WO2025177724A1 - Information processing device, program, and information processing method - Google Patents

Abstract

Provided is an information processing device comprising: a teacher data acquisition unit that acquires teacher data including the positional relationship between a first antenna provided to a mobile body and an array antenna including a plurality of second antennas, and channel response data indicating a channel response, sampled by space and frequency, between the first antenna and the array antenna when the first antenna and the array antenna are in the aforementioned positional relationship; and a training execution unit that uses a plurality of items of the teacher data acquired by the teacher data acquisition unit to execute training of a neural network that receives input of the channel response data and outputs an estimated direction of the first antenna with reference to the array antenna.

Description

Information processing device, program, and information processing method

The present invention relates to an information processing device, a program, and an information processing method.

Non-Patent Document 1 describes robust Loss-MIMO (Line of Sight-Multi Input Multi Output) transmission.
[Prior art documents]
[Non-Patent Documents]
[Non-patent Document 1] M. Tawada, Y. Ohta and A. Nagate, "Design of Robust LoS-MIMO Transmission in HAPS Feeder Link," 2022 IEEE 96th Vehicular Technology Conference (VTC2022-Fall), London, United Kingdom, 2022, pp. 1-7, doi: 10.1109/VTC2022-Fall57202.2022.10012950.

General disclosure

(Means for solving problems)

According to one embodiment of the present invention, an information processing device is provided. The information processing device may include a training data acquisition unit that acquires training data including the positional relationship between a first antenna possessed by a mobile object and an array antenna including a plurality of second antennas, and channel response data indicating the channel response sampled in space and frequency between the first antenna and the array antenna when the first antenna and the array antenna are in the positional relationship. The information processing device may also include a learning execution unit that uses the plurality of training data acquired by the training data acquisition unit to execute training of a neural network that receives the channel response data as input and outputs an estimated direction of the first antenna relative to the array antenna. The channel response data may include channel vectors for each of a plurality of frequencies.

In the information processing device, the learning execution unit may input the channel response data included in the training data to the neural network, and update the neural network using the cosine similarity between the estimated direction vector output from the neural network and the vector of the direction of the first antenna relative to the array antenna indicated by the positional relationship included in the training data as a loss function. The learning execution unit may update the neural network so that the cosine similarity becomes 1.

In any of the information processing devices, the first antenna and the array antenna may be used for LoS-MIMO transmission, and the multiple second antennas may be arranged at intervals that satisfy the requirements for LoS-MIMO transmission.

In the information processing device, the mobile object may be a HAPS, and the array antenna may be located on the ground.

According to one embodiment of the present invention, a program is provided for causing a computer to function as the information processing device.

According to one embodiment of the present invention, there is provided an information processing method executed by a computer. The information processing method may include a training data acquisition step of acquiring training data including the positional relationship between a first antenna possessed by a mobile object and an array antenna including a plurality of second antennas, and channel response data indicating the channel response sampled in space and frequency between the first antenna and the array antenna when the first antenna and the array antenna are in the positional relationship. The information processing method may also include a learning execution step of using the plurality of training data acquired in the training data acquisition step to execute training of a neural network that inputs the channel response data and outputs an estimated direction of the first antenna relative to the array antenna.

According to one embodiment of the present invention, there is provided an information processing device. The information processing device may include a storage unit that stores a neural network that receives channel response data generated using a plurality of training data including the positional relationship between a first antenna provided on a first mobile object and a first array antenna including a plurality of second antennas, and channel response data indicating a channel response sampled in space and frequency between the first antenna and the first array antenna when the first antenna and the first array antenna are in the positional relationship, and outputs the direction of the first antenna relative to the first array antenna. The information processing device may include a channel response acquisition unit that acquires channel response data sampled in space and frequency between a third antenna provided on a second mobile object and a second array antenna including a plurality of fourth antennas. The information processing device may include a direction acquisition unit that inputs the channel response data acquired by the channel response acquisition unit to the neural network and acquires the direction of the third antenna relative to the second array antenna output from the neural network. The information processing device may include a communication adjustment unit that adjusts wireless communication between the third antenna and the second array antenna based on the direction acquired by the direction acquisition unit.

The third antenna and the second array antenna may be used for LoS-MIMO transmission, and the multiple fourth antennas may be arranged at intervals that meet the requirements for LoS-MIMO transmission.

According to one embodiment of the present invention, a program is provided for causing a computer to function as the information processing device.

According to one embodiment of the present invention, there is provided an information processing method executed by a computer. The information processing method may include an NN acquisition step of acquiring a neural network that receives as input the channel response generated using a plurality of training data sets including the positional relationship between a first antenna equipped on a first mobile body and a first array antenna including a plurality of second antennas, and a channel response sampled in space and frequency between the first antenna and the first array antenna when the first antenna and the first array antenna are in the positional relationship, and outputs the direction of the first antenna relative to the first array antenna. The information processing method may include a channel response acquisition step of acquiring a channel response sampled in space and frequency between a third antenna equipped on a second mobile body and a second array antenna including a plurality of fourth antennas. The information processing method may include a direction acquisition step of inputting the channel response acquired in the channel response acquisition step to the neural network, and acquiring the direction of the third antenna relative to the second array antenna output from the neural network. The information processing method may include a communication adjustment step of adjusting wireless communication between the third antenna and the second array antenna based on the direction acquired in the direction acquisition step.

Note that the above summary of the invention does not list all of the necessary features of the present invention. Subcombinations of these features may also constitute inventions.

10A and 10B show an example of the relationship between a moving body 200 and an array antenna 302 in the system 10 according to the present embodiment. FIG. 1 is an explanatory diagram for schematically explaining DOA estimation by a neural network used by the information processing device 100. FIG. 10 is an explanatory diagram illustrating a method for acquiring a channel response. FIG. 2 is an explanatory diagram for roughly explaining the processing content of the information processing device 100. FIG. 2 is an explanatory diagram for explaining input data 110. FIG. 2 is an explanatory diagram for explaining input data 110. FIG. 2 is an explanatory diagram for explaining input data 110. FIG. 2 is an explanatory diagram for explaining input data 110. 2 illustrates an example of a functional configuration of an information processing device 100. 1 shows a schematic diagram of an example of the configuration of a HAPS 400, which is an example of a moving body 200. 1 shows an example of a hardware configuration of a computer 1200 that functions as the information processing device 100.

The present invention will be explained below through embodiments of the invention, but the following embodiments do not limit the scope of the invention as claimed. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

In recent years, mobile communication services using HAPS (High-Altitude Platform Station), which travels in the stratosphere at an altitude of 20 km, have been attracting attention. The feeder link between the fixed terrestrial station and the HAPS requires a high-capacity communication line to accommodate the communication traffic of terminals within the area formed. An effective solution to this problem is to improve channel capacity through the spatial multiplexing effect of LoS-MIMO, which accommodates changes in the distance and angle between the transmitter and receiver as the HAPS moves. Incidentally, millimeter wave bands such as 39 GHz are globally specified as frequencies allocated to HAPS communication systems. Considering that HAPS orbits at an altitude of 20 km, propagation loss increases, making the use of high-gain antennas effective in ensuring the desired SNR (Signal-to-Noise Ratio). However, increasing antenna gain narrows the beamwidth, making it important to estimate the direction of arrival of radio waves between the ground station and HAPS and to track the beam accordingly while the HAPS is moving. While direction calculations can be performed by acquiring information on the HAPS's position and attitude using sensors installed on the HAPS, accuracy and reliability depend on the sensors. Furthermore, there are inherent risks in operating precision sensors in the harsh environment of the HAPS. While introducing a radar system is an option, adding new hardware to the HAPS is undesirable due to payload weight limitations, and introducing positioning signals is undesirable due to power and frequency usage considerations. To achieve a compact system configuration, it is desirable to perform direction estimation using an array antenna and signals for LoS-MIMO transmission, without adding dedicated antennas or signals for direction estimation. Generally, in array signal processing such as MUSIC (Multiple Signal Classification) and ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques), the antenna element spacing is set to half a wavelength or less to suppress images, but to achieve LoS-MIMO transmission, the element spacing becomes several hundred wavelengths. In this case, it becomes difficult to apply conventional direction estimation algorithms. The information processing device 100 according to this embodiment employs a direction estimation method using a neural network (NN) for array antennas with wide element spacing, such as those used in LoS-MIMO transmission systems.

FIG. 1 shows a schematic example of the relationship between a mobile object 200 and an array antenna 302 in a system 10 according to this embodiment. The system 10 includes an information processing device 100, which is not shown in FIG. 1. The system 10 may include an array antenna 302 including a plurality of antennas 310, and a communication device 300 that performs communication using the array antenna 302.

In FIG. 1, the mobile object 200 is illustrated as a HAPS, but this is not limiting. The mobile object 200 may be an airplane or unmanned aerial vehicle that moves through the air. The mobile object 200 may also be a car or the like that moves on the ground. The mobile object 200 is equipped with an antenna 210. The antenna 210 may be an array antenna. The antenna 210 does not have to be an array antenna. The antenna 210 may be an example of a first antenna, and the antenna 310 may be an example of a second antenna.

In FIG. 1, the array antenna 302 is illustrated as a UXA (Uniform Cross Array), but this is not limited to this. The array antenna 302 may also be a ULA (Uniform Linear Array). The communication device 300 may be any device that can communicate with the mobile unit 200 using the array antenna 302. If the mobile unit 200 is a HAPS, the communication device 300 may be a ground station, and the mobile unit 200 and the communication device 300 may perform feeder link communication.

As mentioned above, conventional subspace methods such as MUSIC and ESPRIT exist, but these methods impose conditions on the element spacing of the array antenna, which is usually required to be set to λ/2 or less.

Array antennas for LoS-MIMO also have element spacing requirements, and in the case of ULA, the following formula 1 must be satisfied.

where d is the element spacing, D is the distance between transmitter and receiver, c is the speed of light, l is the number of ULA elements, and fc is the center frequency. If D = 20 km, l = 5, and fc = 39 GHz, then d = 721λ, which violates the element spacing requirements for conventional array signal processing.

Figure 2 is an explanatory diagram that provides an overview of DOA estimation using a neural network, as used by the information processing device 100.

The information processing device 100 employs DOA estimation using a neural network as an alternative to conventional array signal processing. A known characteristic of neural networks is the Universal Approximation Theorem, which states that any function can be expressed with any degree of precision by increasing the number of nodes, as long as there is one or more hidden layers. The information processing device 100 takes advantage of this characteristic and uses a neural network as a black box for DOA estimation.

The information processing device 100 takes the explanatory variable of the direction of arrival, which is the objective variable, as input data, and treats this input data as a channel response sampled at spatial frequency. The information processing device 100 takes the direction of arrival, which is the objective variable, as output data, treats the output data as a direction vector, and uses cosine similarity as the loss function.

Figure 3 is an explanatory diagram for explaining how to acquire a channel response. If the mobile body 200 side is a transmitting array and the array antenna 302 is a receiving array, the array antenna 302 acquires a channel response from any antenna on the mobile body 200 by channel sounding as shown in Figure 3. The information processing device 100 acquires data on the direction of arrival of the radio waves when acquiring this channel response as training data.

The information processing device 100 may collect data on the channel response and direction of arrival of radio waves as actual data. The information processing device 100 may collect data on the channel response and direction of arrival of radio waves through simulation.

4 is an explanatory diagram for outlining the processing contents of the information processing device 100. The information processing device 100 uses channel response information included in the teacher data as input data 110 of the DOA estimation NN 120. The information processing device 100 outputs data 130 as a three-dimensional direction vector v and an antenna position vector of a mobile object 200, which is a source of radio waves.
The loss function L is the cosine similarity between the two, and the DOA estimation NN 120 is optimized so that L becomes 1.

5, 6, 7, and 8 are explanatory diagrams for explaining input data 110. A channel vector h between one transmitting antenna and n receiving antennas can be expressed by equation 112 in Fig. 5 using the direct component h _LoS , the multipath component h _NLoS , and the power ratio K factor K, assuming Nakagami-Rice fading. Normally, in a line-of-sight environment such as a HAPS communication system, the K factor takes a large value, so the LoS component becomes dominant, and under the condition of K>>1,
(h and h _LoS are nearly equal.) In LoS-MIMO, h _LoS is determined by the relative positions of the transmitting and receiving antennas, and can be expressed by equation 114 in FIG. 6 using the distance di from the transmitting antenna to the i-th receiving antenna.

Although this is channel response data sampled at the antenna arrangement position, array signal processing cannot accurately estimate data when the element spacing is large due to ambiguity that occurs. Therefore, the information processing device 100 may use a channel vector of m frequency components contained in the communication signal between the mobile object 200 and the communication device 300. With frequency indices as subscripts, the input data u can be expressed by equation 116 in Figure 7. The information processing device 100 may convert each element into an argument as shown in equation 118 in Figure 8, and use this as input data u.

Figure 9 shows an example of the functional configuration of the information processing device 100. The information processing device 100 includes a memory unit 140, a measurement data receiving unit 142, a teacher data acquisition unit 144, a learning execution unit 146, an NN output unit 148, a channel response acquisition unit 150, a direction acquisition unit 152, and a communication adjustment unit 154. Note that it is not essential for the information processing device 100 to include all of these units.

The measurement data receiving unit 142 receives measurement data. The measurement data includes the positional relationship between the antenna 210 of the mobile unit 200 and the array antenna 302 including multiple antennas 310, and channel response data indicating the channel response sampled in space and frequency between the antenna 210 and the array antenna 302 when the antenna 210 and the array antenna 302 are in this positional relationship. The measurement data receiving unit 142 may receive measurement data from the communication device 300 including channel response data indicating the channel response measured by channel sounding using the array antenna 302 by the communication device 300 in a situation where the positional relationship between the antenna 210 and the array antenna 302 is known. The measurement data receiving unit 142 stores the received measurement data in the memory unit 140 as training data.

Antenna 210 and array antenna 302 may be used for LoS-MIMO transmission. Multiple antennas 310 may be arranged at intervals that meet the requirements for LoS-MIMO transmission.

The teacher data acquisition unit 144 acquires multiple pieces of teacher data from the storage unit 140. The teacher data acquisition unit 144 may acquire all of the teacher data stored in the storage unit 140. The teacher data acquisition unit 144 may also acquire multiple pieces of teacher data selected by a user of the information processing device 100, etc., from the teacher data stored in the storage unit 140.

The learning execution unit 146 uses multiple pieces of training data acquired by the training data acquisition unit 144 to perform neural network training, taking channel response data as input and outputting the estimated direction of the antenna 210 relative to the array antenna 302. As described in Figures 5 to 8, the input channel response data may include channel vectors for each of multiple frequencies.

The learning execution unit 146 may input the channel response data included in the training data into the neural network, and update the neural network using the cosine similarity between the estimated direction vector output from the neural network and the vector of the direction of the antenna 210 relative to the array antenna 302, as indicated by the positional relationship included in the training data, as a loss function. The learning execution unit 146 may optimize the neural network so that the cosine similarity becomes 1.

The learning execution unit 146 is not limited to this, and may use any known learning method that enables a neural network to output an estimated direction that is closest to the vector of the direction of antenna 210 relative to array antenna 302, which is indicated by the positional relationship included in the training data, when channel response data included in the training data is input. The learning execution unit 146 may also input channel response data included in the training data to a known neural network and perform learning to update the neural network so that the output indicates the positional relationship included in the training data.

The NN output unit 148 outputs the neural network after learning by the learning execution unit 146. The NN output unit 148, for example, transmits the neural network after learning by the learning execution unit 146 to the outside.

The information processing device 100 may use the neural network that has been trained by the learning execution unit 146 to perform control to adjust communication between the mobile body 200 and the communication device 300 that are currently communicating using the antenna 210 and the array antenna 302.

The channel response acquisition unit 150 acquires channel response data indicating the channel response sampled in space and frequency between the antenna 210 (sometimes referred to as the third antenna) equipped on the mobile body 200 (sometimes referred to as the second mobile body) during communication and the array antenna 302 (sometimes referred to as the second array antenna) including multiple antennas 310 (sometimes referred to as the fourth antenna) used by the communication device 300.

The direction acquisition unit 152 inputs the channel response data acquired by the channel response acquisition unit 150 into the neural network that has been trained by the learning execution unit 146, and acquires the direction of the third antenna relative to the second array antenna output from the neural network.

The communication adjustment unit 154 adjusts the wireless communication between the third antenna and the second array antenna based on the direction acquired by the direction acquisition unit 152. The communication adjustment unit 154 may send an adjustment instruction to the communication device 300 according to the direction acquired by the direction acquisition unit 152. The communication adjustment unit 154 may send an adjustment instruction to the second mobile body according to the direction acquired by the direction acquisition unit 152.

If the information processing device 100 does not perform such adjustment processing, the information processing device 100 does not need to include the channel response acquisition unit 150, direction acquisition unit 152, and communication adjustment unit 154.

Figure 10 shows a schematic diagram of an example of the configuration of a HAPS 400, which is an example of a mobile object 200. The HAPS 400 provides wireless communication services to user terminals 70 within a communication area 404 formed by emitting a beam 402 toward the ground. The HAPS 400 includes wings 420, a photovoltaic power generation unit 430, a propeller 440, an elevator 450, a central unit 460, and a pod 470.

The photovoltaic power generation unit 430 includes a photovoltaic panel that receives light and generates electricity. The photovoltaic panel may be a so-called solar power generation panel. The HAPS 400 is equipped with multiple batteries (not shown). The multiple batteries are distributed and arranged in all or some of the wing sections 420, central section 460, and pod 470. The multiple batteries are charged with electricity generated by the photovoltaic power generation unit 430.

A flight control unit 462 and a communications control unit 464 are located within the central section 460. The flight control unit 462 controls the flight of the HAPS 400. The communications control unit 464 controls communications for the HAPS 400.

The flight control device 462 controls the flight of the HAPS 400, for example, by controlling the rotation of the propeller 440. The flight control device 462 also controls the flight of the HAPS 400, for example, by changing the angle of the elevator 450. The flight control device 462 may be equipped with various sensors, such as a positioning sensor such as a GPS sensor, a gyro sensor, an acceleration sensor, and a wind speed sensor, and may manage the position, attitude, direction of movement, speed of movement of the HAPS 400, and the wind speed around the HAPS 400.

The communication control device 464 uses a service link (SL) antenna to form a communication area 404 on the ground. The communication control device 464 may use the SL antenna to form a service link with a terrestrial user terminal 70.

The communication control device 464 may use an FL (Feeder Link) antenna to form a feeder link with the terrestrial gateway 40. The communication control device 464 may access the network 30 via the gateway 40. The FL antenna may be an example of an antenna 210. The gateway 40 may be an example of a communication device 300.

The user terminal 70 may be any type of communications terminal capable of communicating with the HAPS 400. For example, the user terminal 70 may be a mobile phone such as a smartphone. The user terminal 70 may also be a tablet terminal or a PC (Personal Computer). The user terminal 70 may also be a so-called IoT (Internet of Things) device. The user terminal 70 may include anything that falls under the so-called IoE (Internet of Everything) category.

HAPS 400 relays communications between network 30 and user terminal 70, for example, via a feeder link and a service link. HAPS 400 may provide wireless communication services to user terminal 70 by relaying communications between user terminal 70 and network 30.

Network 30 includes a mobile communication network. The mobile communication network may conform to any of the following communication methods: LTE (Long Term Evolution), 5G (5th Generation), 3G (3rd Generation), and 6G (6th Generation) or later. Network 30 may also include the Internet.

The HAPS 400 transmits data received from a user terminal 70 within the communication area 404 to the network 30. Furthermore, when the HAPS 400 receives data addressed to a user terminal 70 within the communication area 404 via the network 30, it transmits the data to the user terminal 70.

The management device 500 manages multiple HAPS 400. The management device 500 may communicate with the HAPS 400 via the network 30 and the gateway 40. The management device 500 controls the HAPS 400 by sending instructions. The management device 500 may cause the HAPS 400 to circle above a target area on the ground so that the communication area 404 covers the target area. For example, while flying in a circular orbit above the target area, the HAPS 400 maintains a feeder link with the gateway 40 by adjusting the direction of the FL antenna, and maintains coverage of the target area by the communication area 404 by adjusting the direction of the SL antenna.

FIG. 11 shows a schematic diagram of an example of the hardware configuration of a computer 1200 that functions as the information processing device 100. A program installed on the computer 1200 can cause the computer 1200 to function as one or more "parts" of an apparatus according to this embodiment, or can cause the computer 1200 to perform operations associated with the apparatus according to this embodiment or one or more "parts," and/or can cause the computer 1200 to perform a process or steps of the process according to this embodiment. Such a program may be executed by the CPU 1212 to cause the computer 1200 to perform specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

The computer 1200 according to this embodiment includes a CPU 1212, RAM 1214, and a graphics controller 1216, which are interconnected by a host controller 1210. The computer 1200 also includes input/output units such as a communications interface 1222, a storage device 1224, a DVD drive, and an IC card drive, which are connected to the host controller 1210 via an input/output controller 1220. The DVD drive may be a DVD-ROM drive, a DVD-RAM drive, or the like. The storage device 1224 may be a hard disk drive, a solid-state drive, or the like. The computer 1200 also includes a ROM 1230 and legacy input/output units such as a keyboard, which are connected to the input/output controller 1220 via an input/output chip 1240.

The CPU 1212 operates according to programs stored in the ROM 1230 and RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires image data generated by the CPU 1212 into a frame buffer provided in the RAM 1214 or into the graphics controller itself, and causes the image data to be displayed on the display device 1218.

The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD drive reads programs or data from a DVD-ROM or the like and provides them to the storage device 1224. The IC card drive reads programs and data from an IC card and/or writes programs and data to an IC card.

ROM 1230 stores therein a boot program and the like that is executed by computer 1200 upon activation, and/or programs that depend on the hardware of computer 1200. I/O chip 1240 may also connect various I/O units to I/O controller 1220 via USB ports, parallel ports, serial ports, keyboard ports, mouse ports, etc.

The programs are provided on a computer-readable storage medium such as a DVD-ROM or IC card. The programs are read from the computer-readable storage medium, installed in storage device 1224, RAM 1214, or ROM 1230, which are also examples of computer-readable storage media, and executed by CPU 1212. The information processing described in these programs is read by computer 1200, resulting in cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be configured by implementing the operation or processing of information in accordance with the use of computer 1200.

For example, when communication is performed between computer 1200 and an external device, CPU 1212 may execute a communication program loaded into RAM 1214 and instruct communication interface 1222 to perform communication processing based on the processing described in the communication program. Under the control of CPU 1212, communication interface 1222 reads transmission data stored in a transmission buffer area provided in RAM 1214, storage device 1224, DVD-ROM, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes received data received from the network to a reception buffer area or the like provided on the recording medium.

The CPU 1212 may also cause all or a necessary portion of a file or database stored on an external recording medium such as the storage device 1224, a DVD drive (DVD-ROM), an IC card, etc. to be read into the RAM 1214, and perform various types of processing on the data on the RAM 1214. The CPU 1212 may then write back the processed data to the external recording medium.

Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. CPU 1212 may perform various types of processing on data read from RAM 1214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search/replacement, etc., as described throughout this disclosure and specified by the program's instruction sequence, and write the results back to RAM 1214. CPU 1212 may also search for information in files, databases, etc. on the recording medium. For example, if multiple entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored on the recording medium, CPU 1212 may search for an entry whose attribute value of the first attribute matches a specified condition from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

The programs or software modules described above may be stored on computer-readable storage media on or near computer 1200. Recording media such as a hard disk or RAM provided within a server system connected to a dedicated communications network or the Internet can also be used as computer-readable storage media, thereby providing the programs to computer 1200 via the network.

The blocks in the flowcharts and block diagrams in this embodiment may represent stages of a process in which an operation is performed or "parts" of a device responsible for performing the operation. Particular stages and "parts" may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable storage medium, and/or a processor provided with computer-readable instructions stored on a computer-readable storage medium. Dedicated circuitry may include digital and/or analog hardware circuits, and may include integrated circuits (ICs) and/or discrete circuits. Programmable circuitry may include reconfigurable hardware circuits, such as field programmable gate arrays (FPGAs) and programmable logic arrays (PLAs), including logical ANDs, ORs, exclusive ORs, NOT ANDs, NOT ORs, and other logical operations, flip-flops, registers, and memory elements.

A computer-readable storage medium may include any tangible device capable of storing instructions that are executed by an appropriate device. As a result, a computer-readable storage medium having instructions stored thereon comprises an article of manufacture, including instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable storage media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable storage media may include floppy disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disc (DVD), Blu-ray disc, memory stick, integrated circuit card, etc.

Computer-readable instructions may include either assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages such as the "C" programming language or similar programming languages.

The computer-readable instructions may be provided to a general-purpose computer, a special-purpose computer, or a processor of another programmable data processing device or programmable circuit, either locally or via a local area network (LAN) or a wide area network (WAN) such as the Internet, so that the processor of the programmable data processing device, such as a computer, or the programmable circuit executes the computer-readable instructions to generate means for performing the operations specified in the flowchart or block diagram. Here, the computer may be a PC (personal computer), tablet computer, smartphone, workstation, server computer, general-purpose computer, special-purpose computer, etc., or may be a computer system in which multiple computers are connected. Such a computer system in which multiple computers are connected is also called a distributed computing system, and is a computer in the broad sense. In a distributed computing system, multiple computers collectively execute a program by each executing a portion of the program and passing data between computers as needed during program execution.

Examples of processors include computer processors, central processing units, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc. A computer may have one processor or multiple processors. In a multiprocessor system with multiple processors, each processor executes a portion of a program and passes data between processors as needed during program execution, allowing the multiple processors to collectively execute a program. For example, in multitasking, each of the multiple processors may execute a portion of each task in small chunks by switching tasks for each time slice. In this case, which portion of a program each processor executes changes dynamically. Which portion of a program each of the multiple processors executes may also be statically determined by programming that takes multiprocessors into consideration.

The present invention has been described above using embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be clear to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that forms incorporating such modifications or improvements can also be included within the technical scope of the present invention.

The order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before" or "prior to," and it should be noted that processes can be performed in any order, unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using terms such as "first," "next," etc. for convenience, this does not mean that the processes must be performed in that order.

10 System, 30 Network, 40 Gateway, 70 User Terminal, 100 Information Processing Device, 110 Input Data, 112 Formula, 114 Formula, 116 Formula, 118 Formula, 120 DOA Estimation NN, 130 Output Data, 140 Memory Unit, 142 Measurement Data Receiving Unit, 144 Teacher Data Acquisition Unit, 146 Learning Execution Unit, 148 NN Output Unit, 150 Channel Response Acquisition Unit, 152 Direction Acquisition Unit, 154 Communication Adjustment Unit, 200 Mobile Object, 210 Antenna, 300 Communication Device, 302 Array Antenna, 310 Antenna, 400 HAPS, 402, beam, 404, communication area, 420, wing section, 430, photovoltaic power generation section, 440, propeller, 450, elevator, 460, center section, 462, flight control unit, 464, communication control unit, 470, pod, 500, management unit, 1200, computer, 1210, host controller, 1212, CPU, 1214, RAM, 1216, graphics controller, 1218, display device, 1220, input/output controller, 1222, communication interface, 1224, storage device, 1230, ROM, 1240, input/output chip

Claims (11)

a training data acquisition unit that acquires training data including a positional relationship between a first antenna of a mobile object and an array antenna including a plurality of second antennas, and channel response data that indicates a channel response sampled in space and frequency between the first antenna and the array antenna when the first antenna and the array antenna are in the positional relationship;
a learning execution unit that uses the plurality of teacher data acquired by the teacher data acquisition unit to execute learning of a neural network that receives the channel response data as an input and outputs an estimated direction of the first antenna relative to the array antenna. The information processing device of claim 1, wherein the channel response data includes channel vectors for each of a plurality of frequencies. The information processing device of claim 1 or 2, wherein the learning execution unit inputs the channel response data included in the training data to the neural network, and updates the neural network using, as a loss function, the cosine similarity between the estimated direction vector output from the neural network and the vector of the direction of the first antenna relative to the array antenna indicated by the positional relationship included in the training data. The first antenna and the array antenna are used for LoS-MIMO (Line of Sight Multiple Input Multiple Output) transmission,
The information processing device according to claim 1 , wherein the plurality of second antennas are arranged at intervals that satisfy requirements for LoS-MIMO transmission. the moving object is a High-Altitude Platform Station (HAPS),
The array antenna is disposed on the ground.
The information processing device according to claim 4 . A program for causing a computer to function as the information processing device described in any one of claims 1 to 5. a training data acquisition step of acquiring training data including a positional relationship between a first antenna and an array antenna including a plurality of second antennas possessed by a mobile object, and channel response data indicating a channel response sampled in space and frequency between the first antenna and the array antenna when the first antenna and the array antenna are in the positional relationship;
a learning execution step of executing learning of a neural network using the plurality of teacher data acquired in the teacher data acquisition step, with the channel response data as input and an estimated direction of the first antenna relative to the array antenna as output. a storage unit configured to store a neural network that receives as input a positional relationship between a first antenna provided to a first moving object and a first array antenna including a plurality of second antennas, the neural network being generated using a plurality of pieces of training data including: channel response data indicating a channel response sampled in space and frequency between the first antenna and the first array antenna when the first antenna and the first array antenna are in the positional relationship; and
a channel response acquisition unit that acquires channel response data sampled in space and frequency between a third antenna provided in the second mobile object and a second array antenna including a plurality of fourth antennas;
a direction acquisition unit that inputs the channel response data acquired by the channel response acquisition unit into the neural network and acquires a direction of the third antenna relative to the second array antenna, the direction being output from the neural network;
a communication adjustment unit that adjusts wireless communication between the third antenna and the second array antenna based on the direction acquired by the direction acquisition unit. The third antenna and the second array antenna are used for LoS-MIMO (Line of Sight Multiple Input Multiple Output) transmission,
The information processing device according to claim 8 , wherein the plurality of fourth antennas are arranged at intervals that satisfy requirements for LoS-MIMO transmission. A program for causing a computer to function as the information processing device described in claim 8 or 9. an NN acquisition step of acquiring a neural network that receives as input a channel response generated using a plurality of training data including a positional relationship between a first antenna provided in a first mobile object and a first array antenna including a plurality of second antennas, and a channel response sampled in space and frequency between the first antenna and the first array antenna when the first antenna and the first array antenna are in the positional relationship, and that outputs a direction of the first antenna relative to the first array antenna;
a channel response acquisition step of acquiring a channel response sampled in space and frequency between a third antenna provided on the second mobile station and a second array antenna including a plurality of fourth antennas;
a direction acquisition step of inputting the channel response acquired in the channel response acquisition step into the neural network and acquiring a direction of the third antenna relative to the second array antenna output from the neural network;
a communication adjustment step of adjusting wireless communication between the third antenna and the second array antenna based on the direction acquired in the direction acquisition step.