第一模型有效。与之相应的,若所述第一距离不等于所述第二距离,或者所述第一距离与所述第二距离之间不满足预设的线性转换关系,或第一距离与第二距离之间的差值大于一定阈值,或者将第一距离和第二距离转换至同一维度的空间内之后,两者的差值大于一定阈值,则可以判断第一模型失效。
作为一种可选的实施方式,所述终端根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:
在所述第一距离和所述第二距离满足第一条件的情况下,所述终端确定所述第一模型失效,所述第一条件包括以下至少一项:
所述第一距离与所述第二距离的差异值大于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,存在F组第一目标测量信息,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比大于第二阈值;
第一距离分布和第二距离分布之间的第三距离大于或等于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
一种实施方式中,上述第一距离与第二距离的差异值,可以是通过将第一距离与第二距离进行直接相减得到的差值,或者,是将第一距离和第二距离变换到同一空间坐标内之后,再进行距离比较得到的距离值。
一种实施方式中,第一目标测量信息表示满足以下条件的目标测量信息:同一组目标测量信息对应的第一距离和第二距离的差异值大于第一阈值。
其中,同一组目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,可以反映第一模型与第二模型的输出结果不一致。
一种实施方式中,可以基于第一模型与第二模型的输出结果不一致的次数达到一定阈值F,或者第一模型与第二模型的输出结果不一致的概率达到第二阈值,则判断该第一模型失效。
另一种实施方式中,可以分别利用第一模型和第二模型获取在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离和第二距离,然后,基于第一距离的分布与第二距离的分布之间的第三距离,来判断第一模型失效。
可选地,所述第三距离包括以下至少一项:
所述第一距离分布和所述第二距离分布的KL散度;
所述第一距离分布和所述第二距离分布的交叉熵;
所述第一距离分布和所述第二距离分布的累积概率密度曲线的最大垂直距离。
其中,第三距离越大,则表示第一距离分布与第二距离分布的差异越大,从而表示第一模型精度越低,这样,在第三距离大于或等于第三阈值时,可以判断第一模式失效。
例如:假设第一模型和第二模型基于第一训练集训练得到,如图5d所示线条X表示:在目标测量信息为与第一训练集在相同环境下测量得到的测量信息的情况下,第三距离的Empirical CDF函数曲线;如图5d所示线条Y表示:在目标测量信息为与第一训练集在不同环境下测量得到的测量信息的情况下,第三距离的Empirical CDF函数曲线。
基于如图5d中的线条X和线条Y可知:当采用与第一训练集在相同环境下测量得到的测量信息作为目标测量信息时,第三距离较小,即第一模型针对该目标测量信息有效的置信度较高;当采用与第一训练集在不同环境下测量得到的测量信息作为目标测量信息时,第三距离明显增大,即第一模型针对该目标测量信息有效的置信度较低。
值得说明的是,本申请实施例中的第一阈值、第二阈值、第三阈值、F等阈值,分别可以由终端决定并上报至网络侧设备,或者,可以采用网络侧设备指示或协议约定的方式指示上述阈值,在此不再赘述。
在另一种可选的实数方式中,所述终端根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:
在所述第一距离和所述第二距离满足第二条件的情况下,所述终端确定所述第一模型有效,所述第二条件包括以下至少一项:
所述第一距离与所述第二距离的差异值小于或等于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,第一目标测量信息的组数小于或等于F,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比小于或等于第二阈值;
第一距离分布和第二距离分布之间的第三距离小于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
本实施方式与上一实施方式中,如何判断第一模型失效的方式相似,不同之处在于:
本实施方式中,是用于判断第一模型是否有效,其具体的解释说明可以参考上一实施方式中的解释说明,在此不再赘述。
可选地,所述第一模型的监督结果包括以下至少一项:
满足所述第一条件的目标测量信息的组数,如第一目标测量信息的组数;
满足所述第一条件的目标测量信息的占比,如第一目标测量信息在指定时间范围内的全部目标测量信息中的占比,或者第一目标测量信息在指定的R组目标测量信息中的占比;
满足所述第二条件的目标测量信息的组数;
满足所述第二条件的目标测量信息的占比;
所述第三距离;
所述第一模型有效的置信度;
所述第一模型有效的指示;
时间戳信息,如得出所述第一模型的监督结果的时间戳信息,或得出所述第一模型的监督结果所根据的目标测量信息的时间戳信息。
作为一种可选的实施方式,所述方法还包括以下至少一项:
所述终端对所述目标测量信息进行第一预处理,其中,所述第一模型输入的数据包括所述第一预处理后的目标测量信息;
所述终端对所述目标测量信息进行第二预处理,其中,所述第二模型输入的数据包括所述第二预处理后的目标测量信息。
其中,第一预处理和第二预处理可以是相同的预处理,如:压缩、量化、截断、滤波、归一化等处理,该第一预处理和第二预处理可以减低目标测量信息的数据量,如降低目标测量信息的bit数,或者,第一预处理和第二预处理可以简化或改变目标测量信息的格式,以便于输入第一模型和第二模型。
一种实施方式中,在所述终端向所述第一节点发送所述目标测量信息的情况下,可以将第二预处理后的目标测量信息发送给第一节点,这样,可以降低传输目标测量信息所消耗的空口资源。
作为一种可选的实施方式,所述方法还包括以下至少一项:
所述终端对所述第一模型的输出信息进行第一后处理,其中,所述第一目标信息包括经所述第一后处理得到的信息;
所述终端对所述第二模型的输出信息进行第二后处理,其中,所述第二目标信息包括经所述第二后处理得到的信息。
一种实施方式中,第一后处理和第二后处理可以是相同或相似的处理,经第一后处理和第二后处理的信息处于同一维度或处于同一空间,这样,便于获取第一目标信息和第二目标信息之间的关联关系。
例如:第一后处理和第二后处理用于将第一模型输出的位置信息和第二模型输出的位置信息转换至全局坐标位置,这样,通比较两个全局坐标位置是否一致,便可以确定第一
模型的输出信息与第二模型的输出信息是否一致。
再例如:第一后处理和第二后处理用于将第一模型输出的时域信息和第二模型输出的时域信息转换为频域信息,这样,通比较两个频域信息是否一致,便可以确定第一模型的输出信息与第二模型的输出信息是否一致。
在本申请实施例中,可以分别向第一模型和第二模型输入相同的目标测量信息,以获得第一模型输出的第一目标信息和第二模型输出的第二目标信息,鉴于第一目标信息和第二目标信息对应相同的目标测量信息,在第一模型有效的情况下,第一目标信息和第二目标信息之间存在一定的关联关系,基于该规则,根据所述第一目标信息和所述第二目标信息之间的关联关系,便可以确定所述第一模型的监督结果。例如:若第一目标信息和第二目标信息对应相同的终端位置信息,则确定第一模型有效,若第一目标信息和第二目标信息对应不相同的终端位置信息,则确定第一模型失效。
请参阅图6,本申请实施例提供的另一种信号处理方法,其执行主体可以是第一节点,该第一节点可以是网络侧设备或服务器,其中,网络侧设备可以是如图1中列举的各种类型的网络侧设备12,如接入网设备或核心网设备,或者是除了如图1所示实施例中列举的网络侧设备类型之外的其他网络侧设备,为了便于说明,本申请实施例中通常以第一节点是基站为例进行举例说明。在此不构成具体限定。如图6所示,该信号处理方法可以包括以下步骤:
步骤601、第一节接收来自终端的第一信息,其中,所述第一信息包括目标测量信息,或者所述第一信息包括第一目标信息和目标测量信息,所述第一目标信息为将所述目标测量信息输入第一模型后,由所述第一模型输出的信息,所述第一目标信息与所述终端的位置相关。
步骤602、所述第一节点将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,所述第二目标信息与所述终端的位置相关。
步骤603、所述第一节点向所述终端发送第二信息和所述第二目标信息中的至少一项,其中,所述第一模型的监督结果基于所述第一目标信息和所述第二目标信息之间的关联关系确定,所述第二信息与所述第一模型的监督结果或有效性相关。
本申请实施例中,第一节点具有第二模型,此时,由第一节点将目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息。
一种实施方式中,由第一节点确定第一模型的监督结果。
可选地,在所述第一节点向所述终端发送第二信息之前,所述方法还包括:
所述第一节点根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果。
本实施方式中,所述第一节点接收的第一信息包括第一目标信息和目标测量信息,这样,在第一节点得出第二目标信息之后,基于与相同目标测量信息对应的第一目标信息和第二目标信息之间的关联关系,来确定第一模型的监督结果。
另一种实施方式中,由终端确定第一模型的监督结果。
本实施方式中,所述第一节点接收的第一信息可以包括目标测量信息,且不包括第一目标信息,且在第一节点得出第二目标信息之后,将该第二目标信息发送给终端,以使终端根据第一模型输出的第一目标信息,和接收的第二目标信息之间的关联关系,来确定第一模型的监督结果。
可选地,在所述第一节点向所述终端发送所述第二目标信息的情况下,所述方法还包括:
所述第一节点接收来自所述终端的第三信息,所述第三信息与所述第一模型的监督结果或有效性相关。
需要说明的是,上述目标测量信息、第一信息、第二信息、第一目标信息、第二目标信息、第一模型、第二模型的含义和作用与如图2所示方法实施例中的目标测量信息、第一信息、第二信息、第一目标信息、第二目标信息、第一模型、第二模型的含义和作用相同,具体可以参考如图2所示方法实施例中的相应说明,在此不再赘述。
本申请实施例中,特指第二模型位于第一节点时,第一节点与终端进行交互,以共同实现对第一模型的模型监督功能。
作为一种可选的实施方式,所述目标测量信息包括以下至少一项:
时域信道脉冲响应、信道频率响应、时延功率谱、信道能量响应、参考信号接收功率RSRP、参考信号接收路径功率RSRPP、参考信号接收质量RSRQ、信号与干扰加噪声比SINR、时延多普勒域信道。
作为一种可选的实施方式,所述第一目标信息包括以下至少一项:
位置信息;
视距传播路径LOS到达时延TOA;
LOS到达角AOA;
LOS离开角AOD;
波束质量;
信道压缩指示;
预编码矩阵指示PMI;
时域信道信息;
频域信道信息;
空域信道信息;
小区通信质量;
小区切换判断结果。
作为一种可选的实施方式,所述第一节点根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,包括:
所述第一节点获取一组第一目标信息之间的第一距离,以及获取一组第二目标信息之
间的第二距离,其中,所述一组第一目标信息与所述一组第二目标信息分别基于所述第一模型和所述第二模型对同一组目标测量信息进行处理得到;
所述第一节点根据所述第一距离和所述第二距离之间的差异或相关性,确定所述第一模型的监督结果。
作为一种可选的实施方式,所述第一节点根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:
在所述第一距离和所述第二距离满足第一条件的情况下,所述第一节点确定所述第一模型失效,所述第一条件包括以下至少一项:
所述第一距离与所述第二距离的差异值大于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,存在F组第一目标测量信息,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比大于第二阈值;
第一距离分布和第二距离分布之间的第三距离大于或等于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
作为一种可选的实施方式,所述第一节点根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:
在所述第一距离和所述第二距离满足第二条件的情况下,所述第一节点确定所述第一模型有效,所述第二条件包括以下至少一项:
所述第一距离与所述第二距离的差异值小于或等于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,第一目标测量信息的组数小于或等于F,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比小于或等于第二阈值;
第一距离分布和第二距离分布之间的第三距离小于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
作为一种可选的实施方式,所述一组目标测量信息包括以下至少一项:
至少两个来自不同终端的目标测量信息;
至少两个来自同一终端的不同时刻的目标测量信息。
作为一种可选的实施方式,所述第一节点将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,包括:
所述第一节点将一组目标测量信息输入第二模型,并获得所述第二模型输出的第二目标信息,其中,所述第二目标信息包括:
所述一组目标测量信息中的至少两个目标测量信息各自对应的位置之间的距离。
作为一种可选的实施方式,所述第三距离包括以下至少一项:
所述第一距离分布和所述第二距离分布的KL散度;
所述第一距离分布和所述第二距离分布的交叉熵;
所述第一距离分布和所述第二距离分布的累积概率密度曲线的最大垂直距离。
作为一种可选的实施方式,所述第一距离包括以下至少一项:欧式距离、曼哈顿距离、余弦距离;
或,
所述第二距离包括以下至少一项:欧式距离、曼哈顿距离、余弦距离。
作为一种可选的实施方式,所述第一模型的监督结果包括以下至少一项:
满足所述第一条件的目标测量信息的组数;
满足所述第一条件的目标测量信息的占比;
满足所述第二条件的目标测量信息的组数;
满足所述第二条件的目标测量信息的占比;
所述第三距离;
所述第一模型有效的置信度;
所述第一模型有效的指示;
时间戳信息。
作为一种可选的实施方式,所述目标测量信息包括T个TRP或小区关联的测量信息,T为大于或等于1的整数。
作为一种可选的实施方式,用于估计得到所述目标测量信息的目标参考信号包括M个TRP或小区关联的参考信号,M为大于或等于1的整数。
本申请实施例中,针对第二模型位于第一节点的情况,第一节点与终端进行交互,以利用第一节点的第二模型实现对终端使用的第一模型的有效性监督,其能够取得与如图2所示方法实施例相似的有益效果,为避免重复,在此不再赘述。
本申请实施例提供的信息处理方法,执行主体可以为信息处理装置。本申请实施例中以信息处理装置执行信息处理方法为例,说明本申请实施例提供的信息处理装置。
请参阅图7,本申请实施例提供的信息处理装置700可以是终端内的装置。
图7所示,该信息处理装置700可以包括以下模块:
第一处理模块701,用于将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与所述终端的位置相关;
获取模块702,用于获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
可选地,在所述终端具有所述第一模型,且第一节点具有所述第二模型的情况下:
获取模块702,包括:第一发送单元、第一接收单元和第一确定单元;
第一发送单元,用于向所述第一节点发送第一信息,所述第一信息包括所述目标测量信息,或者,所述第一信息包括所述第一目标信息和所述目标测量信息;
第一接收单元,用于接收来自所述第一节点的第二目标信息;
第一确定单元,用于根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果;
或者,
获取模块702,包括:所述第一发送单元和第二接收单元;
所述第二接收单元,用于接收来自所述第一节点的第二信息,所述第二信息与所述第一模型的监督结果或有效性相关。
可选地,在所述终端具有所述第一模型和所述第二模型的情况下,获取模块702,包括:
第一处理单元,用于将所述目标测量信息输入所述第二模型,获得所述第二模型输出的所述第二目标信息;
第二确定单元,用于根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果。
可选地,信息处理装置700还包括:
第二发送模块,用于向第一节点发送第三信息,所述第三信息与所述第一模型的监督结果或有效性相关。
可选地,所述目标测量信息包括以下至少一项:
时域信道脉冲响应、信道频率响应、时延功率谱、信道能量响应、参考信号接收功率RSRP、参考信号接收路径功率RSRPP、参考信号接收质量RSRQ、信号与干扰加噪声比SINR、时延多普勒域信道。
可选地,所述第一目标信息包括以下至少一项:
位置信息;
视距LOS径到达时间TOA;
LOS径到达角AOA;
LOS径离开角AOD;
LOS径参考信号时间差RSTD;
波束质量;
信道压缩指示;
预编码矩阵指示PMI;
时域信道信息;
频域信道信息;
空域信道信息;
小区通信质量;
小区切换判断结果。
可选地,所述第一确定单元,具体用于:
获取一组第一目标信息之间的第一距离,以及获取一组第二目标信息之间的第二距离,其中,所述一组第一目标信息与所述一组第二目标信息分别基于所述第一模型和所述第二模型对同一组目标测量信息进行处理得到;
根据所述第一距离和所述第二距离之间的差异或相关性,确定所述第一模型的监督结果。
可选地,所述第一确定单元,具体用于:
在所述第一距离和所述第二距离满足第一条件的情况下,确定所述第一模型失效,所述第一条件包括以下至少一项:
所述第一距离与所述第二距离的差异值大于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,存在F组第一目标测量信息,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比大于第二阈值;
第一距离分布和第二距离分布之间的第三距离大于或等于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
可选地,所述第一确定单元,具体用于:
在所述第一距离和所述第二距离满足第二条件的情况下,确定所述第一模型有效,所述第二条件包括以下至少一项:
所述第一距离与所述第二距离的差异值小于或等于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,第一目标测量信息的组数小于或等于F,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比小于或等于第二阈值;
第一距离分布和第二距离分布之间的第三距离小于第三阈值,所述第一距离分布为在
目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
可选地,所述一组目标测量信息包括以下至少一项:
至少两个来自不同终端的目标测量信息;
至少两个来自同一终端的不同时刻的目标测量信息。
可选地,第一处理单元,具体用于:
将一组目标测量信息输入第二模型,并获得所述第二模型输出的第二目标信息,其中,所述第二目标信息包括:
所述一组目标测量信息中的至少两个目标测量信息各自对应的位置之间的距离。
可选地,所述第三距离包括以下至少一项:
所述第一距离分布和所述第二距离分布的KL散度;
所述第一距离分布和所述第二距离分布的交叉熵;
所述第一距离分布和所述第二距离分布的累积概率密度曲线的最大垂直距离。
可选地,所述第一模型的监督结果包括以下至少一项:
满足所述第一条件的目标测量信息的组数;
满足所述第一条件的目标测量信息的占比;
满足所述第二条件的目标测量信息的组数;
满足所述第二条件的目标测量信息的占比;
所述第三距离;
所述第一模型有效的置信度;
所述第一模型有效的指示;
时间戳信息。
可选地,信息处理装置700还包括以下至少一项:
第三处理模块,用于对所述目标测量信息进行第一预处理,其中,所述第一模型输入的数据包括所述第一预处理后的目标测量信息;
第四处理模块,用于对所述目标测量信息进行第二预处理,其中,所述第二模型输入的数据包括所述第二预处理后的目标测量信息。
可选地,信息处理装置700还包括以下至少一项:
第五处理模块,用于对所述第一模型的输出信息进行第一后处理,其中,所述第一目标信息包括经所述第一后处理得到的信息;
第六处理模块,用于对所述第二模型的输出信息进行第二后处理,其中,所述第二目标信息包括经所述第二后处理得到的信息。
可选地,所述目标测量信息包括T个TRP或小区关联的测量信息,T为大于或等于1的整数。
可选地,用于估计得到所述目标测量信息的目标参考信号包括M个TRP或小区关联的参考信号,M为大于或等于1的整数。
本申请实施例中的信息处理装置700可以是电子设备,例如具有操作系统的电子设备,也可以是电子设备中的部件,例如集成电路或芯片。该电子设备可以是终端,也可以为除终端之外的其他设备。示例性的,终端可以包括但不限于上述所列举的终端11的类型,其他设备可以为服务器、网络附属存储器(Network Attached Storage,NAS)等,本申请实施例不作具体限定。
本申请实施例提供的信息处理装置700能够实现图2所示方法实施例实现的各个过程,并达到相同的技术效果,为避免重复,这里不再赘述。
请参阅图8,本申请实施例提供的信息处理装置800可以是第一节点内的装置,该第一节点可以是网络侧设备或是除了网络侧设备之外的其他设备,如服务器。
图8所示,该信息处理装置800可以包括以下模块:
第一接收模块801,用于接收来自终端的第一信息,其中,所述第一信息包括目标测量信息,或者所述第一信息包括第一目标信息和目标测量信息,所述第一目标信息为将所述目标测量信息输入第一模型后,由所述第一模型输出的信息,所述第一目标信息与所述终端的位置相关;
第二处理模块802,用于将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,所述第二目标信息与所述终端的位置相关;
第一发送模块803,用于向所述终端发送第二信息和所述第二目标信息中的至少一项,其中,所述第一模型的监督结果基于所述第一目标信息和所述第二目标信息之间的关联关系确定,所述第二信息与所述第一模型的监督结果或有效性相关。
可选地,信息处理装置800还包括:
第一确定模块,用于根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果。
可选地,在所述第一节点向所述终端发送所述第二目标信息的情况下,信息处理装置800还包括:
第二接收模块,用于接收来自所述终端的第三信息,所述第三信息与所述第一模型的监督结果或有效性相关。
可选地,所述目标测量信息包括以下至少一项:
时域信道脉冲响应、信道频率响应、时延功率谱、信道能量响应、参考信号接收功率RSRP、参考信号接收路径功率RSRPP、参考信号接收质量RSRQ、信号与干扰加噪声比SINR、时延多普勒域信道。
可选地,所述第一目标信息包括以下至少一项:
位置信息;
视距传播路径LOS到达时延TOA;
LOS到达角AOA;
LOS离开角AOD;
波束质量;
信道压缩指示;
预编码矩阵指示PMI;
时域信道信息;
频域信道信息;
空域信道信息;
小区通信质量;
小区切换判断结果。
可选地,所述第一确定模块,具体用于:
获取一组第一目标信息之间的第一距离,以及获取一组第二目标信息之间的第二距离,其中,所述一组第一目标信息与所述一组第二目标信息分别基于所述第一模型和所述第二模型对同一组目标测量信息进行处理得到;
根据所述第一距离和所述第二距离之间的差异或相关性,确定所述第一模型的监督结果。
可选地,所述第一确定模块,具体用于:
在所述第一距离和所述第二距离满足第一条件的情况下,确定所述第一模型失效,所述第一条件包括以下至少一项:
所述第一距离与所述第二距离的差异值大于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,存在F组第一目标测量信息,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比大于第二阈值;
第一距离分布和第二距离分布之间的第三距离大于或等于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
可选地,所述第一确定模块,具体用于:
在所述第一距离和所述第二距离满足第二条件的情况下,确定所述第一模型有效,所述第二条件包括以下至少一项:
所述第一距离与所述第二距离的差异值小于或等于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,第一目标测量信息的组数小于或等于F,其中,同一组第一目标测量信息对应的第一距离和第二
距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比小于或等于第二阈值;
第一距离分布和第二距离分布之间的第三距离小于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
可选地,所述一组目标测量信息包括以下至少一项:
至少两个来自不同终端的目标测量信息;
至少两个来自同一终端的不同时刻的目标测量信息。
可选地,第二处理模块802,具体用于:
所述第一节点将一组目标测量信息输入第二模型,并获得所述第二模型输出的第二目标信息,其中,所述第二目标信息包括:
所述一组目标测量信息中的至少两个目标测量信息各自对应的位置之间的距离。
可选地,所述第三距离包括以下至少一项:
所述第一距离分布和所述第二距离分布的KL散度;
所述第一距离分布和所述第二距离分布的交叉熵;
所述第一距离分布和所述第二距离分布的累积概率密度曲线的最大垂直距离。
可选地,所述第一模型的监督结果包括以下至少一项:
满足所述第一条件的目标测量信息的组数;
满足所述第一条件的目标测量信息的占比;
满足所述第二条件的目标测量信息的组数;
满足所述第二条件的目标测量信息的占比;
所述第三距离;
所述第一模型有效的置信度;
所述第一模型有效的指示;
时间戳信息。
可选地,所述目标测量信息包括T个TRP或小区关联的测量信息,T为大于或等于1的整数。
可选地,用于估计得到所述目标测量信息的目标参考信号包括M个TRP或小区关联的参考信号,M为大于或等于1的整数。
本申请实施例提供的信息处理装置800能够实现图6所示方法实施例实现的各个过程,并达到相同的技术效果,为避免重复,这里不再赘述。
如图9所示,本申请实施例还提供一种通信设备900,包括处理器901和存储器902,存储器902上存储有可在所述处理器901上运行的程序或指令,例如,该通信设备900为
终端时,该程序或指令被处理器901执行时实现上述信息传输方法实施例的各个步骤,且能达到相同的技术效果。该通信设备900为网络侧设备时,该程序或指令被处理器901执行时实现上述信息处理方法实施例的各个步骤,且能达到相同的技术效果,为避免重复,这里不再赘述。
本申请实施例还提供一种终端,包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现如图2所示方法实施例中的步骤。该终端实施例与上述终端侧方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中,且能达到相同的技术效果。具体地,图10为实现本申请实施例的一种终端的硬件结构示意图。
该终端1000包括但不限于:射频单元1001、网络模块1002、音频输出单元1003、输入单元1004、传感器1005、显示单元1006、用户输入单元1007、接口单元1008、存储器1009以及处理器1010等中的至少部分部件。
本领域技术人员可以理解,终端1000还可以包括给各个部件供电的电源(比如电池),电源可以通过电源管理系统与处理器10 10逻辑相连,从而通过电源管理系统实现管理充电、放电以及功耗管理等功能。图10中示出的终端结构并不构成对终端的限定,终端可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置,在此不再赘述。
应理解的是,本申请实施例中,输入单元1004可以包括图形处理器(Graphics Processing Unit,GPU)10041和麦克风10042,图形处理器10041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元1006可包括显示面板10061,可以采用液晶显示器、有机发光二极管等形式来配置显示面板10061。用户输入单元1007包括触控面板10071以及其他输入设备10072中的至少一种。触控面板10071,也称为触摸屏。触控面板10071可包括触摸检测装置和触摸控制器两个部分。其他输入设备10072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆,在此不再赘述。
本申请实施例中,射频单元1001接收来自网络侧设备的下行数据后,可以传输给处理器1010进行处理;另外,射频单元1001可以向网络侧设备发送上行数据。通常,射频单元1001包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器1009可用于存储软件程序或指令以及各种数据。存储器1009可主要包括存储程序或指令的第一存储区和存储数据的第二存储区,其中,第一存储区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外,存储器1009可以包括易失性存储器或非易失性存储器。其中,非易失性存储器可以是只读存储器(Read-Only Memory,ROM)、可编程只读存储器(Programmable ROM,PROM)、可擦除可编程只读存储器(Erasable PROM,EPROM)、电可擦除可编程只读存储器(Electrically EPROM,EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory,RAM),静态随机存取存储器(Static RAM,SRAM)、动态随机存取存储
器(Dynamic RAM,DRAM)、同步动态随机存取存储器(Synchronous DRAM,SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM,DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM,ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM,SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM,DRRAM)。本申请实施例中的存储器1009包括但不限于这些和任意其它适合类型的存储器。
处理器1010可包括一个或多个处理单元;可选的,处理器1010集成应用处理器和调制解调处理器,其中,应用处理器主要处理涉及操作系统、用户界面和应用程序等的操作,调制解调处理器主要处理无线通信信号,如基带处理器。可以理解的是,上述调制解调处理器也可以不集成到处理器1010中。
其中,处理器1010,用于将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与所述终端的位置相关;
处理器1010和射频单元1001中的至少一者,用于获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和所述第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
可选地,在所述终端具有所述第一模型,且第一节点具有所述第二模型的情况下,处理器1010和射频单元1001执行的所述获取所述第一模型的监督结果,包括:
射频单元1001,用于向所述第一节点发送第一信息,所述第一信息包括所述目标测量信息,或者,所述第一信息包括所述第一目标信息和所述目标测量信息;
射频单元1001,还用于接收来自所述第一节点的第二目标信息,处理器1010,用于根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,或者,射频单元1001,还用于接收来自所述第一节点的第二信息,所述第二信息与所述第一模型的监督结果或有效性相关。
可选地,在所述终端具有所述第一模型和所述第二模型的情况下,处理器1010和射频单元1001执行的所述获取所述第一模型的监督结果,包括:
处理器1010,用于将所述目标测量信息输入所述第二模型,获得所述第二模型输出的所述第二目标信息;
处理器1010,还用于根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果。
可选地,射频单元1001,还用于向第一节点发送第三信息,所述第三信息与所述第一模型的监督结果或有效性相关。
可选地,所述目标测量信息包括以下至少一项:
时域信道脉冲响应、信道频率响应、时延功率谱、信道能量响应、参考信号接收功率RSRP、参考信号接收路径功率RSRPP、参考信号接收质量RSRQ、信号与干扰加噪声比SINR、时延多普勒域信道。
可选地,所述第一目标信息包括以下至少一项:
位置信息;
视距LOS径到达时间TOA;
LOS径到达角AOA;
LOS径离开角AOD;
LOS径参考信号时间差RSTD;
波束质量;
信道压缩指示;
预编码矩阵指示PMI;
时域信道信息;
频域信道信息;
空域信道信息;
小区通信质量;
小区切换判断结果。
可选地,处理器1010执行的所述根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,包括:
获取一组第一目标信息之间的第一距离,以及获取一组第二目标信息之间的第二距离,其中,所述一组第一目标信息与所述一组第二目标信息分别基于所述第一模型和所述第二模型对同一组目标测量信息进行处理得到;
根据所述第一距离和所述第二距离之间的差异或相关性,确定所述第一模型的监督结果。
可选地,处理器1010执行的所述根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:
在所述第一距离和所述第二距离满足第一条件的情况下,确定所述第一模型失效,所述第一条件包括以下至少一项:
所述第一距离与所述第二距离的差异值大于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,存在F组第一目标测量信息,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比大于第二阈值;
第一距离分布和第二距离分布之间的第三距离大于或等于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
可选地,处理器1010执行的所述根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:
在所述第一距离和所述第二距离满足第二条件的情况下,确定所述第一模型有效,所述第二条件包括以下至少一项:
所述第一距离与所述第二距离的差异值小于或等于第一阈值;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,第一目标测量信息的组数小于或等于F,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;
在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比小于或等于第二阈值;
第一距离分布和第二距离分布之间的第三距离小于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
可选地,所述一组目标测量信息包括以下至少一项:
至少两个来自不同终端的目标测量信息;
至少两个来自同一终端的不同时刻的目标测量信息。
可选地,处理器1010执行的所述将所述目标测量信息输入所述第二模型,获得所述第二模型输出的第二目标信息,包括:
将一组目标测量信息输入第二模型,并获得所述第二模型输出的第二目标信息,其中,所述第二目标信息包括:
所述一组目标测量信息中的至少两个目标测量信息各自对应的位置之间的距离。
可选地,所述第三距离包括以下至少一项:
所述第一距离分布和所述第二距离分布的KL散度;
所述第一距离分布和所述第二距离分布的交叉熵;
所述第一距离分布和所述第二距离分布的累积概率密度曲线的最大垂直距离。
可选地,所述第一模型的监督结果包括以下至少一项:
满足所述第一条件的目标测量信息的组数;
满足所述第一条件的目标测量信息的占比;
满足所述第二条件的目标测量信息的组数;
满足所述第二条件的目标测量信息的占比;
所述第三距离;
所述第一模型有效的置信度;
所述第一模型有效的指示;
时间戳信息。
可选地,处理器1010,还用于执行以下至少一项:
对所述目标测量信息进行第一预处理,其中,所述第一模型输入的数据包括所述第一预处理后的目标测量信息;
对所述目标测量信息进行第二预处理,其中,所述第二模型输入的数据包括所述第二预处理后的目标测量信息。
可选地,处理器1010,还用于执行以下至少一项:
对所述第一模型的输出信息进行第一后处理,其中,所述第一目标信息包括经所述第一后处理得到的信息;
对所述第二模型的输出信息进行第二后处理,其中,所述第二目标信息包括经所述第二后处理得到的信息。
可选地,所述目标测量信息包括T个TRP或小区关联的测量信息,T为大于或等于1的整数。
可选地,用于估计得到所述目标测量信息的目标参考信号包括M个TRP或小区关联的参考信号,M为大于或等于1的整数。
可以理解,本实施例中提及的各实现方式的实现过程可以参照如图2所示信息处理方法实施例的相关描述,并达到相同或相应的技术效果,为避免重复,在此不再赘述。
本申请实施例还提供一种网络侧设备,包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现如图6所示的方法实施例的步骤。该网络侧设备实施例与上述网络侧设备方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该网络侧设备实施例中,且能达到相同的技术效果。
具体地,本申请实施例还提供了一种网络侧设备。如图11所示,该网络侧设备1100包括:天线1101、射频装置1102、基带装置1103、处理器1104和存储器1105。天线1101与射频装置1102连接。在上行方向上,射频装置1102通过天线1101接收信息,将接收的信息发送给基带装置1103进行处理。在下行方向上,基带装置1103对要发送的信息进行处理,并发送给射频装置1102,射频装置1102对收到的信息进行处理后经过天线1101发送出去。
以上实施例中网络侧设备执行的方法可以在基带装置1103中实现,该基带装置1103包括基带处理器。
基带装置1103例如可以包括至少一个基带板,该基带板上设置有多个芯片,如图11所示,其中一个芯片例如为基带处理器,通过总线接口与存储器1105连接,以调用存储器1105中的程序,执行以上方法实施例中所示的网络设备操作。
该网络侧设备还可以包括网络接口1106,该接口例如为通用公共无线接口(Common Public Radio Interface,CPRI)。
具体地,本申请实施例的网络侧设备1100还包括:存储在存储器1105上并可在处理器1104上运行的指令或程序,处理器1104调用存储器1105中的指令或程序执行如图8所示
各模块执行的方法,并达到相同的技术效果,为避免重复,故不在此赘述。
具体地,本申请实施例还提供了一种网络侧设备。如图12所示,该网络侧设备1200包括:处理器1201、网络接口1202和存储器1203。其中,网络接口1202例如为通用公共无线接口(Common Public Radio Interface,CPRI)。
具体地,本申请实施例的网络侧设备1200还包括:存储在存储器1203上并可在处理器1201上运行的指令或程序,处理器1201调用存储器1203中的指令或程序执行如图8所示各模块执行的方法,并达到相同的技术效果,为避免重复,故不在此赘述。
本申请实施例还提供一种可读存储介质,所述可读存储介质上存储有程序或指令,该程序或指令被处理器执行时实现如图2或图6所示方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
其中,所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质,包括计算机可读存储介质,如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。在一些示例中,可读存储介质可以是非瞬态的可读存储介质。
本申请实施例另提供了一种芯片,所述芯片包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现如图2或图6所示方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
应理解,本申请实施例提到的芯片还可以称为系统级芯片,系统芯片,芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品,所述计算机程序/程序产品被存储在存储介质中,所述计算机程序/程序产品被至少一个处理器执行以实现如图2或图6所示方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
本申请实施例还提供了一种通信系统,包括:终端及第一节点,所述终端可用于执行如图2所示的信息处理方法的步骤,所述第一节点可用于执行如图6所示的信息处理方法的步骤。
需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外,需要指出的是,本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能,还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能,例如,可以按不同于所描述的次序来执行所描述的方法,并且还可以添加、省去或组合各种步骤。另外,参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助计算机软件产品加必需的通用硬件平台的方式来实现,当然也可以通过硬件。该计算
机软件产品存储在存储介质(如ROM、RAM、磁碟、光盘等)中,包括若干指令,用以使得终端或者网络侧设备执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述,但是本申请并不局限于上述的具体实施方式,上述的具体实施方式仅仅是示意性的,而不是限制性的,本领域的普通技术人员在本申请的启示下,在不脱离本申请宗旨和权利要求所保护的范围情况下,还可做出很多形式的实施方式,这些实施方式均属于本申请的保护之内。
Claims (36)
- 一种信息处理方法,包括:终端将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与所述终端的位置相关;所述终端获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
- 根据权利要求1所述的方法,其中,在所述终端具有所述第一模型,且第一节点具有所述第二模型的情况下,所述终端获取所述第一模型的监督结果,包括:所述终端向所述第一节点发送第一信息,所述第一信息包括所述目标测量信息,或者,所述第一信息包括所述第一目标信息和所述目标测量信息;所述终端接收来自所述第一节点的第二目标信息,并根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,或者,所述终端接收来自所述第一节点的第二信息,所述第二信息与所述第一模型的监督结果或有效性相关。
- 根据权利要求1所述的方法,其中,在所述终端具有所述第一模型和所述第二模型的情况下,所述终端获取所述第一模型的监督结果,包括:所述终端将所述目标测量信息输入所述第二模型,获得所述第二模型输出的所述第二目标信息;所述终端根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果。
- 根据权利要求3所述的方法,其中,所述方法还包括:所述终端向第一节点发送第三信息,所述第三信息与所述第一模型的监督结果或有效性相关。
- 根据权利要求1至4中任一项所述的方法,其中,所述目标测量信息包括以下至少一项:时域信道脉冲响应、信道频率响应、时延功率谱、信道能量响应、参考信号接收功率RSRP、参考信号接收路径功率RSRPP、参考信号接收质量RSRQ、信号与干扰加噪声比SINR、时延多普勒域信道。
- 根据权利要求1至5中任一项所述的方法,其中,所述第一目标信息包括以下至少一项:位置信息;视距LOS径到达时间TOA;LOS径到达角AOA;LOS径离开角AOD;LOS径参考信号时间差RSTD;波束质量;信道压缩指示;预编码矩阵指示PMI;时域信道信息;频域信道信息;空域信道信息;小区通信质量;小区切换判断结果。
- 根据权利要求2至6中任一项所述的方法,其中,所述终端根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,包括:所述终端获取一组第一目标信息之间的第一距离,以及获取一组第二目标信息之间的第二距离,其中,所述一组第一目标信息与所述一组第二目标信息分别基于所述第一模型和所述第二模型对同一组目标测量信息进行处理得到;所述终端根据所述第一距离和所述第二距离之间的差异或相关性,确定所述第一模型的监督结果。
- 根据权利要求7所述的方法,其中,所述终端根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:在所述第一距离和所述第二距离满足第一条件的情况下,所述终端确定所述第一模型失效,所述第一条件包括以下至少一项:所述第一距离与所述第二距离的差异值大于第一阈值;在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,存在F组第一目标测量信息,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比大于第二阈值;第一距离分布和第二距离分布之间的第三距离大于或等于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
- 根据权利要求7或8所述的方法,其中,所述终端根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:在所述第一距离和所述第二距离满足第二条件的情况下,所述终端确定所述第一模型有效,所述第二条件包括以下至少一项:所述第一距离与所述第二距离的差异值小于或等于第一阈值;在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,第一目标测量信息的组数小于或等于F,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比小于或等于第二阈值;第一距离分布和第二距离分布之间的第三距离小于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
- 根据权利要求7至9中任一项所述的方法,其中,所述一组目标测量信息包括以下至少一项:至少两个来自不同终端的目标测量信息;至少两个来自同一终端的不同时刻的目标测量信息。
- 根据权利要求10所述的方法,其中,所述终端将所述目标测量信息输入所述第二模型,获得所述第二模型输出的第二目标信息,包括:所述终端将一组目标测量信息输入第二模型,并获得所述第二模型输出的第二目标信息,其中,所述第二目标信息包括:所述一组目标测量信息中的至少两个目标测量信息各自对应的位置之间的距离。
- 根据权利要求8或9所述的方法,其中,所述第三距离包括以下至少一项:所述第一距离分布和所述第二距离分布的KL散度;所述第一距离分布和所述第二距离分布的交叉熵;所述第一距离分布和所述第二距离分布的累积概率密度曲线的最大垂直距离。
- 根据权利要求8至12中任一项所述的方法,其中,所述第一模型的监督结果包括以下至少一项:满足第一条件的目标测量信息的组数;满足所述第一条件的目标测量信息的占比;满足第二条件的目标测量信息的组数;满足所述第二条件的目标测量信息的占比;第三距离;所述第一模型有效的置信度;所述第一模型有效的指示;时间戳信息。
- 根据权利要求1至13中任一项所述的方法,其中,所述方法还包括以下至少一项:所述终端对所述目标测量信息进行第一预处理,其中,所述第一模型输入的数据包括 所述第一预处理后的目标测量信息;所述终端对所述目标测量信息进行第二预处理,其中,所述第二模型输入的数据包括所述第二预处理后的目标测量信息。
- 根据权利要求1至14中任一项所述的方法,其中,所述方法还包括以下至少一项:所述终端对所述第一模型的输出信息进行第一后处理,其中,所述第一目标信息包括经所述第一后处理得到的信息;所述终端对所述第二模型的输出信息进行第二后处理,其中,所述第二目标信息包括经所述第二后处理得到的信息。
- 根据权利要求1至15中任一项所述的方法,其中,所述目标测量信息包括T个TRP或小区关联的测量信息,T为大于或等于1的整数。
- 根据权利要求1至16中任一项所述的方法,其中,用于估计得到所述目标测量信息的目标参考信号包括M个TRP或小区关联的参考信号,M为大于或等于1的整数。
- 一种信息处理方法,所述方法包括:第一节点接收来自终端的第一信息,其中,所述第一信息包括目标测量信息,或者所述第一信息包括第一目标信息和目标测量信息,所述第一目标信息为将所述目标测量信息输入第一模型后,由所述第一模型输出的信息,所述第一目标信息与所述终端的位置相关;所述第一节点将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,所述第二目标信息与所述终端的位置相关;所述第一节点向所述终端发送第二信息和所述第二目标信息中的至少一项,其中,所述第一模型的监督结果基于所述第一目标信息和所述第二目标信息之间的关联关系确定,所述第二信息与所述第一模型的监督结果或有效性相关。
- 根据权利要求18所述的方法,其中,在所述第一节点向所述终端发送第二信息之前,所述方法还包括:所述第一节点根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果。
- 根据权利要求18所述的方法,其中,在所述第一节点向所述终端发送所述第二目标信息的情况下,所述方法还包括:所述第一节点接收来自所述终端的第三信息,所述第三信息与所述第一模型的监督结果或有效性相关。
- 根据权利要求18至20中任一项所述的方法,其中,所述目标测量信息包括以下至少一项:时域信道脉冲响应、信道频率响应、时延功率谱、信道能量响应、参考信号接收功率RSRP、参考信号接收路径功率RSRPP、参考信号接收质量RSRQ、信号与干扰加噪声比SINR、时延多普勒域信道。
- 根据权利要求18至21中任一项所述的方法,其中,所述第一目标信息包括以下 至少一项:位置信息;视距传播路径LOS到达时延TOA;LOS到达角AOA;LOS离开角AOD;波束质量;信道压缩指示;预编码矩阵指示PMI;时域信道信息;频域信道信息;空域信道信息;小区通信质量;小区切换判断结果。
- 根据权利要求19至22中任一项所述的方法,其中,所述第一节点根据所述第一目标信息和所述第二目标信息之间的关联关系,确定所述第一模型的监督结果,包括:所述第一节点获取一组第一目标信息之间的第一距离,以及获取一组第二目标信息之间的第二距离,其中,所述一组第一目标信息与所述一组第二目标信息分别基于所述第一模型和所述第二模型对同一组目标测量信息进行处理得到;所述第一节点根据所述第一距离和所述第二距离之间的差异或相关性,确定所述第一模型的监督结果。
- 根据权利要求23所述的方法,其中,所述第一节点根据所述第一距离和所述第二距离之间的差异,确定所述第一模型的监督结果,包括:在所述第一距离和所述第二距离满足第一条件的情况下,所述第一节点确定所述第一模型失效,所述第一条件包括以下至少一项:所述第一距离与所述第二距离的差异值大于第一阈值;在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,存在F组第一目标测量信息,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比大于第二阈值;第一距离分布和第二距离分布之间的第三距离大于或等于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
- 根据权利要求23或24所述的方法,其中,所述第一节点根据所述第一距离和所 述第二距离之间的差异,确定所述第一模型的监督结果,包括:在所述第一距离和所述第二距离满足第二条件的情况下,所述第一节点确定所述第一模型有效,所述第二条件包括以下至少一项:所述第一距离与所述第二距离的差异值小于或等于第一阈值;在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,第一目标测量信息的组数小于或等于F,其中,同一组第一目标测量信息对应的第一距离和第二距离的差异值大于第一阈值,F为大于或等于1的整数,R为大于或等于F的整数;在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息内,所述第一目标测量信息的组数占比小于或等于第二阈值;第一距离分布和第二距离分布之间的第三距离小于第三阈值,所述第一距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第一距离的分布;所述第二距离分布为在目标时间范围测量得到的全部组目标测量信息或预设的R组目标测量信息各自对应的第二距离的分布。
- 根据权利要求23至25中任一项所述的方法,其中,所述一组目标测量信息包括以下至少一项:至少两个来自不同终端的目标测量信息;至少两个来自同一终端的不同时刻的目标测量信息。
- 根据权利要求26所述的方法,其中,所述第一节点将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,包括:所述第一节点将一组目标测量信息输入第二模型,并获得所述第二模型输出的第二目标信息,其中,所述第二目标信息包括:所述一组目标测量信息中的至少两个目标测量信息各自对应的位置之间的距离。
- 根据权利要求24或25所述的方法,其中,所述第三距离包括以下至少一项:所述第一距离分布和所述第二距离分布的KL散度;所述第一距离分布和所述第二距离分布的交叉熵;所述第一距离分布和所述第二距离分布的累积概率密度曲线的最大垂直距离。
- 根据权利要求25至28中任一项所述的方法,其中,所述第一模型的监督结果包括以下至少一项:满足第一条件的目标测量信息的组数;满足所述第一条件的目标测量信息的占比;满足第二条件的目标测量信息的组数;满足所述第二条件的目标测量信息的占比;第三距离;所述第一模型有效的置信度;所述第一模型有效的指示;时间戳信息。
- 根据权利要求18至29中任一项所述的方法,其中,所述目标测量信息包括T个TRP或小区关联的测量信息,T为大于或等于1的整数。
- 根据权利要求18至30中任一项所述的方法,其中,用于估计得到所述目标测量信息的目标参考信号包括M个TRP或小区关联的参考信号,M为大于或等于1的整数。
- 一种信息处理装置,包括:第一处理模块,用于将目标测量信息输入第一模型,获得所述第一模型输出的第一目标信息,所述第一目标信息与终端的位置相关;获取模块,用于获取所述第一模型的监督结果,其中,所述第一模型的监督结果基于所述第一目标信息和第二目标信息之间的关联关系确定,所述第二目标信息为将所述目标测量信息输入第二模型后,由所述第二模型输出的信息,所述第二目标信息与所述终端的位置相关。
- 一种信息处理装置,包括:第一接收模块,用于接收来自终端的第一信息,其中,所述第一信息包括目标测量信息,或者所述第一信息包括第一目标信息和目标测量信息,所述第一目标信息为将所述目标测量信息输入第一模型后,由所述第一模型输出的信息,所述第一目标信息与所述终端的位置相关;第二处理模块,用于将所述目标测量信息输入第二模型,获得所述第二模型输出的第二目标信息,所述第二目标信息与所述终端的位置相关;第一发送模块,用于向所述终端发送第二信息和所述第二目标信息中的至少一项,其中,所述第一模型的监督结果基于所述第一目标信息和所述第二目标信息之间的关联关系确定,所述第二信息与所述第一模型的监督结果或有效性相关。
- 一种终端,包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至17中任一项所述的信息处理方法的步骤。
- 一种网络侧设备,包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求18至31中任一项所述的信息处理方法的步骤。
- 一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如权利要求1至17中任一项所述的信息处理方法的步骤,或者实现如权利要求18至31中任一项所述的信息处理方法的步骤。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310371177.1 | 2023-04-07 | ||
| CN202310371177.1A CN118780013A (zh) | 2023-04-07 | 2023-04-07 | 信息处理方法、信息处理装置、终端及网络侧设备 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2024208167A1 true WO2024208167A1 (zh) | 2024-10-10 |
Family
ID=92971077
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2024/085422 Pending WO2024208167A1 (zh) | 2023-04-07 | 2024-04-02 | 信息处理方法、信息处理装置、终端及网络侧设备 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN118780013A (zh) |
| WO (1) | WO2024208167A1 (zh) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112748453A (zh) * | 2020-08-13 | 2021-05-04 | 腾讯科技(深圳)有限公司 | 道路侧定位方法、装置、设备及存储介质 |
| WO2021206189A1 (ko) * | 2020-04-07 | 2021-10-14 | 엘지전자 주식회사 | 학습 기반의 신호 수신 방법 및 장치 |
| CN113543305A (zh) * | 2020-04-22 | 2021-10-22 | 维沃移动通信有限公司 | 定位方法、通信设备和网络设备 |
| CN114077912A (zh) * | 2020-08-14 | 2022-02-22 | 华为技术有限公司 | 数据预测方法以及数据预测装置 |
| CN114521012A (zh) * | 2020-11-18 | 2022-05-20 | 维沃移动通信有限公司 | 定位方法、装置、终端设备、基站及位置管理服务器 |
-
2023
- 2023-04-07 CN CN202310371177.1A patent/CN118780013A/zh active Pending
-
2024
- 2024-04-02 WO PCT/CN2024/085422 patent/WO2024208167A1/zh active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021206189A1 (ko) * | 2020-04-07 | 2021-10-14 | 엘지전자 주식회사 | 학습 기반의 신호 수신 방법 및 장치 |
| CN113543305A (zh) * | 2020-04-22 | 2021-10-22 | 维沃移动通信有限公司 | 定位方法、通信设备和网络设备 |
| CN112748453A (zh) * | 2020-08-13 | 2021-05-04 | 腾讯科技(深圳)有限公司 | 道路侧定位方法、装置、设备及存储介质 |
| CN114077912A (zh) * | 2020-08-14 | 2022-02-22 | 华为技术有限公司 | 数据预测方法以及数据预测装置 |
| CN114521012A (zh) * | 2020-11-18 | 2022-05-20 | 维沃移动通信有限公司 | 定位方法、装置、终端设备、基站及位置管理服务器 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118780013A (zh) | 2024-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116567806B (zh) | 基于人工智能ai模型的定位方法及通信设备 | |
| WO2023098662A1 (zh) | 定位方法及通信设备 | |
| WO2023098661A1 (zh) | 定位方法及通信设备 | |
| US20240373257A1 (en) | Information interaction method and apparatus, and communication device | |
| US20240323741A1 (en) | Measurement method and apparatus, device, and storage medium | |
| US20250227507A1 (en) | Ai model processing method and apparatus, and communication device | |
| WO2024008111A1 (zh) | 数据采集方法及装置 | |
| WO2023040888A1 (zh) | 数据传输方法及装置 | |
| WO2024125525A1 (zh) | Ai算力上报方法、终端及网络侧设备 | |
| WO2024083004A1 (zh) | Ai模型配置方法、终端及网络侧设备 | |
| WO2024208167A1 (zh) | 信息处理方法、信息处理装置、终端及网络侧设备 | |
| WO2024093997A1 (zh) | 确定模型适用性的方法、装置及通信设备 | |
| WO2024067434A1 (zh) | Rs配置方法、装置、终端及网络侧设备 | |
| WO2024027576A1 (zh) | 一种ai网络模型的性能监督方法、装置和通信设备 | |
| WO2023155839A1 (zh) | Ai模型的在线学习方法、装置、通信设备及可读存储介质 | |
| WO2025108195A1 (zh) | 模型确定方法、装置及通信设备 | |
| WO2025036223A1 (zh) | 信息上报方法、信息接收方法及设备 | |
| WO2025092999A1 (zh) | 模型性能监督方法、装置及设备 | |
| WO2024235140A1 (zh) | 定位处理方法、装置及设备 | |
| CN120456087A (zh) | 一种定位测量处理方法、装置、通信设备及存储介质 | |
| WO2025140432A1 (zh) | 信息上报方法及装置 | |
| WO2025092998A1 (zh) | 信息传输方法、装置及设备 | |
| WO2025108390A1 (zh) | 基于ai模型的定位方法、装置、设备及可读存储介质 | |
| CN120602970A (zh) | 测量处理方法、测量配置方法、装置、设备及可读存储介质 | |
| WO2024217405A1 (zh) | 终端定位方法、装置、终端、网络侧设备及介质 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 24784251 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |