CN1868157B - Methods for forward error correction coding above a radio link control layer and related apparatus - Google Patents

Abstract

Transmission techniques are provided that improve service continuity and reduce interruptions in delivery of content that can be caused by transitions that occur when the User Equipment (UE) moves from one cell to the other, or when the delivery of content changes from a Point-to-Point (PTP) connection to a Point-to-Multipoint (PTM) connection in the same serving cell, and vice-versa. Such transmission techniques enable seamless delivery of content across cell borders and/or between different transmission schemes such as Point-to-Multipoint (PTM) and Point-to-Point (PTP). Mechanisms for adjusting different streams and for recovering content from each data block during such transitions are also provided so that data is not lost during a transition. In addition, mechanisms for realigning data during decoding at a receiving terminal are also provided.

Description

Forward error correction coding method and related device on radio link control layer

Priority Claim Pursuant to 35 U.S.C. §119

This application requests Provisional Application No. 60/497,457, filed August 21, 2003, entitled "Method and Apparatus for Seamless Delivery of Broadcast and Multicast Content Across Cell Borders and/or Between Different Transmission Schemes" and filed August 21, 2003 Priority to Provisional Application No. 60/497,456, entitled "L2 Design for Outer Coding Scheme," both of which are assignable to their assignees so that they are expressly incorporated herein by reference.

technical field

The present invention relates generally to communication systems, and more particularly to the delivery of broadcast and multicast content.

Background technique

Wireless communication systems have traditionally been used to carry voice traffic and low data rate non-voice traffic. Now, wireless communication systems that also transmit high data rate (HDR) multimedia services such as video, data, and other types of services are being implemented. Multimedia Broadcast and Multicast Service (MBMS) channels can be used to deliver streaming applications based on voice, audio and video data sources, such as radio broadcasts, television broadcasts, movies and other types of audio or video content. Streaming data sources can tolerate delay and a certain amount of loss or bit errors because these streaming data sources are sometimes intermittent and typically compressed. In this regard, the rate of transmitted data to the radio access network (RAN) can be very variable. Since application layer buffers are typically limited, MBMS transport mechanisms that support variable source data rates are required.

Typically, base stations provide such multimedia traffic services to subscriber stations by transmitting information signals, which can typically be organized into packets. A packet may be a group of bytes arranged in a particular format including data (payload) and control elements. Control elements may include, for example, a preamble and quality metrics, which may include cyclic redundancy checks (CRC), parity bits, and other types of metrics. Packets are usually formatted into messages according to the communication channel structure. Messages propagate between an originating terminal and a destination terminal and can be affected by communication channel characteristics, such as signal-to-noise ratio, fading, time variation, and other such characteristics. These characteristics can affect the modulated signal differently in different communication channels. One of them is that the transmission of modulated information signals over wireless communication channels requires the selection of appropriate methods to protect the information in the modulated signals. These methods may include, for example, encoding, symbol repetition, interleaving, and other methods known to those of ordinary skill in the art. However, these methods add overhead. Therefore, a design trade-off must be made between the reliability of message delivery and the amount of overhead.

Typically, the operator selects a point-to-point (PTP) connection or a point-to-multipoint (PTM) connection on a cell-by-cell basis depending on the number of subscriber stations or user equipment (UE) interested in receiving MBMS content.

Point-to-point (PTP) transmission uses dedicated channels to deliver services to selected users in the coverage area. A "dedicated" channel carries information to and from a single subscriber station. In point-to-point (PTP) transmissions, separate channels are available for transmissions to each mobile station. Dedicated user traffic for one user service, in the forward link or downlink direction, may be carried over eg a logical channel called a Dedicated Traffic Channel (DTCH). For example, point-to-point (PTP) communication services are typically most efficient if there are not enough users requiring a particular Multimedia Broadcast and Multicast Service (MBMS) in the coverage area. In such cases where the base station only transmits the service to specific users who have requested the service, point-to-point (PTP) transmission may be used. For example, in a WCDMA system, it may be more efficient to use dedicated channels or point-to-point (PTP) transmissions until there are more than a predetermined number of mobile stations.

"Broadcast communication" or "point-to-multipoint (PTM) communication" is communication to multiple mobile stations over a common communication channel. A "common" channel carries information to and from multiple subscriber stations and can be used by several terminals simultaneously. In point-to-multipoint (PTM) communication services, a cellular base station may broadcast a multimedia service on a common channel if, for example, the number of users requiring the service exceeds a predetermined threshold number within the coverage area of the base station. In CDMA2000 systems, broadcast or point-to-multipoint (PTM) transmission is typically used instead of PTP transmission because PTM radio bearers are almost as efficient as PTP radio bearers. Common channel transmissions from a particular base station may not necessarily be synchronized with common channel communications from other base stations. In a typical broadcast system, one or more central stations provide content to (subscribers' broadcast network). The central station can transmit information to all subscriber stations or to a specific group of subscriber stations. Each subscriber station interested in the broadcast service monitors a common forward link signal. Point-to-multipoint (PTM) transmissions may be sent on the downlink or forward common channel. Typically, common broadcast forward link signals are broadcast on a unidirectional channel, such as a common traffic channel (CTCH) that exists in the forward link or "downlink" direction. Since this channel is unidirectional, subscriber stations typically do not communicate with the base station since allowing all subscriber units to communicate back to the base station could overload the communication system. Thus, in the context of point-to-multipoint (PTM) communication services, when there is a bit error in the information received by the subscriber station, the subscriber station may not be able to communicate back to the base station. Therefore, other information protection measures may be desirable.

In CDMA 2000 systems, subscriber stations can soft combine in point-to-multipoint (PTM) transmissions. Even when measures are taken to protect the information signal, the condition of the communication channel may deteriorate so that the destination station cannot decode some packets transmitted over the dedicated channel. In these cases, one approach may be to simply retransmit packets that cannot be decoded by using Automatic Repeat Request (ARQ) made by the destination (user) station to the origin (base) station. Retransmissions help ensure delivery of data packets. If the data cannot be transmitted correctly, the user of the RLC at the transmitting end can be notified.

Typically, a user station undergoes transitions in a number of situations. These transitions can be categorized in different ways. For example, conversion can be divided into "cross conversion" and "direct conversion". Conversion can also be divided into "inter-cell" conversion and "intra-cell" conversion.

Switching between cells or transmission schemes results in interruption of service which may not be desired by the user. When a subscriber station or user equipment (UE) moves from one cell to another, or when the delivery of Multimedia Broadcast and Multicast Service (MBMS) content changes from one mode to another within a serving cell, there may be problem appear. Transmissions from neighboring cells may be time-shifted relative to each other by an amount Δt1. Also, an additional delay may be introduced during the handover, since the mobile station needs to determine the system information in the target cell, which requires a certain amount of processing time Δt2. Data streams transmitted from different cells (or different transport channel types Point-to-Point (PTP)/Point-to-Multipoint (PTM)) may be offset relative to each other. Consequently, during point-to-multipoint (PTM) transmissions from different cells, the mobile station may receive the same chunk twice, or some chunks may be lost. This is less than satisfactory in terms of service quality. Switching between cells and/or between point-to-point (PTP) transmissions and point-to-multipoint (PTM) transmissions, depending on the duration of the switching and delays or skews between transmissions, can cause interruptions in service.

Therefore, there is a need in the art for transmission techniques that provide service continuity and reduce interruptions in content delivery that may be caused by transitions that occur when a user equipment (UE) moves from one cell to The other, or when the content transfer in the same cell changes from a point-to-point (PTP) connection to a point-to-multipoint (PTM) connection, and when a switch in the opposite direction occurs. These transmission techniques will preferably enable seamless transfer of content across cell boundaries and/or between different transmission schemes such as point-to-multipoint (PTM) and point-to-point (PTP). Mechanisms for aligning different data streams and restoring content from each data block during these transitions are also desirable so that data is not lost during transitions. It would also be desirable to provide a mechanism for rearranging data during decoding at the receiving terminal.

Description of drawings

Figure 1 is a diagram of a communication system;

Figure 2 is a block diagram of the UMTS signaling protocol stack;

Fig. 3 is a block diagram of the packet switching user plane of the UMTS protocol stack;

Figure 4 is a block diagram of the access layer portion of the UMTS signaling protocol stack;

5A is a block diagram of the data transmission mode used in the radio link control (RLC) layer of the UMTS signaling protocol stack and the various channels used in each layer;

5B is a block diagram showing the structure of a radio link control (RLC) layer including various RLC data transmission modes;

Figure 5C is a block diagram showing entities for implementing Radio Link Control (RLC) Acknowledged Mode (AM);

Fig. 6 is a block diagram of an improved UMTS protocol stack with a forward error correction layer;

FIG. 7A shows an embodiment of a protocol structure of an access layer including a forward error correction (FEC) layer;

FIG. 7B shows another embodiment of the protocol structure of the access layer including the forward error correction (FEC) layer;

Fig. 8 is a block diagram of an information block and an external code block corresponding to the information block;

FIG. 9A is a block diagram showing an external code block structure applicable to Multimedia Broadcast and Multicast Service (MBMS) data;

Figure 9B is a block diagram showing the structure of the outer code block of Figure 9A, wherein multiple lines are transmitted per transmission time interval (TTI);

Figure 9C is a block diagram showing the structure of the outer code block of Figure 9A, wherein each row is sent in multiple TTIs;

10A and 10B are block diagrams showing outer code blocks generated by a forward error correction layer;

Fig. 11 is the embodiment of the Forward Error Correction (FEC) layer used in RLC UM+ entity;

Figure 12A shows an encoding process for generating an outer code block from a data unit, where the line size of the outer code block is fixed;

Figure 12B shows an example of transmitting information over the air in Figure 12A;

Figure 13 shows the encoding process for generating outer code blocks with variable line sizes;

Figure 14 is a diagram of an embodiment of a forward error correction (FEC) header format;

Figure 15 is an algorithm for enabling a mobile station to delay decoding by a time offset between different logical streams;

Figure 16 is a diagram showing the difference between outer code blocks received by a mobile station when it transitions between a point-to-multipoint (PTM) transmission received from cell A and another point-to-multipoint (PTM) transmission from cell B The graph of the time relationship;

17 is a diagram showing the time relationship between external code blocks received by a mobile station when switching between point-to-multipoint (PTM) transmission and point-to-point (PTP) transmission occurs;

Fig. 18 is a diagram showing the transition or relocation between a point-to-point (PTP) transmission from radio network controller (RNC) A and another point-to-point (PTP) transmission from radio network controller (RNC) B, by a mobile station A graph of the temporal relationship between received external code blocks.

Detailed ways

The word "exemplary" is used herein to mean "serving as an example, instance or illustration". Any embodiment described herein as "exemplary" is not necessarily to be construed as a preferred embodiment or as an advantage over other embodiments.

The term "mobile station" is used herein interchangeably with the terms "destination station", "subscriber station", "subscriber unit", "terminal" and "user equipment (UE)" and is used herein to refer to hardware, Such as a base station, which communicates with an access network, such as the UMTS Terrestrial Radio Access Network (UTRAN). In a UMTS system, a user equipment (UE) is a device that allows a user to access UMTS network services, and preferably also includes a USIM containing all user subscription information. A mobile station may be mobile or stationary and may generally include any communication device, data device or terminal that communicates through a wireless channel or through a wired channel, for example using fiber optic or coaxial cables. A mobile station may be implemented in a device including a PC card, a compact flash memory, an external or internal modem, or a wireless or wireline phone, but the included components are not limited to these components.

The term "connection establishment state" refers to a state where a mobile station is in the process of establishing an active traffic channel connection with a base station.

The term "traffic state" refers to a state in which a mobile station has an established active traffic channel connection with a base station.

The term "communication channel" is used herein to refer to either a physical channel or a logical channel, depending on the context.

The term "physical channel" is used herein to refer to a channel that conveys user data or control information over the air interface. The physical channel is the "transmission medium" that provides the wireless platform over which information is actually transmitted and is used to convey signaling and user data over the wireless link. Typically, a physical channel includes a combination of a scrambling code and a channelization code. In the uplink direction, the relative phase may also be included. There are many different physical channels that can be used on the uplink based on what the mobile station is trying to do. In UMTS systems, the term physical channel can also refer to different kinds of bandwidth allocated over the Uu interface for different purposes. The physical channel constitutes the physical presence of the Uu interface between the user equipment (UE) domain and the network access domain. A physical channel may be defined by the physical mapping and attributes used to communicate data over the air interface.

The term "transport channel" is used herein to refer to a communication route used for data transmission between peer physical layer entities. A transport channel refers to the manner in which information is transmitted. In general, there can be two types of transport channels known as common transport channels and dedicated transport channels. The transport channel can be defined by how or what characteristic data can be transmitted through the air interface on the physical layer, such as whether to use a dedicated or common physical channel, or multiplexing of logical channels. The transport channel can be used as a service access point (SAP) for the physical layer. In the UMTS system, transport channels describe how logical channels can be transported and map these information streams to physical channels. Transport channels can be used to transfer signaling and user data between the medium access control (MAC) layer and the physical layer (L1). The Radio Network Controller (RNC) looks at the transport channel. Information is passed from the MAC layer to the physical layer over any of a number of transport channels that can be mapped to physical channels.

The term "logical channel" is used herein to refer to a stream of information dedicated to a particular type of information or air interface transfer. Logical channels relate to information being conveyed. The logical channel can be defined by the type of information to be transmitted, for example, it can be defined by signaling or user data, and it can be understood as different tasks that the network and the terminal should perform at different time points. Logical channels can be mapped into transport channels that perform the actual transfer of information between the mobile station domain and the access domain. Information is transferred via logical channels that may be mapped through transport channels that can be mapped to physical channels.

The term "dedicated channel" is used herein to refer to a channel typically dedicated to or reserved for a particular user and carrying information to or from a particular mobile station, subscriber unit or user equipment A channel for device information. Typically, dedicated channels carry information intended for use by a given user, including data for the actual service as well as higher layer control information. A dedicated channel can be identified by a certain code on a certain frequency. Dedicated channels can be bi-directional to potentially facilitate feedback.

The term "common channel" is used herein to refer to a transmission channel that conveys information to or from multiple mobile stations. In a common channel, information can be shared among all mobile stations. Common channels can be divided among all users or among a group of users in a cell.

The term "point-to-point (PTP) communication" is used herein to refer to a communication delivered to a single mobile station over a dedicated physical communication channel.

The terms "broadcast communication" or "point-to-multipoint (PTM) communication" may be used herein to refer to communication to multiple mobile stations over a common communication channel.

The term "reverse link or uplink channel" is used herein to refer to the communication channel/link through which a mobile station sends signals to a base station in a radio access network. This channel can also be used to transmit signals from a mobile station to a mobile base station, or from a mobile base station to a base station.

The term "forward link or downlink channel" is used herein to refer to the communication channel/link through which the radio access network sends signals to the mobile station.

The term "transmission time interval (TTI)" is used herein to refer to how often data arrives at the physical layer from higher layers. Transmission Time Interval (TTI) may refer to the time between arrivals of Transport Block Sets (TBS), and is approximately equal to the period of TBS transmitted by the physical layer on the radio interface. Data transmitted on the transport channel during a TTI may be coded and interleaved together. A TTI can last for multiple radio frames and can be a multiple of the minimum interleaving period. The starting positions of TTIs for different transport channels that may be multiplexed together for a single connection are time aligned. TTI has a common starting point. Media Access Control Transmits a transport block set to the physical layer every TTI. Different transport channels mapped on the same physical channel may have different transmission time interval (TTI) durations. Multiple PDUs can be transmitted in one TTI.

The term "packet" is used herein to refer to a group of bits arranged in a particular format including data or payload and control elements. Control elements may include, for example, preambles, quality metrics, and other control elements known to those skilled in the art. Quality metrics include, for example, cyclic redundancy check (CRC), parity bits, and other quality metrics known to those skilled in the art.

The term "access network" is used herein to refer to the equipment necessary for accessing a network. An access network may include a collection or network of base stations (BS) and one or more base station controllers (BSC). The access network transports data packets between a plurality of subscriber stations. The access network may further connect to additional networks outside the access network, such as an intranet or the Internet, and may communicate data packets between the access terminals and such external networks. In the UMTS system, the access network may be referred to as the UMTS Terrestrial Radio Access Network (UTRAN).

The term "core network" is used herein to refer to the network used to connect to the Public Switched Telephone Network (PSTN) for circuit-switched calls in the circuit-switched (CS) domain, or for packet-switched calls in the packet-switched (PS) domain. Switching and routing capabilities connected to a Packet Data Network (PSDN). The term "core network" also refers to routing capabilities for mobility and subscriber location management and for authentication services. The core network includes the network elements required for switching and user control.

The term "base station" is used herein to refer to an "origin station," which includes the hardware with which a mobile station communicates. In UMTS systems, the term "Node B" is used interchangeably with the term "Base Station". Base stations can be fixed or mobile.

The term "cell" is used herein to refer to either a hardware or a geographic coverage area depending on the context in which the term is used.

The term "service data unit (SDU)" is used herein to refer to a data unit exchanged with a protocol overlying the protocol of interest.

The term "payload data unit (PDU)" is used herein to refer to a data unit exchanged with a protocol underlying the protocol of interest. If the identity of the protocol of interest is ambiguous, it will be explicitly mentioned in the name. For example, FEC-PDU is a PDU of the FEC layer.

The term "soft handoff" is used herein to refer to communications between a subscriber station and two or more sectors, where each sector belongs to a different cell. Reverse link communications can be received by two sectors, and forward link communications can be transmitted on the forward links of two or more sectors simultaneously.

The term "softer handoff" is used herein to refer to communications between a subscriber station and two or more sectors, where each sector belongs to the same cell. Reverse link communications may be received by both sectors, and forward link communications may be simultaneously transmitted on one of the forward links of two or more sectors.

The term "erasing" is used herein to refer to unrecognized messages, and may also be used to refer to sets of bits that can be lost when decoding.

The term "crossover" may be defined as a transition from point-to-point (PTP) transmission to point-to-multipoint (PTM) transmission, or vice versa. The four possible crossovers are: transition from point-to-point (PTP) transmission in cell A to point-to-multipoint (PTM) transmission in cell B, transition from point-to-multipoint (PTM) transmission in cell A to cell B Conversion of point-to-point (PTP) transmission in cell A, conversion from point-to-point (PTP) transmission in cell A to point-to-multipoint (PTM) transmission in cell A, conversion from point-to-multipoint (PTM) transmission in cell A to Switching of point-to-point (PTP) transmissions in cell A.

The term "direct switching" may be defined as switching from one point-to-point transmission to another point-to-point transmission and from point-to-multipoint transmission to point-to-multipoint transmission. Two possible direct conversions are the conversion from point-to-point (PTP) in cell A to point-to-point (PTP) transmission in cell B and from point-to-multipoint (PTM) in cell A to point-to-multipoint in cell B Conversion of point (PTM) transfers.

The term "inter-cell handover" is used to refer to handovers across cell boundaries. The four possible inter-cell transitions are: transition from point-to-point (PTP) transmission in cell A to point-to-point (PTP) transmission in cell B, transition from point-to-multipoint (PTM) transmission in cell A to Conversion of point-to-multipoint (PTM) transmission, conversion from point-to-point (PTP) transmission in cell A to point-to-multipoint (PTM) transmission in cell B, and conversion from point-to-multipoint (PTM) transmission in cell A Conversion of point-to-point (PTP) transmissions into cell B. Typically, the most frequent transitions are from point-to-multipoint (PTM) transmissions to point-to-multipoint (PTM) transmissions across cell boundaries.

The term "intra-cell switching" is used to refer to switching from one mode to another within a cell. Two possible intra-cell transitions are: transition from point-to-point (PTP) transmission in cell A to point-to-multipoint (PTM) transmission in cell A, and transition from point-to-multipoint (PTM) transmission in cell A to Switching of point-to-point (PTP) transmissions in cell A.

The term "radio bearer" is used to refer to the services provided by Layer 2 for user data transfer between the User Equipment (UE) and the UMTS Terrestrial Radio Access Network (UTRAN).

Embodiments of the invention will now be discussed in which the aspects discussed above are implemented in a WCDMA or UMTS communication system. Figures 1-5C illustrate some aspects of a conventional UMTS or WCDMA system, wherein aspects of the invention described herein as applicable in this description are provided for purposes of illustration and limitation only. It should be understood that aspects of the present invention are also applicable in other systems that carry both voice and data, such as GSM systems and CDMA 2000 systems that conform to standards including 3G TS 25.211, 3G TS 25.212, 3G TS 25.213 and 3G TS 25.214 (W-CDMAB Standard) in a set of documents "Third Generation Partnership Project" (3GPP), or "TR-45.5 Physical Layer Standard for cdma2000 Spread Spectrum Systems" (IS- 2000 standard) and GSM specifications such as TS 04.08 (Mobile Radio Interface Layer 3 Specification), TS 05.08 (Radio Subsystem Link Control) and TS 05.01 (Physical Layer over Radio Path (General Description)).

For example, although the description indicates that the radio access network 20 may be implemented using a Universal Terrestrial Radio Access Network (UTRAN) air interface, alternatively, in a GSM/GPRS system, the access network 20 may be a GSM/EDGE radio The Access Network (GERAN), or in the intersystem case it may comprise cells of the UTRAN air interface and cells of the GSM/EDGE air interface.

UMTS network layout

Fig. 1 is a block diagram of a communication system according to a UMTS network layout. The UMTS system includes a user equipment (UE) 10 , an access network 20 and a core network 30 . The UE 10 is connected to an access network, the access network is connected to a core network 30, and the core network 30 may be connected to an external network.

The UE 10 includes a mobile device 12 and a Universal Subscriber Identity Module (USIM) 14 that contains user subscription information. The (not shown) Cu interface is the electrical interface between the USIM 14 and the mobile device 12. UE 10 is typically a device that allows a user to access UMTS network services. UE 10 may be a mobile device such as a cellular telephone, a fixed station, or other data terminal. A mobile device may be eg a wireless terminal for wireless communication over an air interface (Uu). The Uu interface is the interface through which the UE accesses the fixed part of the system. A USIM is typically an application program installed on a "smart card" or other logical card that includes a microprocessor. The smart card saves the user identification code, executes the authentication algorithm, and stores the authentication in the encryption key and user information required by the terminal.

Access network 20 includes wireless devices for accessing the network. In a WCDMA system, the access network 20 is the Universal Terrestrial Radio Access Network (UTRAN) air interface. The UTRAN includes at least one Radio Network Subsystem (RNS) comprising at least one Base Station or “Node B” 22 connected to at least one Radio Network Controller (RNC) 24 .

RNC controls the radio resources of UTRAN. The RNC 24 of the access network 20 communicates with the core network 30 through the Iu interface. The Uu interface, the Iu interface 25, the Iub interface and the Iur interface allow Internet interconnection between devices from different manufacturers, and are specified in 3GPP standards. The implementation of the radio network controller (RNC) varies with different manufacturers, so it will be described in general terms below.

A Radio Network Controller (RNC) 24 acts as a switching and control element of the UMTS Terrestrial Radio Access Network (UTRAN) and is located between the Iub interface and the Iu interface 25 . The RNC acts as a service access point for all services provided by the UTRAN to the core network 30, eg for managing connections to user equipments. The Iub interface 23 connects Node B 22 and Radio Network Controller (RNC) 24. The Iu interface connects the UTRAN to the core network. The Radio Network Controller (RNC) provides the switching point between the Iu bearer and the base station. A user equipment (UE) 10 may have several radio bearers between itself and a radio network controller (RNC) 24 . The radio bearer refers to the user equipment (UE) context, which is a set of definitions required by Iub in order to arrange public and private connections between the user equipment (UE) and the radio network controller (RNC) . The respective RNCs 24 can communicate with each other via an optional Iur interface allowing soft handover between cells connected to different nodes 22. Thus, the Iur interface allows inter-RNC connectivity. In these cases, when the drift RNC transmits to the mobile station 10, via one or more base stations 22, frames that can be exchanged over the Iur interface, the serving RNC maintains the Iu connection 25 to the core network 30 and executes the selector and Outer loop power control function.

The RNC controlling one Node B 22 may be referred to as the controlling RNC of the Node B, which controls the load and congestion of its own cells and also performs admission control code allocation for new radio links to be established in those cells.

The RNC and the base station (or Node B) can connect and communicate via the Iub interface 23 . The RNC controls the use of radio resources by each base station 22 connected to a particular RNC 24. Each base station 22 controls one or more cells and provides a radio link to mobile stations 10 . The base station can perform interface processing such as channel coding and interleaving, rate adaptation and spreading. The base station also performs basic radio resource management operations, such as inner loop power control. The base station 22 switches the data flow between the Iub and Uu interfaces 23,26. Base station 22 also participates in radio resource management. An air interface Uu 26 connects each base station 22 to a mobile station 10 . A base station may be responsible for radio transmission to mobile station 10 in one or more cells and may be responsible for radio reception from mobile station 10 in one or more cells.

The core network 30 includes all of the following switching and routing capabilities: (1) connection to the PSTN 42 if the current call is circuit switched, or to a packet data network (PDN) if the current call is packet switched, (2) Mobility and user location management, and (3) authentication services. The core network 30 may include a Home Location Register (HLR) 32, a Mobile Switching Service Center/Visitor Location Register (MSC/VLR) 34, a Gateway Mobile Switching Center (GMSC) 36, a Serving General Packet Radio Service Support Node (SGSN) 38, and Gateway GPRS Support Node (GGSN) 40 .

The core network 30 may be connected to an external circuit switched (CS) network 42 that provides a circuit switched connection, such as the public switched telephone network (PSTN) or (ISDN) in the present case of a packet switched call, or may be connected to To a PS network 44, such as the Internet providing a connection for packet data services in the present case of a packet switched call.

UMTS signaling protocol stack

FIG. 2 is a block diagram of the UMTS signaling protocol stack 110 . The UMTS signaling protocol stack 110 includes an access stratum and a non-access stratum (NAS).

Typically, the access layer includes a physical layer 120 , a layer 2130 and a radio resource control (RRC) layer 160 , wherein the layer 2130 includes a medium access control (MAC) layer 140 and a radio link control (RLC) layer 150 . Hereinafter, each layer of the access layer will be described in more detail.

The UMTS non-access stratum is essentially the same as the GSM upper stratum and can be divided into a circuit switched part 170 and a packet switched part 180 . The circuit switched portion 170 includes a connection management (CM) layer 172 and a mobility management (MM) layer 178 . The CM layer 172 handles circuit switched calls and includes various sublayers. Call Control (CC) sublayer 174 performs functions such as setup and release. Supplementary Services (SS) sublayer 176 performs functions such as call forwarding and three-way calling. Short message service (SMS) sublayer 177 performs short message service. The MM layer 178 handles location updates and authentication for circuit switched calls. The packet switching section 180 includes a session management (SM) sublayer 182 and a GPRS mobility management (GMM) sublayer 184 . A session management (SM) sublayer 182 handles packet-switched calls by performing functions such as setup and release, and may also include a short message service (SMS) portion 183 . The GMM sublayer 184 handles location updates and authentication for packet switched calls.

Figure 3 is a block diagram of the packet switched user plane of the UMTS protocol stack. The stack includes the Access Stratum (AS) and the Non-Access Stratum (NAS). The NAS layer includes an application layer 80 and a packet data protocol (PDP) layer 90 . Application layer 80 is provided between user equipment (UE) 10 and remote user 42 . A PDP layer 90, such as IP or PPP, is provided between the GGSN 40 and the user equipment (UE) 10. A Low Layer Packet Protocol (LLPP) 39 is provided between the remote user 42 and the SGSN 38. The Iu interface protocol 25 is provided between the Radio Network Controller (RNC) 24 and the SGSN 38, and the Iub interface protocol is provided between the Radio Network Controller (RNC) 24 and the Node B 22. Other parts of the AS layer will be described below.

Access Layer (AS)

Figure 4 is a block diagram of the access layer portion of the UMTS signaling protocol stack. The traditional access layer includes physical layer (L1) 120, data link layer (L2) 130, radio link control (RLC) layer 150, packet data convergence protocol (PDCP) layer 156, broadcast/multicast control (BMC) layer 158 and a radio resource control (RRC) layer 160 , wherein the data link layer (L2) 130 has sublayers including a medium access control (MAC) layer 140 . These layers are further described below.

The radio bearer transports user data 163 between the application layer and Layer 2 (L2) 130 . Control plane signaling 161 can be used for all UMTS-specific control signaling, and includes application protocols in signaling bearers used to transmit application protocol messages. The application protocol can be used to establish a bearer for the UE 10. The user plane transports all user plane information 163 sent and received by the user, such as encoded speech in a voice call or packets in an Internet connection. User plane information 163 conveys data flows and data bearers for those data flows. Each data stream may be characterized by one or more frame protocols specified for that interface.

The Radio Resource Control (RRC) layer 160 acts as an overall controller of the access stratum and configures all other layers in the access stratum. RRC layer 160 generates control plane signaling 161, control plane signaling 161 controls radio link control unit 152, physical layer (L1) 120, medium access control (MAC) layer 140, radio link control (RRC) layer 150, Packet Data Convergence Protocol (PDCP) layer 156 and Broadcast/Multicast Control (BMC) layer 158 . A radio resource control (RRC) layer 160 determines the types of measurements made and reports the results of those measurements. The RRC layer 160 also serves as a control and signaling interface to the non-access stratum.

More specifically, the RRC layer 160 broadcasts a system information message including access stratum and non-access stratum information elements to all user equipments (UEs) 10 . The RRC layer 160 establishes, maintains and releases a radio resource control (RRC) connection between the UTRAN 20 and the UE 10. UE RRC requests connection, while UTRAN RRC establishes and releases connection. The RRC layer 160 also establishes, reconfigures and releases radio bearers between the UTRAN 20 and the UE 10, and initiates these operations through the UTRAN 20.

The RRC layer 160 also handles aspects of user equipment (UE) 10 mobility. These procedures depend on the UE state, whether the call is circuit-switched or packet-switched, and the radio access technology (RAT) of the new cell. The RRC layer 160 also pages the UE 10. UTRAN RRC pages the UE regardless of whether the UE is listening to the paging channel or the paging indicator channel. The RRC of the UE notifies the upper layer of the core network (CN) 30 .

Data Link Layer (L2) 130 includes Media Access Control (MAC) sublayer 40, Radio Link Control (RLC) sublayer 150, Packet Data Convergence Protocol (PDCP) sublayer 156, and Broadcast/Multicast Control (BMC) Sublayer 158.

The broadcast and multicast control protocol (BMC) 158 transmits messages from the cell broadcast center via the wireless interface by conforming to the broadcast/multicast service from the broadcast domain on the wireless interface. The BMC protocol 158 provides a service called "radio bearer" and exists in the user plane. The BMC protocol 158 and the RNC store cell broadcast messages received over the CBC-RNC interface for scheduled transmissions. On the UTRAN side, the BMC 158 calculates the required transmission rate for the cell broadcast service based on messages receivable via the (not shown) CBC-RNC interface, and requests appropriate CTCH/FACH resources from the RRC. The BMC protocol 158 also receives scheduling information along with each cell broadcast message through the CBC-RNC interface. Based on this scheduling information, on the UTRAN side, the BMC generates a scheduling message, thereby generating a sequence of scheduling BMC messages. On the user equipment side, the BMC evaluates the scheduling messages and indicates the scheduling parameters to the RRC, which can then use these scheduling parameters to configure the lower layers for discontinuous reception. The BMC also transmits BMC messages, such as schedule and cell broadcast messages according to the schedule. Non-corrupted cell broadcast messages may be delivered to upper layers. Part of the control signaling between the UE 10 and the UTRAN 20 is a Radio Resource Control (RRC) 160 message that can convey all parameters required for establishing, modifying and releasing the Layer 2 protocol 130 and Layer 1 protocol 120 entities. RRC messages convey all higher layer signaling in their payload. Radio Resource Control (RRC) controls the mobility of user equipment in connected mode through signaling such as measurements, handover and cell update.

Packet Data Convergence Protocol (PDCP) 156 exists in the user plane for services from the PS domain. Services provided by PDCP may be referred to as radio bearers. Packet Data Convergence Protocol (PDCP) provides header compression services. Packet Data Convergence Protocol (PDCP) 156 includes compression methods that provide better spectral efficiency for services that transport IP packets over the air. Any of several header compression algorithms may be used. PDCP compresses redundant protocol information at the transmitting entity and decompresses it at the receiving entity. Header compression methods may be specific to a particular network layer, transport layer, or combination of upper layer protocols such as TCP/IP and RTP/UDP/IP. PDCP also transports user data it receives from the Non-Access Stratum in the form of PDCP Service Data Units (SDUs) and forwards these data to the RLC entity and vice versa. PDCP also provides support for lossless SRNS relocation. When PDCP uses Acknowledged Mode (AM) RLC for in-sequence delivery, PDCP entities that can be configured to support lossless RSRNS relocation have protocol data unit (PDU) sequence numbers that, during relocation, can be The confirmed PDCP packets are forwarded to the new SRNC together.

The RLC layer 150 provides services to higher layers (eg, non-access stratum) through a service access point (SAP) that can be used by higher layer protocols in the UE side and IURNAP protocol in the UTRAN side. Service Access Points (SAPS) describe how the RLC layer handles data packets. All higher layer signaling, such as mobility management, call control, session management, etc., can be encapsulated in RLC messages for transmission over the radio interface. The RLC layer 150 includes various radio link control entities 152 connected to the MAC layer 140 through logical channels that convey signaling information and user data.

On the control plane 161, RLC services may be used by the RLC layer for signaling. On the user plane 163, RLC services may be used by service-specific protocol layers PDCP or BMC, or by other higher layer user plane functions. For services that do not use PDCP 156 or user plane protocols, RLC services may be referred to as signaling radio bearers in the control plane 161 and as radio bearers in the user plane 163. In other words, if the service cannot use the PDCP and BMC protocols, the RLC layer 150 provides a service called a Signaling Radio Bearer (SRB) in the control plane 161 and a service called a Radio Bearer (RB) in the user plane 163. Serve. Otherwise, RB service may be provided by PDCP layer 156 or BMC layer 158 .

A radio link control (RLC) layer 150 performs framing functions for user and control data, including segmentation/splicing and padding functionality. Typically, the RLC layer 150 provides segmentation and retransmission services to the Radio Resource Control (RRC) 160 layer for control data in the control plane 161 and to the application layer for user data in the user plane 163 . Typically, the RLC layer segments variable-length higher-level protocol data units (PDUs) into smaller RLC protocol data units (PDUs), and reassembles variable-length higher-level protocol data units (PDUs) from smaller RLC protocol data units (PDUs) data unit (PDU). Typically, one radio link control (RLC) protocol data unit (PDU) conveys one PDU. For example, the size of a Radio Link Control (RLC) PDU may be set according to the minimum possible bit rate for a service using Radio Link Control (RLC). As will be discussed below, for variable rate services, when using any bit rate above the lowest bit rate, several Radio Link Control (RLC) PDUs may be transmitted during one transmission time interval (TTI). The RLC transport entity also performs splicing. If the content of the Radio Link Control (RLC) Service Data Unit (SDU) does not fill an integer number of Radio Link Control (RLC) PDUs, the first segment of the next Radio Link Control (RLC) SDU can be placed in the radio In a link control (RLC) PDU, concatenated with the last segment of the previous RLC SDU. Typically, the RLC transmitting entity also performs padding functions. When the remaining data to be transmitted does not fill an entire Radio Link Control (RLC) PDU of a given size, the remainder of that data field may be filled with stuffing bits. For example, techniques for reducing or eliminating the amount of fill used may be provided in accordance with aspects of the invention discussed below with reference to FIGS. 11-13.

The RLC receiving entity detects repetitions of received Radio Link Control (RLC) PDUs and ensures that the results in higher layer PDUs are delivered once to upper layers. The RLC layer also controls the rate at which the PRLC transmitting entity can send information to the RLC receiving entity.

Figure 5A is a block diagram illustrating the data transmission scheme used in the Radio Link Control (RLC) layer of the UMTS signaling protocol stack, showing possible mappings of logical, transport and physical UMTS channels with respect to the access layer. Those skilled in the art will understand that for a given user equipment (UE), all mappings will not necessarily be defined at the same time, and multiple instances of some mappings may occur simultaneously. For example, a voice call may use three dedicated traffic channel (DTCH) logical channels mapped to three dedicated channel (DCH) transport channels. Also, some channels shown in FIG. 5, such as CPICH, SCH, DPCCH, AICH, and PICH, exist in the physical layer background and do not transmit upper layer signaling or user data. The content of these channels may be defined at the physical layer 120 (L1).

Each RLC instance in the radio link control (RLC) layer can be configured by the radio resource control (RRC) layer 160 to operate in one of the following three modes: transparent mode (TM), unacknowledged mode (UM) or Acknowledgment Mode (AM). They will be described in detail below with reference to FIG. 5B . The three data transmission modes indicate the mode in which the radio link control (RLC) is configured for the logical channel. Transparent and unacknowledged mode RLC entities are defined as unidirectional, while acknowledged mode entities are bidirectional. Normally, for all RLC modes, CRC error detection is performed on the physical layer, and the result of the CRC check, along with the actual data, is transmitted to the RLC. Depending on the specific requirements of each mode, these modes perform some or all of the functions of the RLC layer 150, including segmentation, reassembly, splicing, stuffing, retransmission control, flow control, duplicate detection, in-sequence delivery, error correction, and encryption. These functions will be described in more detail below with reference to Figures 5B and 5C. According to aspects of the invention discussed herein, a new radio link control (RLC) data transmission mode may be provided.

The MAC layer 140 provides services to the RLC layer 150 through a logical channel characterized by the type of data transferred. The medium access control (MAC) layer 140 maps and multiplexes logical channels into transport channels. The MAC layer 140 identifies user equipment (UE) on a common channel. The MAC layer 140 also multiplexes higher layer PDUs into transport blocks delivered to the physical layer on the common transport channel, and demultiplexes transport blocks delivered from the physical layer on the common transport channel into higher layer PDUs. The MAC handles multiplexing of services for common transport channels since this cannot be done in the physical layer. When the common transport channel transports data from a dedicated type of logical channel, the medium access control (MAC) header includes the identification of the UE. The MAC layer also multiplexes higher layer PDUs into transport block sets delivered to the physical layer on dedicated transport channels, or demultiplexes transport block sets delivered from the physical layer on dedicated transport channels into higher layer PDUs.

The MAC layer 140 receives the RLC PDU along with status information about the amount of data in the RLC transmit buffer. The MAC layer 140 compares the amount of data corresponding to the transport channel with a threshold set by the RRC layer 160 . If the amount of data is too high or too low, the MAC sends a measurement report on the status of the traffic to the RRC. The RRC layer 160 may also request the MAC layer 160 to periodically send these measurements. The RRC layer 160 uses these reports to trigger reconfiguration of radio bearers and/or transport channels.

The MAC layer also selects an appropriate transport format (TF) for each transport channel depending on the instantaneous source rate of the logical channel. The MAC layer 140 provides priority handling of data flows by selecting "high bit rate" and "low bit rate" transport formats (TFs) for different data flows. Packet Switched (PS) data is inherently bursty, so the amount of data available for transmission varies from frame to frame. When more data is available, the MAC layer 140 can choose one of the higher data rates, however when both signaling and user data are available, the MAC layer 140 chooses between them so that the higher priority channel The maximum amount of data sent. The transport format (TF) can be selected relative to the transport format combination (TFC) that can be defined by admission control for each connection.

The Media Access Control (MAC) layer also performs encryption. Each radio bearer may be encrypted separately. The details of encryption are described in 3GPP TS 33.102.

In a system such as WCDMA, there are three types of transport channels that can be used to transmit packet data. These channels are commonly referred to as common transport channels, dedicated transport channels and shared transport channels. In the downlink, the transport channel packet data is selected by a packet scheduling algorithm. In the uplink, the transport channel is selected by the mobile device 10 based on parameters set by the packet scheduling algorithm.

Common channels may be, for example, the Random Access Channel RACH in the uplink and the Forward Access Channel FACH in the downlink. They both carry signaling data and user data. Common channels have low settling times. Since the common channel is available for signaling before the connection is established, the common channel is immediately available for transmitting packets without a long setup time. Typically, there are several RACH or FACH per sector. The common channel does not have a feedback channel, so typically open loop power control or fixed power is used. Also, common channels cannot use soft handoff. Therefore, the link level performance of the common channel will be worse than that of the dedicated channel and will cause more interference than the dedicated channel. Therefore, common channels may be better suited for transmitting small individual packets. Applications used in public channels will be applications such as short message service and short text mail. Sending a single request to a web page also fits well with the concept of a common channel, but in the case of larger data volumes, the common channel suffers from poor wireless performance.

Dedicated channels can use fast power control and soft handoff features that improve wireless performance, and typically cause less interference than common channels. However, setting up a dedicated channel takes more time than accessing a common channel. Dedicated channels can have variable bit rates from a few kilobytes per second up to 2 megabytes per second. Since the bit rate varies during transmission, the downlink orthogonal codes must be assigned according to the highest bit rate. Therefore, variable bit rate dedicated channels consume valuable downlink orthogonal code space.

The physical layer (L1) 120 is connected to the MAC layer 140 through a transport channel that transfers signaling information and user data. The physical layer 120 provides services to the MAC layer through a transport channel that can be characterized by how and what characteristic data is transmitted.

The physical layer (L1) 120 receives signaling and user data on the radio link through physical channels. Typically, the physical layer (L1) performs multiplexing and channel coding including CRC calculations, forward error correction (FEC), rate matching, interleaving transport channel data, and multiplexing transport channel data, and other physical layer processes such as obtaining , access, paging and wireless link establishment/failure. The physical layer (L1) may also be responsible for spreading and scrambling, modulation, measurements, transmit diversity, power weighting, handover, compressed mode and power control.

FIG. 5B is a block diagram showing the structure of a Radio Link Control (RLC) layer. As mentioned above, each RLC entity or instance 152 in the Radio Link Control (RLC) layer 150 may apply to the Radio Resource Control (RRC) layer 160 to be configured to operate in one of the following three data transmission modes: : Transparent Mode (TM), Unacknowledged Mode (UM) or Acknowledged Mode (AM). Quality of Service (OoS) settings control the data transfer mode used for user data.

A TM is unidirectional and includes a transmitting TM entity 152A and a receiving TM entity 152B. In transparent mode, no protocol order is added to the higher layer data. Bad protocol data units (PDUs) may be discarded or marked as bad. Where higher layer data is typically not fragmented, streaming may be used, although in certain cases, delivery with limited fragmentation/reassembly capabilities may be achieved. When fragmentation/reassembly is used, it can be negotiated during radio bearer establishment.

A UM is also unidirectional and includes a transmitting UM entity 152C and a receiving UM entity 152D. UM RLC entities are defined as unidirectional, since no link between uplink and downlink is required. Data delivery is not guaranteed in UM. UM can be used, for example, in certain RRC signaling procedures where acknowledgments and retransmissions are not part of the RRC procedure. Examples of user services using unacknowledged mode RLC are cell broadcast services and Voice over IP. Depending on configuration, received erroneous data can be marked or discarded. Timer-based discarding can be applied without an explicit signaling function, so that RLC PDUs that cannot be transmitted within a specified time can simply be removed from the transmit buffer. In the unacknowledged data transfer mode, the PDU construction includes a sequence number, and a sequence number check can be performed. Sequence Number Checking Helps ensure the integrity of reassembled PDUs by verifying the sequence numbers in Radio Link Control (RLC) PDUs as they are reassembled into Radio Link Control (RLC) SDUs , and provide a means of detecting corrupt Radio Link Control (RLC) SDUs. Any damaged Radio Link Control (RLC) SDUs may be discarded. Segmentation and splicing are also available in Unacknowledged Mode (UM).

In Acknowledged Mode, the RLC AM entity is bidirectional and can piggyback link status indications in user data in the opposite direction. Figure 5C is a block diagram showing entities for implementing a Radio Link Control (RLC) Acknowledged Mode (AM) entity and how an AM PDU can be constructed. Data packets (RLC SDUs) received from higher layers via the AM-SAP may be segmented and/or concatenated 514 into fixed-length protocol data units (PDUs). The length of the protocol data unit is a semi-static value determined during the establishment of the radio bearer, and can be changed through the RRC radio bearer reconfiguration process. Bits carrying information about the length and extension may be inserted into the beginning of the last PDU for concatenation or padding purposes, or include data from the SDU. If several SDUs fit into one PDU, they can be concatenated. An appropriate Length Indicator (LI) may be inserted in the header of the PDU. The PDU may then be placed in transmit buffer 520, which also maintains retransmission management.

The PDU can be constructed in the following manner: get a PDU from the transmit buffer 520, add a header to it, if the data in the PDU does not fill the entire RLC PDU, then you can add a padding field or add a piggybacked status message. A piggybacked status message may be sent from the receiver or from the sender to indicate the discard of the RLC SDU. The header contains the RLC PDU sequence number (SN), a query bit (P) that can be used to request status from a peer entity, and an optional length indicator (LI), if SDU splicing, stuffing, or piggybacking has occurred in the RLC PDU PDU, the length indicator can be used.

Typically, Acknowledged Mode (AM) is used for packet type services such as Internet browsing and mail downloading. In acknowledged mode, an automatic repeat request (ARQ) mechanism can be used for error correction. Any packets received with errors may be retransmitted. The quality versus delay performance of RLC can be controlled by RRC by configuring the number of retransmissions provided by RLC. If the RLC cannot transmit data correctly, for example, if the maximum number of retransmissions has been reached, or the retransmission time has elapsed, upper layers are notified and the Radio Link Control (RLC) SDU may be discarded. The peer entity can also be notified of the discarding of the SDU by sending a Move Receive Window command in a status message, so that the receiver also removes all PDUs belonging to the discarded Radio Link Control (RLC) SDU.

RLC can be configured for both in-order and out-of-sequence transfers. With in-order delivery, the order of the higher layer PDUs is maintained, while out-of-order delivery sends the higher layer PDUs once they have been completely received. The RLC layer provides in-sequence delivery of higher layer PDUs. This function preserves the order of higher layer PDUs submitted for RLC transmission. If this feature is not used, out-of-order transfers are provided. In addition to data PDU transfers, status and reset control procedures may be signaled between peer RLC entities. The control process can even use separate logical channels, so that one AM RLC entity can use one or two logical channels.

Encryption can be performed in the RLC layer for both acknowledged and unacknowledged RLC modes. In FIG. 5C , the AM RLC PDU is encrypted 540 except for the first two digits including the PDU sequence number and the query bit. The PDU sequence number is an input parameter to the encryption algorithm and must be readable by the peer entity in order to perform encryption. The 3GPP specification TS33.102 describes encryption.

The PDU may then be forwarded to the MAC layer 140 via a logical channel. In Figure 5C, additional logical channels (DCCH/DTCH) are shown with dashed lines, illustrating that an RLC entity can be configured to use different logical channels for sending control PDUs and data PDUs. The receiver 530 of the AM entity receives the RLC AMPDU from the MAC layer through one of the logical channels. The physical layer CRC, which can be calculated on the entire RLC PDU, can be used to check the bit error. The actual CRC check can be performed in the physical layer and the RLC entity receives the result of the CRC check together with the entire header deciphered data and possible piggybacked status information can be extracted from the RLC PDU. If the received PDU is a robust message, or if status information is piggybacked in the AM PDU, control information (status messages) can be passed to the sender, which checks its retransmission buffer against the received status information . The PDU number from the RLC header is used for decryption 550, and the PDU sequence number is also used when storing the encrypted PDU in the receive buffer. Once all PDUs belonging to a complete SDU are in the receive buffer, the SDU can be reassembled. Although not shown, checks may be performed for in-sequence delivery and duplicate detection before the RLC SDU is delivered to higher layers.

The RLC entity 152 is re-initialized when a user equipment (UE) or mobile station moves (or changes cells) between PTM transmissions and point-to-point (PTP) transmissions. This would undesirably result in the loss of any data located in the radio link control (RLC) buffers. As mentioned above, when a mobile station moves from one cell to another, or when the transmission of Multimedia Broadcast and Multicast Service (MBMS) content within a serving cell changes from point-to-point (PTP) transmission mode to point-to-multipoint ( Problems may arise when using PTM) transfer mode.

It is desirable to be able to maintain Multimedia Broadcast and Multicast Service (MBMS ) continuity, and expects to avoid the submission of duplicate information. In order to maintain continuity of MBMS services and avoid duplicate message submissions, layer 2150 should be able to rearrange data from both streams. This synchronization cannot be provided by the physical layer, since the network endpoints may be different in each mode. If forward error correction (FEC) is performed below the RLC layer 150, as is the case in 3GPP2, during any transition between point-to-multipoint (PTM) transmission and point-to-point (PTP) transmission and in the reverse direction During this time, data may be lost. Furthermore, this would require physical layer synchronization and sharing of the same Medium Access Control (MAC) among multiple cells (eg, with common scheduling). Thus, this can cause problems in 3GPP2 where these assumptions do not apply.

Point-to-point (PTP) transmission

Assuming the application is significantly delay tolerant, the most efficient data transfer mode for point-to-point (PTP) transfers is Radio Link Control (RLC) Acknowledged Mode (AM). For example, RLC Acknowledged Mode (AM) is typically used for Packet Switched Data Transfer (PTP) on dedicated logical channels. RLC operates in Acknowledged Mode (AM) on dedicated logical channels. As shown in Figure 5A, dedicated user traffic for one user service in the downlink direction may be sent over a logical channel called a dedicated traffic channel (DTCH).

In Acknowledged Mode (AM), the reverse link can be used to retransmit a request if the data is in error. RLC transmits Service Data Units (SDUs) and guarantees delivery to its peers by retransmissions. If the RLC cannot transmit the data correctly, the user of the RLC at the transmitting end is notified. Operating in RLC AM is generally more power efficient, at the cost of introducing additional latency.

Point-to-multipoint (PTM) transmission

A Common Traffic Channel (CTCH) is a unidirectional channel existing in the downlink direction and may be used when transmitting information to all terminals or a specific group of terminals. These data transfer modes all use a unidirectional common channel that does not establish a reverse link channel.

It is desirable to provide a structure that allows MBMS services to be switched transparently between point-to-point (PTP) and point-to-multipoint (PTM) transmission modes. In order to achieve good performance when switching between point-to-point (PTP) and point-to-multipoint (PTM) transmission modes, it is also desirable to provide an architecture that allows switching between different radio link control (RLC) modes. This can help, for example, to reduce power requirements.

Aspects of the invention will now be described on the basis of the embodiment shown and described with reference to FIGS. 6 to 19 . These features, in particular, help maintain service continuity during these transitions through the use of a new forward error correction (FEC) layer.

6 is a diagram of a modified UMTS protocol stack with a forward error correction (FEC) layer capable of operating in forward error correction (FECd) mode and forward error correction (FECc) mode. The forward error correction (FEC) layer allows the underlying radio link control (RLC) entity 152 to switch from one radio link to another when the user equipment (UE) changes from point-to-point (PTP) to point-to-multipoint The control (RLC) data transmission mode becomes another radio link control (RLC) data transmission mode while maintaining service continuity. According to this embodiment, the FEC layer can operate in the first mode (FECc) or the second mode (FECd). In one implementation, the first mode (FECc) can use parity blocks and the second mode (FECd) can operate without using parity blocks. Changing between FECd and FECc modes will have a much lower impact than changing between RLC modes and can be seamless so that no data is lost during the transition.

Forward Error Correction (FECc) mode protects user data using outer coding techniques. This can be especially effective on public channels. Forward Error Correction (FECc) mode allows, on the Radio Link Control (RLC) layer, functions typically present in Unacknowledged Mode (UM), such as framing (segmentation and concatenation) and sequence number addition. As a result, the Radio Link Control (RLC) layer can use Transparent Mode (TM) for Point-to-Multipoint (PTM) transmissions, since conventional Unacknowledged Mode (UM) functions can be performed at the Forward Error Correction (FEC) layer. Although this function can be duplicated in Radio Link Control (RLC) Acknowledged Mode (AM), the gain due to ARQ compensates for this duplication.

By placing a forward error correction (FEC) layer or an outer coding layer above the radio link control (RLC) layer, the sequence number can be added in a layer independent of the radio link control (RLC). Using additional overhead such as sequence numbers for unacknowledged transmissions, protocol data units (PDUs) with encoder packets (EPs) can be rearranged during asynchronous transmission of MBMS data. Because the sequence number is added to the layer above the radio link control (RLC), the sequence number is common in point-to-point (PTP) transmission and point-to-multipoint (PTM) transmission, so when transferring from point-to-multipoint (PTM) to Sequence number continuity is maintained when transitions in point-to-point (PTP) transmissions occur. This allows data to be rearranged so that duplication of data and/or loss of data can be avoided.

Outer coding can also be used in point-to-point (PTP) transmissions, which can potentially gain some power for the system and/or reduce the delay of retransmissions. Multimedia Broadcast and Multicast Service (MBMS) data can be somewhat delay tolerant. In point-to-point (PTP) transmission, a feedback path is provided. The use of Radio Link Control (RLC) Acknowledgment Mode (AM) is made more efficient due to the use of ARQ retransmissions, which are generally more radio efficient than FEC schemes that always send additional parity blocks, when necessary. Thus, adding parity blocks to MBMS payload data is unnecessary on dedicated logical channels such as Point-to-Point (PTP).

7A and 7B show an embodiment of a protocol structure for an access layer including a forward error correction (FEC) layer 157 placed above a radio link control (RLC) layer 150 . One embodiment of a forward error correction (FEC) layer will be described with reference to FIG. 11 .

A forward error correction (FEC) layer 157 receives user plane information 163 directly over a user plane radio bearer. Since the forward error correction (FEC) layer is on top of the radio link control (RLC) layer, the FEC protocol data unit (PDU) corresponds to the RLC service data unit (SDU). The FEC layer preferably supports: arbitrary (limited to multiples of 8 bits) SDU size, variable rate source, out-of-order reception of packets from lower layers, and reception of duplicate packets from lower layers. The size of the FEC PDU can be limited to multiples of 8 bits.

As described in more detail below with reference to FIG. 9A, the FEC layer 157 segments and concatenates higher layer user data blocks, such as SDUs, into equally sized rows. Each row may also be referred to as an inner block. Each protocol data unit (PDU) may include overhead. The overhead may include a length indicator (LI) that indicates where the last protocol data unit (PDU) starts so that data from a particular user data block, such as a service data unit (SDU), can be located. A collection of PDUs comprises an encoder packet (EP) or "encoder matrix". The number of PDUs included in the Encoder Packet (EP) depends inter alia on the outer code used. Packing each encoder "matrix" row into a separate or separate Transmission Time Interval (TTI) enhances physical layer performance. To reduce the buffering burden, shorter transmission time interval (TTI) durations may be used.

Then, an encoder packet (EP) may be passed through the outer code encoder to generate parity lines. As will be described in more detail below with reference to FIG. 9A , the FEC layer 157 can perform outer encoding by providing the functionality of a Reed Solomon (RS) encoder in the UMTS Terrestrial Radio Access Network (UTRAN) 20, and can perform outer encoding by The function of ReedSolomon decoder is provided in (UE) 10 to perform external decoding.

The parity lines generated by the outer encoder can be added to the encoder packet (EP) and can be placed in the transmit buffer as a set of inner blocks. Each inner block has information added to it to generate a Protocol Data Unit (PDU). The set of PDUs can then be transmitted.

The FEC layer 157 also allows recovery of data belonging to a single EP even if different inner blocks are received from different cells. This is achieved by conveying a sequence number (SN) in the header of each protocol data unit (PDU). In one embodiment, a System Frame Number (SFN) can help maintain the alignment of data relative to an encoder packet (EP). Sequence numbers are discussed in more detail throughout this document, for example, with reference to FIGS. 10A and 10B .

The FEC layer 157 also performs padding and reassembly; delivery of user data; and performs in-sequence delivery of upper layer PDUs, duplicate detection, and sequence number checking.

In the embodiment shown in FIGS. 6-7A, a forward error correction (FEC) layer 157 is shown between the packet data convergence protocol (PDCP) layer 156 and the radio link control (RLC) layer 150 (eg, with ( BMC) layer at the same layer and below the Packet Data Convergence Protocol (PDCP) layer). By placing the forward error correction (FEC) layer 157 just above the radio link control (RLC) layer 150, the performance of the outer code can be optimized because the inner block size is comparable to the "golden" packet size for packets sent over the air. match. It should be understood, however, that a forward error correction (FEC) layer is shown here for purposes of illustration only and not limitation. Packet Data Convergence Protocol (PDCP) layer 156 may be used on top of Forward Error Correction (FEC) layer 157 to use its header compression capabilities. It should be noted that the current Packet Data Convergence Protocol (PDCP) layer 156 is defined for point-to-point (PTP) transmission using dedicated logical channels. As shown in Figure 7B, a forward error correction (FEC) layer may be provided anywhere within the access layer above the radio link control (RLC) layer or in the application layer. A Forward Error Correction (FEC) layer may be below or above a Packet Data Convergence Protocol (PDCP) layer. If FEC is performed at the application layer 80, it can be applied equally to GSM and WCDMA, even though the "golden" packet size will be different for the two.

external code design

A new Forward Error Correction (FEC) layer can perform outer coding on user plane information. FIG. 8 is a diagram showing an information block 91 and an external code block 95 to explain the concept of the external code block structure. FIG. 9A is a diagram showing an example of how an external code block structure can be applied to Multimedia Broadcast and Multicast Service (MBMS) data 91. FIG. Outer coding can improve physical layer performance when broadcasting delay tolerant content throughout a cell. External codes can eg help avoid data loss during inter-cell handovers and during handovers between point-to-point (PTP) and point-to-multipoint (PTM) transmission modes.

The outer code block 95 can be represented in the form of a matrix including k protocol data units 91 and N−k parity rows 93 . In outer block encoding, data can be assembled into large encoder packets or information blocks by organizing user data into k payload lines via segmentation, concatenation, and data padding (including insertion of overhead into inner blocks) 91, the resulting information block 91 is then encoded to generate N-k parity rows 93 which can be added to the information block 91 to generate an outer code block 95. The parity block 93 adds redundant information to the information block 91 . A single row in the outer code block may then eventually be transmitted over a single or multiple Transmission Time Intervals (TTIs). Redundant information on sets of Protocol Data Units (PDUs) can allow the original information to be reconstructed even if some PDUs are lost during transmission.

Figure 9A shows a schematic outer code structure known as a Reed-Solomon (RS) block code. Reed-Solomon (RS) codes can be used to detect and correct channel errors. The external code shown in FIG. 9A is a systematic (n, k) block code, where each Reed-Solomon (RS) code symbol includes one byte of information defined by rows and columns. Each column includes a Reed-Solomon (RS) codeword. If n lost blocks are to be recovered, at least n parity blocks are required. Thus, the amount of storage required increases with the number of parity blocks. In Reed-Solomon (RS) coding, Nk parity symbols can be added to k systematic symbols to generate a codeword. In other words, a codeword [N, k] of a Reed-Solomon (RS) code has k information or "systematic" symbols and Nk parity symbols. N is the length of the code and k is the dimensionality of the code. For every k information bytes, the code generates n coded symbols, the first k of which may be identical to the information symbols. Each row may be referred to as an "inner block", which represents the payload per Transmission Time Interval (TTI). In a conventional WCDMA system, transmissions may take place over a basic WCDMA structure such as 20ms frames (TTI). The parity symbols can be obtained from the systematic symbols using the generator matrix G _k×N defined as follows:

m _1×k G _k×k =c _1×N (Equation 1)

m _1×k = information word = [m ₀ m ₁ . . . m _k-1 ] (Equation 2)

c _1×N =codeword=[c ₀ c ₁ . . . c _N-1 ] (Equation 3)

Among them, m _i and _ci belong to any Galois field. For example, if the symbols of a Reed-Solomon (RS) codeword are bits, a 2-dimensional Galois field (GF(2)) will be used to describe the decoding operation. In one embodiment, if the symbols are octets, the 256-dimensional Galois field GF(256) may be used to describe the decoding operation. In this case, each column of information per row consists of 1 byte. Each information column can be encoded using a [N, k] Reed-Solomon (RS) code on the 256-dimensional Galois field GF(256). If there are M bytes per row, the outer block is encoded M times. Therefore, each outer block 95 has N*M bytes.

delete decoding

External code structures allow deletion corrections. If the decoder already knows which symbols are erroneous, relatively little computation is required to reconstruct the erroneous system symbols. An encoder packet (EP) or matrix refers to the entire collection of data at the output of an outer encoder. Redundant information is fetched column-wise from each row, and each row is transmitted with a CRC attached to it, which must be checked to confirm that the data was sent correctly. In the case of MBMS transmission, a CRC can be used in each transport channel block, which CRC indicates whether the inner block 91 is in error, and if the CRC fails, all symbols in the block can be assumed to be erroneous. In one embodiment, if a given internal block is erroneous, all bits for that block may be deleted. The term "delete" refers to every symbol that belongs to the erroneous block that failed the CRC. It can be assumed that no deleted symbols are correct. Neglecting the probability that the CRC does not detect an error, each N×1 column contains both correct and deleted symbols.

The received vector r can be written as:

r _1×N =[c ₀ e e c ₃ c ₄ e c ₆ c ₈ ... c _N-1 ] (equation 4)

Where e indicates deletion.

Erasure decoding allows correction of up to Nk erroneous symbols. The error correction performance of RS codes is generally much better than that of typical RS codes because symbols that are not erased can be assumed to be correct. The size of the CRC used in each inner block should be large enough to ensure that the probability of an undetected bit error does not exceed the residual outer block probability. For example, if a 16-bit CRC is used in the inner block, the lower bound for the residual outer block error rate will be 2 ⁻¹⁶ =1.5·10 ⁻⁵ . If there are no bit errors in the first k inner blocks, no RS decoding needs to be performed because the systematic symbols and information symbols are the same.

It may be noted that once k blocks with a good CRC are received, the decoding of the outer block can be performed without waiting for all N inner blocks to be received. To perform erasure decoding, an improved generator matrix Ω k _×k can be obtained from the generator matrix G _k×N by removing all columns corresponding to deleted or unnecessary blocks, e.g., only the first k good received , to identify the improved generator matrix Ω _k×k . The source information word m can be recovered in the following way:

$m_{1\×k}={\left[{\mathrm{\Ω}}_{k\×k}\right]}^{-1}\·r_{1\×k}^{\′}$ (equation 5)

where r _{1 x k} ' is the improved received vector obtained with the first k good symbols. Therefore, the complexity of erasure decoding can be reduced to the complexity of k*k matrix inversion. Therefore, the use of RS erasure decoding can greatly simplify the computational complexity of RS decoding.

Impact of Data Packing on External Code Performance

As will be discussed below with reference to Figures 11-13, outer coding can be used with variable rate data sources without incurring excessive overhead if the amount of padding and overhead sent over the air is limited by a particular outer coding scheme . In the external code scheme discussed above, data can be packed into blocks of a given size, and shortened ReedSolomon codes can be operated on these blocks. The encoded packet data can be packed into TTIs in at least two different ways which will be described with reference to Figures 9A and 9B.

FIG. 9B is a diagram showing the outer code block structure of FIG. 9A, where multiple lines may be sent per transmission time interval (TTI). According to another aspect of the invention, data from one row is transmitted in a single TTI. In another embodiment, data from one encoder packet (EP) line is put into one TTI so that each TTI contains data from that encoder packet (EP) line. Thus, each row can be transmitted in a separate WCDMA frame or Transmission Time Interval (TTI). Transmitting each row in one TTI will provide better performance. In Figure 9B, both k and n are divided by the number of rows per TTI, and the errors in rows can all be correlated. This makes a noticeable difference when looking at EP BER versus TTI BER.

Figure 9C is a diagram showing the outer block structure of Figure 9A, where each row can be sent in multiple TTIs. It should be appreciated that while FIG. 9C illustrates that each row of an encoder packet (EP) is sent over four TTIs (TTI0-TFI3), each row may actually be sent over any number of TTIs. Since each column is an outer code word, each of the four different transmission "phases" (TTI0-TTI3) corresponds to an independent outer code. In order to recover the entire packet, it is necessary to correctly decode all these independent external codes.

10A and 10B are diagrams showing outer code blocks generated by a forward error correction layer.

FECc mode can be used on common or point-to-multipoint (PTM) logical channels to construct external condition blocks 95 by adding odd and even lines or blocks 93 to MBMS payload data 91 . Each outer block 95 includes a plurality of inner blocks 91 , 93 . Identifying the order of the inner blocks and their position relative to the encoder block allows each available inner block to be placed in the correct position for correct outer decoding. In one embodiment, each internal block includes a header 94 that identifies the internal block by an internal block number m and an external block number n. For example, outer block n includes a data portion 91 having m inner Multimedia Broadcast and Multicast Service (MBMS) payload blocks, and a redundant portion 93 having M-(m+1) inner parity blocks. According to this embodiment, the sequence number space can be optimized for MBMS and can be defined with many different sequence numbers, for example, 0 to 127. The sequence number space should be large enough so that the same sequence number does not appear after a receive gap caused by any kind of conversion. Even if some blocks are missing, the receiving UE should be able to determine the order of the internal blocks. If the UE misses more internal blocks than can be identified by the entire sequence number space, the UE will not be able to reorder the internal blocks correctly. The sequence number of the same internal block is the same for the FECd block and the FECc block. The FECd block does not include the redundant portion 93 used in the FECc block. The FECd entity and the FECc entity may use the same over-the-air bit rate.

sender

Transmit Forward Error Correction (FEC) entity 410 includes service data unit (SDU) buffer 412 for receiving SDUs, segmentation and concatenation unit 414, outer encoder 416 that performs Reed Solomon (RS) encoding, adds sequence numbers to Sequence number generator 418 on encoded PDUs, transmit buffer 420 for transmitting PDUs over logical channel 406 , and scheduling unit 422 .

As indicated by the arrow, a service data unit (SDU) buffer 412 receives user data (FEC SDUs) on the radio bearer 402 in the form of service data units (SDUs) and stores FEC SDUs from higher layers. Receive buffer 412 communicates to scheduling unit 422 how much data is to be transmitted.

As discussed above, typically, the amount of time it takes to fill an encoder packet (EP) will vary since the source data rate typically varies. As explained with reference to FIG. 13, frame filling efficiency can be improved by flexibly deciding when to start packing data. By delaying the generation of EPs as much as possible based on the jitter tolerance of the receiving FEC entity 430, the amount of padding introduced can be reduced.

The scheduling entity 422 can decide when to start encoding. The scheduler 422 determines how long a packet may have to wait before it needs to be sent out, preferably based on the Qos policy for that particular service. Once the scheduler 422 determines that enough data has accumulated, or the maximum acceptable packet transfer delay has been exhausted, the scheduler 422 triggers the generation of an encoder packet (EP) 91 . Segmentation and concatenation unit 414 divides Service Data Units (SDUs) into lines and generates Length Indicators (LIs).

Scheduling unit 422 preferably determines the optimal number of lines for EPs or Protocol Data Units (PDUs) so that the SDU fits into exactly that number of lines (eg, 12 lines). Optionally, the scheduler 422 selects the FEC PDU size that will result in the smallest possible padding from those configured by RRC, and requests the Segmentation & Concatenation function 414 to format the SDU into k blocks of size PDU_size-FEC_Header_size. This formatting can vary. Examples of different types of formatting are discussed below with reference to FIGS. 12-13. The total amount of data to consider should include the overhead that will be added by the stitching and segmentation function 414 . To generate an encoder packet (EP), the scheduler 422 requests the concatenation and segmentation function 414 to generate k PDUs of that size. This size scale includes reorganization information. In one embodiment, a PDU may have a size that is a multiple of 8 bits, and the data of successive PDUs correspond to different symbols in the codeword.

The k PDU blocks may then pass through an outer encoder 416 that performs Reed Solomon (RS) encoding. The outer encoder 416 encodes data in the encoder packet (EP) matrix by generating and appending redundancy or parity information to the encoder packet (EP) matrix to generate outer code blocks. In one embodiment, the outer code may be assumed to be an (n, k) erasure decoded block code, and the outer encoder generates n-k parity blocks. The encoder performs encoding on k lines of information of equal length and delivers them to lower sublayer n Protocol Data Units (PDUs) of the same size. The first k blocks are exactly the same as it received, and the next n-k blocks correspond to parity information.

The scheduler 422 also monitors the time alignment or relative timing of the PTM streams and performs transmissions to adjust the alignment of the different logical streams. For example, during reconfiguration, time alignment between PTP and PTM logical flows may be adjusted to facilitate service continuity. Best performance is achieved when the streams are fully synchronized.

Different base stations (or different transmission modes PTP, point-to-multipoint (PTM)) transmit the same content stream, but these streams may be misaligned. However, if the Encoder Packet (EP) format of the data streams is the same, the information on each stream is exactly the same. Sequencing each outer block allows the user equipment (UE) to combine the two streams, since the user equipment (UE) will know the relationship between the two streams.

Sequence number generator 418 prepends each block with a sequence number in the same order as used in encoder 416 to generate the PDUs. In one embodiment, the sequence number generator prepends each outer code block with, for example, an 8-bit sequence number to generate a PDU. Additional overhead information may also be added to outer code blocks. The sequence number space should be large enough to accommodate the worst-case time-difference between these streams. Therefore, in another embodiment, a sequence number space of 20 may be used, and at least 5 bits may be reserved for sequence numbers in each header. After performing Reed Solomon (RS) encoding, this header can be appended to the outer code block, so this "outer" header is not protected by the outer block. It is also preferable to sequence number the parity blocks, even if they cannot be transmitted. In one embodiment, sequence number phases may be aligned with encoder packet boundaries. A flip of the sequence number will correspond to the receipt of a new encoder packet.

Forward Error Correction (FEC) Header Format

As mentioned above, synchronization of the data streams can be obtained by introducing sequence numbers comprising information related to the ordering of the PDUs. In addition to reordering and duplicate detection, sequence numbers allow data from the respective sources included in the encoder packet to be rearranged. This sequence number unambiguously identifies the order in which each packet should be considered. This sequence number can constitute a "FEC header" that can be appended to both an information payload unit (PDU) and a parity block after encoding is performed. This ordinal should not be protected by external code as it is needed for decoding.

14 is a diagram of an embodiment of a forward error correction (FEC) header format. To facilitate alignment of data with encoder packets (EPs), the sequence number may be split to include a reserved portion (R) 402, an encoder packet (EP) portion (EPSN) identifying an EP, and identifying a particular inner block within an encoder packet Inner Encoder Packet (IEPSN) 406 for the position of .

It is expected that the FEC layer 400 can operate between all radio link control (RLC) modes. Since both Radio Link Control (RLC) AM and Radio Link Control (RLC) UM require Service Data Units (SDUs) to have a size that is a multiple of 8 bits, it is expected that the FEC layer 400 also complies with this requirement. Since the external code for the FEC layer 400 operates in byte-sized data increments, the encoder packet (EP) line size also needs to be an integer number of bytes. Therefore, the FEC header size 401 for the FEC Protocol Data Unit (PDU) size should also be a multiple of 8 bits in order to be acceptable by the Radio Link Control (RLC). In one embodiment, the forward error correction (FEC) header 401 may be one byte, the reserved portion (R) 402 includes a single bit, the portion identifying the EP (EPSN) 404 includes 3 bits, and identifies the PDU at the encoder The IEP part (IEPSN) 406 of the location within the packet consists of 4 bits. In this embodiment, an 8-bit sequence number is used because it is expected that one PDU will be sent per TTI, and because it is not expected that the transmission timing of different cells will drift by more than 100 ms.

The transmit buffer 420 stores PDUs until one frame of data is accumulated. When a PDU is requested, the transmit buffer 420 transmits frames one by one to the MAC layer via a logical channel through the wireless interface (Uu). The MAC layer then delivers the PDU to the physical layer via the transport channel, where the PDU can finally be delivered to the UE 10.

Receiving end

Still referring to FIG. 11 , the receiving forward error correction (FEC) entity 430 includes a receiving buffer/reordering/duplication detection unit 438, a sequence number clearing unit 436, an outer decoder 434 performing Reed Solomon (RS) decoding, and a reassembly unit/ A service data unit (SDU) transmit buffer 432 .

The information rows of the EP matrix correspond to PDUs. To support outer encoding, the receiving forward error correction (FEC) entity 430 accumulates sequence numbers of FEC PDUs before triggering outer decoding. To achieve continuous reception, despite the need to decode the encoder packets, the user equipment (UE) buffers incoming protocol data units (PDUs) while performing the decoding.

The receive buffer 438 may accumulate PDUs until the entire encoder packet (EP) is received, or until a scheduling unit (not shown) deems that there are no more retransmissions for the encoder packet (EP). Once it is decided that no more data will be received for a given encoder packet, the missing PDU can be marked as deleted. In other words, in the decoding process, PDUs that do not pass the CRC check will be replaced by deletions.

Since some blocks may be lost during transmission, and also because different data streams may have different delays, the receiving forward error correction (FEC) entity 430, in the receive buffer/reordering/duplication detection unit 438, performs repetition Detect and possibly perform reordering of received blocks. Sequence numbers may be used in each FEC protocol data unit (PDU) to aid in reordering/duplication detection. Sequence numbers may be used in receive buffer 438 to reorder data received out of order. Once the PDUs have been reordered, the duplicate detection unit detects duplicate PDUs in the Encoder Packet (EP) based on their sequence numbers and removes any duplicates.

These sequence numbers can then be cleared. Sequence number removal unit 436 removes sequence numbers from the Encoder Packet (EP) since sequence numbers cannot be part of a block sent to a Reed Solomon (RS) decoder.

The data can then be passed to an external decoding function 434 to recover the lost information. The outer decoder 434 receives the encoder packet (EP) and, if necessary, the Reed Solomon (RS) decodes the encoder packet (EP) using the parity information to regenerate any erroneous or missing lines. For example, if all k protocol data units (PDUs) containing information were not received correctly, or less than k PDUs out of n PDUs were not received correctly, then for protocol data units (PDUs) up to the size of parity PDUs ), an external decode can then be performed to recover the missing information PDU. Whenever outer decoding is performed, at least one parity PDU will be available at the receiver. If all k protocol data units (PDUs) containing information are received correctly, or less than k PDUs out of n PDUs are received correctly, decoding is not necessary. The message protocol data unit (PDU) may then be passed to reassembly function 432 .

Regardless of whether the outer decoding was successful, the line of information may then be passed to the reassembly unit/function 432 . The reassembly unit 432 uses the length indicator (LI) to reassemble or reconstruct the SDUs from the information rows of the encoder packet (EP) matrix. Once the SDUs are successfully put together, the service data unit (SDU) transmit buffer 432 transmits the service data unit (SDU) over the radio bearer 440 to transmit the SDU to higher layers.

At the receiving Forward Error Correction (FEC) entity 430, enabling the UE to delay decoding by a time offset between different logical streams may allow the system to take advantage of possible gaps in data due to lack of synchronization between logical streams Received out of order. This smoothes out service during handoffs and during transitions between PTP and PTM. An algorithm for enabling a UE to delay decoding by a time offset between different logical streams will be discussed with reference to FIG. 15 .

Encoder grouping (EP) options: fixed or variable row size

The FEC or external code entity has flexibility as to when a Protocol Data Unit (PDU) can be constructed, since the Protocol Data Unit (PDU) does not need to be sent continuously every Transmission Time Interval (TTI). This can lead to better frame-fill efficiency and less padding overhead.

The external code entity can generate the payload every Transmission Time Interval (TTI) if required. Protocol Data Units (PDUs) can be constructed in real time as Service Data Units (SDUs) can be received from higher layers. RLC may add padding if there is not enough data to construct a protocol data unit (PDU).

Encoder Packet (EP) with fixed row size

When decoding SDUs 201-204, it is desirable to minimize the amount of padding that will be transmitted.

In one embodiment, the row size of encoder packet (EP) matrix 205 may be a fixed size. A priori knowledge of the size of the 205 rows of the Encoder Packet (EP) matrix allows to arrange the data back to their original configuration. Because the row size of the SDUs 201-204 to be sent is known in advance, the transfer can begin as soon as the data is received without waiting to see how much data will be sent.

Figure 12A shows an example of a decoding process for generating an outer code block 214 from data units 201-204, where the row size of the outer code block 214 may be fixed. In this example, user data is in the form of a number of Service Data Units (SDUs) 201-204 comprising bit blocks of arbitrary size, where the size of the bit blocks depends on the particular application (video, voice, etc.).

In order to be able to transmit FEC SDUs of arbitrary size, segmentation, concatenation and padding can be performed at the FEC level. Although splicing is not strictly necessary, its absence can result in a significant drop in data throughput for higher layers.

The higher layer SDUs 201-204 may first be formatted into a fixed PDU size. In this embodiment, the segmenting/stitching function generates fixed size internal blocks that can be indicated to subscriber units. At step 220, the set of internal blocks may be segmented and concatenated to become part of an encoder grouping matrix 205 comprising the internal blocks, the necessary degree of padding 208, and a length indicator (LI) 206 , the length indicator 206 can be used to indicate where a Service Data Unit (SDU) 201-204 ends by indicating how many SDUs end up in a given line of the EP. The outer encoder, discussed below, uses these inner blocks to generate redundant blocks.

In radio link control (RLC), the length indicator (LI) indicates where each service data unit (SDU) ends, where each service data unit is relative to a protocol data unit (PDU) rather than a service data unit (SDU) identified. This helps reduce overhead since the PDU size is usually smaller than the Service Data Unit (SDU) size. For example, a Length Indicator (LI) may be used to indicate the last octet of each FEC Service Data Unit (SDU) ending within a Payload Data Unit (PDU). The "length indicator" may be set to the number of octets between the end of the FEC header and up to the last octet of the FEC SDU segment. A length indicator (LI) may be completely included in the PDU to which the length indicator (LI) refers. In other words, the Length Indicator (LI) preferably refers to the same Payload Data Unit (PDU), and preferably in the same order as the FECSDU referred to by the Length Indicator (LI)

When an outer block is received, information such as a length indicator (LI) can be used to let the receiver know where the service data unit (SDU) and/or padding starts and ends.

Since it is impossible to indicate the presence of the Length Indicator (LI) with 1 bit in the FEC header, the FEC layer adds a fixed header indicating the presence of the Length Indicator (LI) within the payload. The inner header or LI provides all the information needed to reconstruct the SDU 201-204. LI may be included in the RLC-PDU it refers to. The presence of the first LI may be indicated by a tag included in the sequence number header of the RLC-PDU. A bit in each LI can be used to indicate its extension. In order to allow the length of the length indicator (LI) to vary with the size of the FEC PDU, a new special value can be introduced for the one-byte length indicator (LI) to indicate that a previously terminated SDU lacked one byte to fill the last PDUs. The Length Indicator (LI) presence bit can be implemented in various ways, two of which are discussed below.

In one embodiment, a length indicator (LI) presence bit may be provided in each protocol data unit (PDU). For example, a byte may be added at the beginning of each encoder packet (EP) line, and a bit in that byte indicates the presence of LI. The entire first byte of each protocol data unit (PDU) may be reserved for this "presence bit". To accommodate this presence bit, the length indicator data may be shortened by one bit. Presence bits are provided in each small unit (PDU), allowing decoding of SDUs when EP decoding fails, even when the first PDU is lost. This results in a lower residual bit error rate. Providing a presence bit in each PDU also allows real-time splicing/segmentation.

In another embodiment, a length indicator (LI) presence bit may be provided in the first PDU. Instead of adding overhead at the beginning of each PDU, the presence bits for all k information PDUs can be added at the beginning of the first PDU of the EP. Presence bits are provided at the beginning of the encoder packet (EP), resulting in less overhead when having large SDUs and/or small PDUs.

After segmentation and splicing, EP 205 includes a number of rows occupied by at least one of a plurality of service data units (SDUs) 201-204 and stuffing blocks. The row size of the outer blocks can be designed so that each row can be transmitted at the peak data rate during one transmission time interval (TTI). Service Data Units (SDUs) with an amount of data sent during a Transmission Time Interval (TTI) generally cannot be arranged in a row. Therefore, as shown in FIG. 11, the second and fourth SDUs 202, 204 do not fit in the transmission time interval (TTI) of the first and second row of the EP, respectively. In this example, the EP has 12 rows available for data, and can group the four SDUs 201-204 into the first three of the 12 rows. The remaining rows of EP 205 may be occupied by filler blocks 208. In this way, the second SDU 202 can be segmented such that a first part of the second service data unit (SDU) 202 begins in the first line of the "chunk" and a second part of the second SDU 202 ends in the second line. Similarly, the third SDU must be segmented so that the first part of the third service data unit (SDU) 203 starts in the second row, and the second part of the third SDU 203 ends in the third row. A fourth service data unit (SDU) 204 fits within the third row, and the remainder of the third row can be filled with filler blocks 208 . In this example, encoder packet (EP) 213 consists primarily of padding 208 .

Encoders use EPs to generate redundant or parity information. In step S240, the encoder encodes the intermediate grouping matrix 205 encoded by adding the outer parity block 214 to generate the outer code block 213 having a length of 16 blocks. The encoder extracts 8 bits of data from each column of each block to generate the resulting data 210 . A Reed Solomon (RS) encoder encodes the resulting data 210 to obtain four rows of redundancy or parity information 212 . Parity information 212 may be used to generate an outer parity block 214 that may be appended to EP matrix 205 to generate 16 outer code blocks 213 .

Figure 12B shows an example of information transmitted over the air in the examples discussed above. In step S260, after adding the additional overhead including the sequence number to each line of the EP 205, the 16 external code blocks 213 can be transmitted over the air as a protocol data unit (PDU) 214. All or the entire encoder packet (EP) 213 matrix is not transmitted in a protocol data unit (PDU) 214 sent on the downlink. Instead, a protocol data unit (PDU) includes information bits 201 - 204 and a length indicator (LI) 206 of an encoder packet (EP) matrix 213 . Since the row size of the encoder packet (EP) 213 is fixed and thus known at the receiver, there is no need to actually transmit the padding 208 over the air. Padding information 208 is not transmitted on the downlink, since the padding value is known, so there is no need to transmit padding information 208 . For example, if the stuffing can consist of a known bit sequence, such as a bit sequence of all 0s, all 1s, or an alternating structure of 0s and 1s, the receiver can stuff protocol data units (PDUs) 214 up to standard encoder packets ( EP) 213 line length. Therefore, during transmission, instead of choosing a PDU size equal to the EP row size, the smallest EP size available that transmits all information bits 201-204 and reassembly overhead 206 (eg LI) can be used.

Although the encoder matrix row size is fixed, the FEC PDU size may be chosen from a given set at each transmission such that each FEC PDU includes all information parts of a single encoder matrix row (padding may be excluded). When receiving a PDU with a size smaller than the encoder matrix row size, the UE can pad up to that size with a known bit sequence. This allows the internal block size to remain fixed without increasing the load on the air interface. Thus, using a fixed row size encoder packet (EP) 213 eliminates the need to wait until all data is available before starting to transmit a Protocol Data Unit (PDU), and also eliminates the need to send padding.

If the above algorithm is implemented to handle variable rate transmissions, a rate equalization scheme can be used in which all encoder block matrix rows have a constant size. Smaller PDUs can be used when padding forms part of the PDU. Padding can consist of a specific sequence of bits and can be placed right at the end of the data. At the receiver, the size of the block received from the lower layer can be equal to the base-line size by appending padding at the end.

If a predetermined bit sequence is available for padding, this padding is not transmitted over the air. The receiver does not need to know the actual encoder packet line size, unless the receiver needs to do external decoding. Basic SDU reassembly does not require knowledge of the amount of padding at the end of the PDU. If all PDUs containing information from the first k Encoder Packet (EP) rows are received, no outer decoding is necessary. Conversely, if at least one PDU containing information from the first k Encoder Packet (EP) lines is lost, then at least one PDU containing data from even and odd lines is required. Since parity rows are usually not padded, the size can be used as a reference for the actual encoder packet size that needs to be assumed.

Encoder Packet (EP) with variable row size

Figure 13 shows the encoding process for generating an outer code block 313 with a variable line size.

This aspect of the invention relates to flexible outer block encoding of data transmitted over the air interface. This encoding process results in less padding being transmitted in order to increase frame filling efficiency. A row of encoder packets (EPs) 305 may be of variable size, and outer blocks of different sizes may be sent each transmission time interval (TTI). The row size of the encoder packet (EP) 305 is preferably varied so that the SDUs fit into exactly a few rows (eg, 12 rows) of the encoder packet (EP) matrix 305 . In this embodiment, the FEC layer must wait for all data to be available before constructing the EP so that the FEC layer can determine the optimal row size. The row size can be chosen from a number of different sizes based on the amount of data available to limit padding. The row size of the encoder packet (EP) can be linked to the set of PDU sizes configured for S-CCPCH. Depending on the amount of data available when the encoder packet 305 needs to be generated, the line size that results in minimal padding may be chosen. By reducing the size of the outer block 313 so that the block size in each frame can be smaller, data can be sent at a reduced transmission rate because less data is sent in the same TTI duration. Using variable row size encoder packets (EPs) 305 helps to stabilize the power requirements for all transmissions of encoder packets (EPs) and also uses less parity overhead 314 . This embodiment is useful for point-to-multipoint (PTM) transmissions in systems such as WCDMA, where the underlying radio protocol allows the size of the transport block sent in each Transmission Time Interval (TTI) to be variable of.

At step 320, a plurality of service data units (SDUs) 201-204 may be segmented and concatenated to generate an encoder packet (EP) matrix 305, where a length indicator (LI) 206 may be used to indicate that a service data unit (SDU) 201 The ending position of -204. A length indicator (LI) may be included in the last line in which each service data unit (SDU) terminates.

In step 330, redundancy or parity information is generated column by column by extracting 8 bits of data from each data block, and the resulting data 310 may be sent to a Reed Solomon (RS) encoder to obtain parity information 312 . Since the rows of the encoder grouping (EP) matrix 305 are smaller, less redundant information may be generated.

Encoding continues at step 340, since the parity information 312 is used to generate the outer parity block 314, which can be appended to the 12-block encoder packet (EP) matrix 305, thereby generating a block of length 16 in this example. the outer block of code. This embodiment avoids the transmission of padding, which improves transmission efficiency since the entire outer code block 313 is occupied by SDUs, Length Indicators (LI) 206 and/or redundant information 314 . In this particular instance, no padding is required. However, it should be understood that in some cases the number will be limited due to the configured size of the PDU and some padding may be required, albeit to a reduced amount. This results in higher frame filling efficiency and may also allow more constant power to be maintained across the entire Encoder Packet (EP). This is desirable in a CDMA system using a power control scheme.

Although not shown, the transmission of the PDU over the air will proceed in a manner similar to that discussed above with respect to step S260 of FIG. 12 .

11 is an embodiment of an outer coding or forward error correction (FEC) layer 400 with RLC Unacknowledged Mode (UM)+ entity (RLC UM+) provided on the Radio Link Control (RLC) layer. Here, the FEC layer located above Radio Link Control (RLC) performs framing.

The outer coding layer 400 comprises a transmitting forward error correction (FEC) entity 410 which communicates with a receiving forward error correction (FEC) entity 430 via a logical channel 406 over a wireless interface (Uu) 404 .

Reorder/Duplicate Detection

Figure 15 is a reordering protocol or algorithm for enabling the mobile station 10 to delay decoding by a time offset between different logical streams.

The receive forward error correction (FEC) entity 430 uses the sequence number to determine the location of a given PDU within the EP matrix. For example, a Part Sequence Number (PSN) identifies the position of a PDU within an Encoder Packet (EP).

This algorithm assumes that at most two encoder packets (EPs) of data are received before decoding can be initiated. In the following description, an encoder packet (EPd) is an encoder packet (EP) to be decoded next in order, and an encoder packet (EPb) is an encoder packet (EP) that is being buffered. The encoder packet (EPb) follows the encoder packet (EPd). UE implementations that require the full encoder packet transfer time to perform RS decoding will require double buffering in order to be able to decode consecutive packets. Therefore, the UE stores at least n+k rows of maximum size of the encoder matrix (k and n being the number of information rows and the total number of rows including odd and even rows, respectively). UEs with faster decoding engines can reduce this requirement, although not below n+1. For example, if a UE has a certain amount of buffer space (XtraBffr) beyond what it needs to receive consecutive packets based on its decoding capabilities, and if a 64kbps stream is assumed, delaying decoding by 100ms without increasing computational requirements would require The buffer size is increased by 800 bytes.

At block 1410, it may be determined whether a new forward error correction (FEC) protocol data unit (PDU) has been received. If no new Forward Error Correction (FEC) Protocol Data Units (PDUs) have been received, processing resumes at block 1410 . If a new forward error correction (FEC) protocol data unit (PDU) has been received, at block 1420, it can be determined whether the new forward error correction (FEC) protocol data unit (PDU) belongs to the next sequence to be decoded in order. An Encoder Packet (EPd).

If the forward error correction (FEC) protocol data unit (PDU) does not belong to the next encoder packet (EP) to be decoded in order, then at block 1421, it may be determined that the forward error correction (FEC) protocol data unit Whether the (PDU) belongs to the encoder packet (EPb) being buffered. If the forward error correction (FEC) protocol data unit (PDU) does not belong to the encoder packet (EPb) being buffered, then at block 1440, the protocol data unit (PDU) may be discarded. If the forward error correction (FEC) protocol data unit (PDU) does belong to the encoder packet (EPb) being buffered, then at block 1423, the protocol data unit (PDU) may be added to the buffer of EPb at the relevant location middle. At block 1425, it may be determined whether the amount of data for EPb exceeds XtraBffr. If it is determined at block 1426 that the amount of data for EPb does not exceed XtraBffr, processing restarts at block 1410. If the amount of data for EPb exceeds XtraBffr, then at block 1428 the transmitting entity attempts to transmit a complete SDU from EPd. Then, at block 1430, the remainder of EPd may be cleared from the buffer, and at block 1434, EPb may be set to EPd.

If at block 1420 it is determined that the forward error correction (FEC) protocol data unit (PDU) belongs to EPd, then at block 1422 the protocol data unit (PDU) may be added to the buffer of EPd at the relevant location. At block 1424, it may be determined whether the buffer has k separate PDUs for EPd. If the buffer does not have k separate PDUs for EPd, at block 1426 , processing resumes at block 1410 . If the buffer does have k separate PDUs for EPd, then at block 1427 the decoder performs outer decoding for EPd, and then at block 1428 the transmitting entity attempts to transmit the complete SDU from EPd. Then, at block 1430, the remainder of EPd may be flushed from the buffer, and at block 1434, EPb may be sent to EPd.

Figure 16 is a diagram showing the outer code blocks received by a mobile station as it transitions between a point-to-multipoint (PTM) transmission received from cell A 99 and another point-to-multipoint (PTM) transmission from cell B 99 A graph of the time relationship between. Some aspects of FIG. 16 are described in U.S. Patent Applications US-2004-0037245-A1 and US-2004-0037246-A1 to Grilli et al., filed August 21, 2002 and to Willenegger et al., filed May 6, 2002. There is further discussion in US Patent Application US-2003-0207696-A1, which is hereby incorporated by reference in its entirety.

The described scenario assumes a certain UMTS Terrestrial Radio Access Network (UTRAN) 20 and User Equipment (UE) 10 requirements. For example, if the UTRAN 20 transmits content coded using the same outer block across cells, the same number shall be used on blocks carrying the same data or payload in neighboring cells. Outer blocks with the same number are transmitted in a relatively time-aligned manner. The maximum misalignment of PTM transmissions across cells is controlled by a radio network controller (RNC) 24 . The UTRAN 20 controls delay jitter on point-to-multipoint (PTM) transmissions across cells. The UE 10 should be able to decode the current outer block while the next outer block is being received. Therefore, the buffer space in the UE should preferably accommodate at least two outer blocks 95A-95C, since memory for one outer block is required to accumulate the current outer block. The memory should also be able to accumulate "rows" of inner blocks to compensate for inaccuracies in time alignment across base stations 22 if the outer blocks are during Reed-Solomon (RS) decoding.

In cell A 98, during the transmission of the outer block n 95A, the transition occurs during the transmission of the second inner Multimedia Broadcast and Multicast Service (MBMS) payload block. The transition of the user equipment (UE) 10 from cell A 98 to cell B 99 is illustrated by the diagonal arrow 96, which is non-horizontal because some time has passed during the transition. By the time the user equipment (UE) 10 arrives at cell B 99, a fifth piece of Multimedia Broadcast and Multicast Service (MBMS) payload data is being transmitted. Thus, the user equipment (UE) 10 misses the second to fourth blocks due to the time offset of the respective transmissions and the time elapsed during the transition. If enough blocks are received in cell B 99, outer block n 95A can still be decoded because the parity blocks can be used to reconstruct missing blocks.

Subsequently, during the transmission of outer block n+295C, the user equipment (UE) 10 undergoes another transition from cell B 99 to cell A 98, which occurs at the fifth inner multimedia broadcast and multicast of outer block n+295C service (MBMS) payload block. In this case, fewer internal blocks are lost during conversion, and external blocks are still recoverable.

The use of external code blocks can help reduce the possibility of any service interruption. To ensure effective error recovery, the same blocks should be sent on each transmission path, which means that parity blocks should be constructed in the same way on each transmission path. (Multimedia Broadcast and Multicast Service (MBMS) payload blocks must be the same on each path since it is a broadcast transmission). Performing forward error correction (FEC) at the upper application layer 80 helps to ensure that the parity blocks are identical on each transmission path, since the encoding takes place in the forward error correction (FEC) layer 157 so that for each outer block Are the same. In contrast, if the encoding is done at a lower layer, eg at a separate Radio Link Control (RLC) entity 152, some coordination is required since the parity blocks will be different in each transmission path.

Conversion from point-to-multipoint (PTM) to point-to-point (PTP)

FIG. 17 is a diagram showing the temporal relationship between outer code blocks received by the mobile station 10 when switching between point-to-multipoint (PTM) transmission and point-to-point (PTP) transmission occurs. The scheme shown in Fig. 17 applies, for example, to systems using point-to-point (PTP) transmission, such as WCDMA and GSM systems.

One aspect of the invention relates to forward error correction by adding parity information or blocks to the internal MBMS "payload" or data blocks during PTM transmission. Each outer code block transmitted in a PTM transmission includes at least one inner payload block and at least one inner parity block. The error correction capability of the outer code block can significantly reduce and often eliminate the loss of MBMS content or "payload" during transitions, such as when the UE moves from one cell to another, or when the MBMS When the transfer of content changes from a PTM connection to a PTP connection, or when a change in the opposite direction occurs.

As mentioned above, a given cell may transmit to users 10 using either a PTP or PTM transmission scheme. For example, if the demand for services in a cell falls below a certain threshold, a cell that normally uses PTM transmission mode to deliver broadcast services may choose to establish a dedicated channel and transmit it in PTP mode (only to a certain user 10). Likewise, a cell that normally transmits content to individual users on a dedicated channel (PTP) may decide to broadcast content to multiple users over a common channel. Furthermore, a given cell may transmit content in a PTP transmission mode, while another cell may transmit the same content in a PTM transmission mode. Transitions occur when the mobile station 10 moves from one cell to another, or when a change in the number of users within a cell triggers a change in the transmission scheme from PTP to PTM or in the opposite direction.

During the point-to-multipoint (PTM) transmission of the outer block n 95A, switching occurs during the transmission of the fourth inner Multimedia Broadcast and Multicast Service (MBMS) payload block. Slashed arrow 101 illustrates a user equipment (UE) transition from point-to-multipoint (PTM) to point-to-point (PTP) transmission, which is non-horizontal because some time elapses during the transition. When the transition from PTM 101 to PTP occurs, the bit rate over the air remains approximately the same. Typically, point-to-point (PTP) transmissions have a bit error rate of less than one percent (eg, one or fewer bit errors in every 100 payload blocks during transmission). In contrast, in point-to-multipoint (PTM) transmission, a higher bit error rate can be assumed. For example, in one embodiment, the base station generates outer blocks every 16 transmission time intervals (TTIs), and 12 of these TTIs may be occupied by payload blocks and 4 TTIs may be occupied by parity blocks. The maximum number of tolerable errored blocks shall be 4 inner blocks out of 16 (12 basic blocks + 4 parity blocks). Thus, the maximum tolerable block error rate will be 1/4.

When a mobile station switches 101 from point-to-multipoint (PTM) to point-to-point (PTP) transmission, some internal blocks may be lost. Assuming point-to-multipoint (PTM) transmission and point-to-point (PTP) transmission have approximately the same bit rate at the physical layer (L1), then PTP transmission will allow MBMS payload blocks to be transmitted faster than PTM transmission because, on average, The percentage of blocks that are retransmitted will typically be lower than the percentage of parity blocks. In other words, point-to-point (PTP) transmissions are typically much faster than point-to-multipoint (PTM) transmissions, statistically speaking, the number of parity blocks is greater than the number of radio link control (RLC) retransmissions (Re-Tx) Much bigger. Since the transition 101 is from point-to-multipoint (PTM) transmission to typically much faster point-to-point (PTP) transmission, when user equipment (UE) 10 transitions 101 to point-to-point (PTP) transmission, multimedia broadcast and multicast The first block of service (MBMS) payload data is being transmitted. Thus, neither the time offset of the corresponding transmission nor the time elapsed during the transition 101 has resulted in the loss of any blocks. Therefore, when moving from point-to-multipoint (PTM) to point-to-point (PTP) transmission, once the PTP link is established in the target cell, it is possible to compensate for the lost payload block. The network can compensate by starting the PTP transmission from the beginning of the same outer block, ie starting transmission with the first inner block. The network can then recover the delay introduced by the transition due to the faster delivery of complete outer blocks. The loss of data during transmission is reduced, reducing the interruption of MBMS content delivery that may be caused by such switching.

Subsequently, during the PTP transmission of outer block n+2, the user equipment (UE) 10 undergoes another transition 103 to point-to-multipoint (PTM) transmission mode. In Figure 12, this transition 103 from point-to-point (PTP) to point-to-multipoint (PTM) occurs in the last inner Multimedia Broadcast and Multicast Service (MBMS) payload block of outer block n+2. In this case, except for the last inner block, many inner Multimedia Broadcast and Multicast Service (MBMS) payload blocks in outer block n+2 have been transmitted. Typically, FEC is used in such cases where feedback is not available. The use of FEC is not beneficial since PTP transmission uses a dedicated channel, thus having feedback capability on the reverse link. To minimize or eliminate data loss during handover, UMTS Terrestrial Radio Access Network (UTRAN) 20 preferably relies on the low residual block error rate of RLC Acknowledged Mode (AM) in PTP transmission to recover from handover to PTM transmission All internal blocks that may be lost during this period. In other words, normal layer 2 retransmissions can be used to retransmit any packet for which an error was detected in the source transmission. Therefore, as shown in Fig. 17, no parity block is required in PTP transmission. If there are bit errors in the payload blocks during point-to-point (PTP) transmission, the outer blocks can still be decoded because the radio link control (RLC) layer will request retransmission of any erroneous blocks. That is, when there is a bit error during PTP transmission, the mobile station 10 either requests a retransmission (Re-Tx), or when all blocks are correct, no retransmission occurs and Transport Format Zero (TF0) may be used. The outer encoding is preferably done in layer 2 of the protocol stack so that each inner block 97 is sized to fit into exactly one transmission time interval (TTI), as this enhances encoding efficiency.

If Forward Error Correction (FEC) outer coding is done at an upper layer of the protocol stack, such as at the application layer, parity blocks will be sent regardless of the retransmission scheme (point-to-point (PTP) or point-to-multipoint (PTM) )). Thus, parity blocks will also be appended to point-to-point (PTP) transmissions.

As mentioned above, in PTP transmission, the use of parity blocks is not necessary, since a more efficient retransmission scheme can be used instead of forward error correction. Since parity blocks are preferably not transmitted in PTP transmission, the transmission of complete outer blocks can be on average faster than in PTM, assuming the same over-the-air bit rate. This allows the UE to compensate for interruptions caused by point-to-multipoint (PTM) to point-to-point (PTP) transitions, since PTP transmissions can be expected relative to PTM transmissions. The user equipment (UE) can be transmitted by combining (1) the internal block received in the new cell or in the point-to-point (PTP) transmission after switching, and (2) the point-to-multipoint transmission in the old cell or before switching (PTM) transfer of received internal blocks to correctly recover external blocks. A user equipment (UE) may combine an internal block received before conversion and an internal block received after conversion belonging to the same external block. For example, the user equipment (UE) 10 may combine the inner Multimedia Broadcast and Multicast Service (MBMS) payload block in the outer block n+2 received via point-to-point (PTP) transmission, and receive via point-to-multipoint (PTM) transmission The outer block n+2 of and the inner Multimedia Broadcast and Multicast Service (MBMS) payload block in the parity block. The UMTS Terrestrial Radio Access Network (UTRAN) 20 can facilitate this by slightly "anticipating" the transmission of the outer chunks for all users receiving MBMS content from the PTP link relative to the transmission over the PTM link.

A "seamless" transition from PTP to PTM is possible because UTRAN anticipates the transmission of outer blocks relative to PTM transmission. As a result, the transfer of MBMS content across cell boundaries and/or between different transmission schemes, such as PTM and PTP, is also "seamless". This "time expectation" can be represented by the number of internal blocks. When the user equipment (UE) 10 transitions to PTM transmission, the user equipment (UE) 10 can lose up to a "time expected" number of internal blocks, even if no communication link exists during the transition time, without compromising the quality of MBMS reception. QoS. If the UE starts MBMS reception directly in PTP, UTRAN can apply "time anticipation" immediately at the start of PTP transmission, because UTRAN 20 can slowly anticipate the transmission of outer blocks by avoiding empty inner blocks (TF 0) until the expectation is reached The "time expected" number of internal blocks requested so far. From that point on, UTRAN can maintain a constant "time expectation".

In point-to-multipoint (PTM) transmission, UE-specific feedback information available in the Radio Network Controller (RNC) cannot be relied upon. In point-to-point (PTP) transmission, the UE 10 can inform the RNC of the number of the last external block that was correctly received before switching. This should apply to any conversion to PTP (either from PTM or from PTP). If this feedback is not considered acceptable, the UTRAN 20 can estimate the last outer block most likely received by the user equipment (UE) 10 before the state transition. This estimate may be based on knowledge of the maximum foreseeable temporal inaccuracy between different cell transmissions, and may be based on outer blocks currently being transmitted or about to be transmitted in the target cell.

Forward Error Correction (FEC) can be performed so that any blocks lost during the transition can be recovered. This produces a "seamless" transition by reducing the likelihood that content will be lost during the transition. This scheme assumes that the transition from point-to-point (PTP) to point-to-multipoint (PTM) transmission occurs when the same external block is being transmitted from each source, which typically occurs at a given external block relative to the duration of the transition. in the case of duration.

The amount of memory in the UE 10 can be traded off with the accuracy in the time alignment of the PTM transmissions across neighboring cells. By relaxing the memory requirements in the user equipment (UE) 10, the time accuracy of PTMUTRAN 20 transmissions can be increased.

Fig. 18 is a diagram showing the transmission by a mobile station during a transition or relocation between a point-to-point (PTP) transmission from a radio network controller (RNC) A and another point-to-point (PTP) transmission from a radio network controller (RNC) B. A graph of the temporal relationship between received external code blocks. The term RNC is used interchangeably with the term "Base Station Controller (BSC)". During "relocation", the user equipment (UE) 10 switches from point-to-point (PTP) transmission of a content flow in the area controlled by the first RNC A 124 to the same content flow in the area controlled by the second RNC B 224 point-to-point (PTP) transmission. Retransmissions (re-Tx) can be used to compensate for any lost MBMS payload blocks. Direct transitions between cells from point-to-point (PTP) to point-to-point (PTP) may be performed similar to Release '99 soft handoff or hard handoff. Even without coordination between the two RNCs A, B, the target RNC A 124 should be able to figure out the latest entire external block received by the UE 10. This estimation may be based on the moment of MBMS content received by the RNC 24 over the Iu interface 25 . When using PTP transport, the RNC 24 can compensate for the initial delay and will not lose parts of the MBMS even if no lossless SRNS relocation is required.

Those skilled in the art will appreciate that although the flow chart may be drawn sequentially for ease of understanding, certain steps may be performed in parallel in an actual implementation. Also, method steps may be interchanged without departing from the scope of the invention unless clearly indicated otherwise.

Those of skill in the art would understand that information and signals may be represented using any of a number of different technologies and techniques. For example, the data, instructions, commands, information, signals, bits, symbols, and chips mentioned in the above description can all be expressed as voltage, current, electromagnetic wave, magnetic field or magnetic particle, light field or light particle, or the above combined.

Those skilled in the art can further appreciate that the logical blocks, modules, circuits, and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described in the foregoing description generally in terms of their functionality. Whether such functionality can be implemented in software or hardware depends upon the particular application and design constraints imposed on the overall system. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered as exceeding the scope of the present invention.

The logic blocks, modules, and circuits of various examples described in conjunction with the embodiments disclosed herein can be implemented with general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or Implemented or performed by other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination of the above designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computer devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors with a DSP core, or any other such configuration The combination.

Each step of the method or algorithm described in conjunction with the embodiments disclosed herein may be directly implemented by hardware, a software module executed by a processor, or a combination of the two. The software module can be placed in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable hard disk, CD-ROM, or any other form of storage medium known in the technical field. An exemplary storage medium is coupled to the processor so the processor can read information from, and write information to, the storage medium. Alternatively, the storage medium may be integrated in the processor. The processor and storage medium can be placed in an ASIC. The ASIC can be placed in the user premises. Alternatively, the processor and the storage medium may be placed in the user terminal as separate components.

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. For example, although the description indicates that the radio access network 20 may be implemented using a Universal Terrestrial Radio Access Network (UTRAN) air interface, alternatively, in a GSM/GPRS system, the access network 20 may be a GSM/EDGE radio The Access Network (GERAN), or in the intersystem case it may comprise cells of the UTRAN air interface and cells of the GSM/EDGE air interface. Accordingly, the present invention will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Portions of this disclosure document contain copyrighted material. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the Patent and Trademark Office patent file or records, otherwise all copyrights are reserved under all circumstances.

Claims (20)

1. A system for transmitting information blocks, comprising: Destination stations including decoders; a first specific content source that transmits a first external block over a common channel using a first data transmission mode for reception by said destination station, wherein said first external block comprises an information block and a redundant block usable to reconstruct said information block; wherein said redundant block is generated above a radio link control layer; and a second specific content source that transmits a second external block comprising information blocks over a dedicated channel using a second data transfer mode for receipt by said destination station when said destination station undergoes a transition, wherein said second outer block has a time shift relative to said first outer block, and Wherein said decoder is configured to use said redundant blocks to reconstruct any information blocks lost during said conversion, wherein each information block includes a sequence number comprising an outer block number and an inner block number. 2. The system of claim 1, wherein the transmission rate of the information chunks from the second specific content source is greater than the transmission rate of the information chunks from the first specific content source, and wherein the Said second specific content source starts transmission from a first information block of said first outer block being transmitted when said transition occurs, thereby reducing loss of information blocks during said transition. 3. The system of claim 1, wherein if during another transition from the second specific content source to the first specific content source, any information block from the second outer block is incorrectly If received, the second specific content source retransmits the second external block through the dedicated channel. 4. The system of claim 1, wherein each of the first outer block and the second outer block occupies a frame. 5. The system of claim 1 , wherein the destination station combines the information blocks received from the first specific content source when the transition occurs while the same external block is being transmitted from each source , and the information block received from the second source to generate a complete external block. 6. The system of claim 1, wherein the first specific content source includes a Reed-Solomon encoder that encodes the information block to generate the redundant block and adds the redundant block to the In the above information block to generate the first outer block. 7. The system of claim 1, wherein the decoder decodes the first outer block and regenerates any missing information blocks from the redundant blocks. 8. A method for a destination station to receive content, wherein the content comprises a plurality of internal content blocks, the method comprising: formatting the content as a first outer block comprising the inner content block and an inner parity block based on the inner content block, the inner parity block being defined above the radio link control layer generate, each internal content block includes a sequence number, the sequence number includes an external block label and an internal block label; transmitting the first outer block over a common channel; receiving at least a portion of the first outer block; formatting said content as a second outer block comprising said inner content block; transmitting the second external block over a dedicated channel; switching from delivery of said content over said common channel to delivery of said content over said dedicated channel; receiving at least a portion of the second outer block, wherein the second outer block is time-shifted relative to the first outer block; and using said portion of a first outer block received before said transformation occurs, together with said portion of a second outer block, to generate a complete outer block comprising said plurality of inner content blocks, thereby compensating for said The time shift between the first outer block and the second outer block. 9. The method of receiving content as claimed in claim 8, wherein when any of said inner content blocks delivered over said common channel were not correctly received, said inner parity block is used to reconstruct all said inner content blocks which were not correctly received. Describe the inner content block. 10. A method of receiving content as claimed in claim 9, wherein said portion of a first outer block received before said conversion occurs is used together with said portion of a second outer block to generate an a complete outer block of said plurality of inner content blocks, thereby compensating for said time shift between said first outer block and said second outer block, comprising: determining a last external block number and a last internal block number corresponding to the sequence number of the last internal content block received over said common channel; and A new external block index and a new internal block index corresponding to said sequence number of the first internal content block received over said dedicated channel are determined. 11. The method of receiving content as recited in claim 10, wherein if said last outer block index is equal to said new outer block index, said received portion of a first outer block and said second outer block are combined. block, thereby compensating for the time shift between the first outer block and the second outer block, wherein the second outer block starts at the first inner block number with the next inner block number following the last inner block number Two outer blocks of the inner content block. 12. The method for receiving content as claimed in claim 11, further comprising: If said last inner block index is greater than said new inner block index, then discard said inner content block of said second outer block until said inner content block of a second outer block has a value equal to said last inner block Labeled inner block numbering, thereby compensating for the time shift between the first outer block and the second outer block. 13. The method of receiving content as recited in claim 11 , further comprising: If said last internal block index is less than said new internal block index, storing said second external block starting at the first internal content block of said second external block, thereby compensating said first external block and The time shift between the second outer block. 14. The method of receiving content of claim 10, wherein the time shift comprises a difference in time alignment between the first outer block and the second outer block. 15. A method for seamlessly delivering broadcast and multicast content to terminals, comprising: point-to-multipoint (PTM) transmission using outer block codes, wherein each outer block code includes an inner data block and an inner parity block, each inner data block includes a sequence number including an outer block number and an inner block number, said inner parity block is generated above a radio link control layer; When the terminal switches from point-to-multipoint (PTM) transmission to point-to-point (PTP) transmission, by starting the point-to-point (PTP) transmission at the beginning of the same outer block code, relative to the point-to-multipoint point-to-point (PTM) transmission, contemplated transmission of an outer block code during said point-to-point (PTP) transmission, wherein the transfer of a complete outer block code via said point-to-point (PTP) transmission is faster than via point-to-multipoint (PTM) transmission transmission of ; and Faster delivery of complete outer block codes via the point-to-point (PTP) transfer, recovering any delay introduced by the conversion to recover any lost inner data blocks, allowing transfer from the point-to-point (PTP) transfer to the Seamless switching of point-to-multipoint (PTM) transmissions. 16. The method of claim 15, wherein the inner parity block is not transmitted in the point-to-point (PTP) transmission. 17. The method of claim 16, wherein starting the point-to-point (PTP) transmission from the beginning of the same outer block code comprises: The point-to-point (PTP) transmission starts at the beginning of the first inner data block. 18. A method for seamless delivery of broadcast and multicast content to terminals during transitions across cell boundaries, comprising: point-to-multipoint (PTM) transmission using outer block codes, wherein each outer block code includes an inner data block and an inner parity block, each inner data block includes a sequence number including an outer block number and an inner block number, said inner parity block is generated above a radio link control layer; By starting point-to-point (PTP) transmission from the beginning of the same outer block code when the terminal performs a transition across cell boundaries, with respect to the point-to-multipoint (PTM) transmission, it is expected that the point-to-point ( transmission of the outer block code during PTP) transmission, wherein the transmission of the complete outer block code via said point-to-point (PTP) transmission is faster than transmission via point-to-multipoint (PTM) transmission; and Faster delivery of complete outer block codes via the point-to-point (PTP) transfer, recovering any delay introduced by the conversion to recover any lost inner data blocks, allowing transfer from the point-to-point (PTP) transfer to the Seamless switching of point-to-multipoint (PTM) transmissions. 19. The method of claim 18, wherein no inner parity blocks are transmitted in the point-to-point (PTP) transmission. 20. The method of claim 19, wherein starting said point-to-point (PTP) transmission from the beginning of said same outer block code comprises: The point-to-point (PTP) transmission starts at the beginning of the first inner data block.

Images