JP5149267B2 - Wireless communication system, wireless communication method, transmitter, and receiver - Google Patents

Description

  The present invention relates to a wireless communication system, a wireless communication method, a transmitter, and a receiver, and more particularly, to a wireless communication system, a wireless communication method, a transmitter, and a receiver that can realize adaptive error recovery.

Because the wireless medium has unique properties, its protocol design is very different and difficult from that of wired networks. The main features of the wireless system and its medium are that the channel varies over time and is prone to errors. Radio signals propagate according to three mechanisms: reflection, diffraction, and scattering. The signal received at the node is a signal after the transmission signal is attenuated over time. As a result, the strength of the received signal varies with time. This phenomenon is called “multiple propagation”. The rate of channel variation is determined by the coherent time of the channel. When the received signal strength falls below a predetermined threshold, the node is considered to be in a “fade state”. In wireless transmission, the channel and signal strength fluctuate over time, and errors are likely to occur due to the effects of obstacles such as natural background noise, interference, and movement. In a wired network, since the bit error rate is less than 10 ^−6, the probability that a packet error will occur is low. On the other hand, since the bit error rate is 10 ⁻³ or more in the wireless channel, the probability that a packet error occurs is much higher. In the case of a wired network, random noise is the main cause of packet errors. On the other hand, errors on the radio link occur in long bursts when the node is in a fade state. Packet loss due to burst errors can be minimized by reducing packets, using forward error correction codes, and retransmitting. The design of the link layer error recovery scheme affects network capacity, transmission station placement, and higher layer protocol design.

  Error recovery problems include wireless local area network (WLAN) (802.11), wireless metropolitan area network (WMAN) (802.16), wireless wide area network (WWAN) (3G, LTE, IMT-advanced), etc. Present in most wireless communication networks. Therefore, it is essential to incorporate an appropriately designed error control scheme in next generation wireless networks.

  A bit error in wireless communication occurs when interference or noise exceeds an encoded or modulated transmission. The reliable communication protocol requires that all intended message recipients receive a message in the same state as when it was sent. Error control techniques that can cope with errors during wireless transmission are roughly classified into two types: “forward error correction (FEC)” and “automatic repeat request (ARQ)”.

  FEC is an error control scheme used in data transmission. This is one of channel coding techniques, and adds redundant information to transmission data so that data lost during transmission can be recovered at the reception destination. The FEC code introduces a controlled amount of redundancy in the transmitted data stream so that the receiver can make an accurate decision about the transmitted sequence.

  FEC is effective when the bandwidth is sufficient, but the effect cannot be expected so much when the bandwidth is insufficient. This is because the use of FEC codes involves the problem of fixed band penalties related to the code rate. Specifically, when the total number of lost packets exceeds the number of redundant packets, data cannot be recovered. Furthermore, since error statistics of the channel are unstable only with the FEC code, transmission without error cannot be guaranteed.

  ARQ is also an error control scheme used in data transmission, and is characterized by realizing highly reliable data transmission by acknowledgment and timeout. In ARQ, when an error is detected, a transmitter retransmits a data block according to a request from a receiver. The receiver can refuse to accept each incoming packet and request a retransmission via the feedback channel. The receiver checks the quality and reliability of the received packet and decides whether to request retransmission. This check is usually performed using an error detection code such as a cyclic redundancy code (CRC). The ARQ scheme is simple but provides a high degree of system reliability. However, the decoder throughput and final output are not constant and drop rapidly as the channel error rate increases.

  ARQ is usually used in applications that require data reliability, such as TCP and wireless MAC. In contrast, FEC is resistant to a certain amount of loss. Therefore, when the FEC code is used, packet loss at the receiver can be arbitrarily reduced if the price of transmitting redundant data is prepared. In a situation where there is too much loss and FEC cannot perform error correction, a hybrid ARQ (HARQ) scheme combined with a normal ARQ technique is used. HARQ is a method of adding an FEC bit to an existing error detection bit. If the FEC fails to correct all bit errors in the data packet, a retransmission is requested.

  Thus, FEC and ARQ have both advantages and disadvantages. When selecting either FEC or ARQ, the packet / frame length is also an important factor. In view of the error characteristics of the wireless channel, it is clear that errors are more likely to occur as the packets become longer. In most cases, it is desirable to avoid frequent retransmissions. Therefore, when solving the problem of which scheme is optimal, examination based on the wireless network environment, packet size, and the like is indispensable.

  As mentioned above, conventional FEC, ARQ, and HARQ either correct a small amount of bit errors or retransmit the entire packet (although most of the received data is assumed to be correct). It is. Therefore, all of FEC, ARQ, and HARQ have a problem that the overhead is large and the flexibility is extremely low, resulting in a waste of network capacity. Some research has already been done on how to design a partial packet recovery scheme.

  Non-Patent Document 1 (K. Jamison and H. Balakrishnan “PPR: Partial Packet Recovery for Wireless Networks” (PPR: Partial Packet Recovery for Wireless Networks), ACM SIGCOMM '07 (Kyoto, Japan, August 2007) In the current wireless mesh network protocol, it is pointed out that a bit of network capacity is wasted because bits already received several times are transmitted in most of packet retransmissions to be executed. In Non-Patent Document 1, an extended physical layer interface (SoftPHY) that provides a higher layer with a physical layer reliability index for each bit to be decoded is used. Non-Patent Document 1 further proposes a link layer ARQ protocol for performing partial packet recovery. The link layer ARQ protocol proposed in Non-Patent Document 1 is called “PP-ARQ”, and the receiver compactly encodes a request for bit retransmission only for a packet that is likely to contain an error. be able to.

K. Jamison and H.C. Balakrishnan "PPR: Partial Packet Recovery for Wireless Networks" (PPR: Partial Packet Recovery for Wireless Networks), ACM SIGCOMM '07 (Kyoto, Japan, August 2007)

The PP-ARQ protocol proposed in Non-Patent Document 1 is constructed based on a conventional communication system in which FEC is frequently used. Therefore, a large amount of overhead is generated in packet transmission, and the throughput is reduced. The PP-ARQ protocol is effective when packet drops due to collisions are serious, but otherwise no performance improvement can be expected.

A first wireless communication system of the present invention includes a transmitter and a receiver, and the transmitter includes transmitter physical layer means for transmitting communication data without performing channel coding. Receiver physical layer means for receiving received data corresponding to communication data transmitted from the layer means and presenting hint information as to whether each bit in the received data is appropriate; and hint information from the receiver physical layer means And the hint information and the received data are passed to the receiver medium access control (MAC) layer means, the interface between the receiver physical layer means and the receiver MAC layer means, and the hint passed from the interface Receiver MAC layer means for analyzing the received data with reference to the information, and the receiver MAC layer means converts the received data into one or more appropriate bit strings and one or more illegal bits based on the hint information. Marking as a sequence of columns, constructing feedback indicating the position of each of the one or more incorrect bit sequences and the checksum of each of the one or more correct bit sequences, the receiver physical layer means encoding the feedback, Transmit the encoded feedback, receive and decode the encoded retransmission data, the transmitter physical layer means receives and decodes the encoded feedback, encodes the retransmission data, and encodes the retransmission Transmitting the data, the transmitter further includes transmitter MAC layer means for determining retransmission data based on the decoded feedback, the transmitter MAC layer means including one or more frauds indicated by the feedback Based on the position of each bit string and the checksum of each of the one or more proper bit strings, one or more errors in the communication data. A portion corresponding to the bit string, in the communication data, determines a retransmission data containing the portion corresponding to the erroneous determination proper bit sequence is properly determined to be the incorrect bits.

The first wireless communication method of the present invention includes a step of transmitting communication data without executing channel coding, a step of receiving received data corresponding to the communication data, and whether each bit in the received data is appropriate. Presenting physical layer hint information about the data, analyzing received data with reference to the physical layer hint information, constructing feedback on the received data based on the analysis result, and encoding the feedback Transmitting encoded feedback, receiving and decoding the encoded feedback, determining retransmission data based on the decoded feedback, encoding and encoding the retransmission data Transmitting the retransmitted data, and receiving and decoding the encoded retransmitted data. And reconstructing the communication data based on the received data and the decoded retransmission data. In the analyzing step, the received data is converted into one or more appropriate bit sequences and 1 based on the physical layer hint information. Marking as a sequence of one or more illegal bit sequences, constructing feedback indicating a position of each of the one or more illegal bit sequences and a checksum of each of the one or more proper bit sequences in a construction step, A portion corresponding to one or more illegal bit sequences in the communication data based on the position of each of the one or more illegal bit sequences indicated by the feedback and the checksum of each of the one or more proper bit sequences; Judgment of retransmission data including a portion corresponding to an erroneously determined appropriate bit string that is an illegal bit string determined to be appropriate in the data That.

The first transmitter of the present invention transmits communication data without performing channel encoding, receives and decodes encoded feedback from the receiver, encodes retransmission data, and performs the encoded retransmission Physical layer means for transmitting data, and medium access control (MAC) layer means for determining retransmission data based on the decoded feedback, wherein the MAC layer means includes one or more frauds indicated by the feedback Based on each position of the bit string and each checksum of one or more appropriate bit strings, a portion corresponding to one or more illegal bit strings in the communication data, and an illegal bit string determined to be appropriate in the communication data Retransmission data including a portion corresponding to a certain erroneous determination appropriate bit string is determined .

The first receiver of the present invention receives the received data corresponding to the communication data transmitted from the transmitter, presents hint information as to whether each bit in the received data is appropriate, and encodes feedback. Transmitting the encoded feedback and receiving and decoding the encoded retransmission data, extracting the hint information from the physical layer means, and combining the hint information and the received data. Analyzing the received data with reference to the interface between the physical layer means and the MAC layer means, which is passed to the MAC layer means, and the hint information passed from the interface, and constructing feedback on the received data based on the analysis result, And MAC layer means for reconstructing communication data based on the received data and the retransmission data decoded by the physical layer means, the receiver M The C layer means marks the received data as a sequence of one or more proper bit strings and one or more illegal bit strings based on the hint information, and positions each of the one or more illegal bit strings and one or more proper bit strings. Construct feedback indicating the checksum of each bit string .

Therefore, according to the present invention, the overhead can be greatly reduced, the amount of transmission data can be reduced, the number of executions of error recovery can be maintained at the same level, and the system throughput can be improved.

The above and other features and advantages of the present invention will be more clearly understood by reading the following detailed description in conjunction with the following drawings, in which: In the drawings, the same or similar elements are denoted by the same or similar reference symbols.
1 is a block diagram of a wireless communication system according to a first embodiment of the present invention. FIG. 6 is a schematic diagram of a process for constructing feedback by a receiver MAC layer according to physical layer hint information. The process of the radio | wireless communication method performed in the radio | wireless communications system shown in FIG. 1 is shown. It is a block diagram of the radio | wireless communications system by the 2nd Example of this invention. The process of the radio | wireless communication method performed in the radio | wireless communications system shown in FIG. 4 is shown. It is a block diagram of the radio | wireless communications system by the 3rd Example of this invention. FIG. 6 is a schematic diagram of threshold rules used when a receiver marks a bit string. 7 shows processing of a wireless communication method executed in the wireless communication system shown in FIG. 2 shows simulation results of the adaptive error recovery scheme of the present invention.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram of a wireless communication system according to a first embodiment of the present invention.

  As shown in FIG. 1, the wireless communication system 1 includes a transmitter 10 and a receiver 20.

  The error recovery scheme used in the wireless communication system 1 is a link layer error recovery scheme that is performed using the medium access control (MAC) layer and the physical layer of the transmitter 10 and the receiver 20.

  The transmitter 10 includes physical layer means 11 and MAC layer means 12. The physical layer means 11 includes an encoding / decoding unit 111 and a transmission / reception unit 112.

  The receiver 20 includes a physical layer means 21, a MAC layer means 22, and a SoftPHY interface 23 disposed between the physical layer means 21 and the MAC layer means 22. The physical layer means 21 includes an encoding / decoding unit 211 and a transmission / reception unit 212.

  The physical layer means 11 and 21 are means for implementing the physical layer functions of the transmitter 10 and the receiver 20 and allow data to be transmitted on various physical media. The physical layer means 11 and 21 provide services such as channel encoding / decoding, modulation / demodulation and the like. In FIG. 1, encoding / decoding sections 111 and 211 for performing channel encoding / decoding, and some physical layer processing (modulation, etc.) are performed on the data, data is transmitted and received, and reception is performed. Only data transmission / reception units 112 and 212 that perform some physical layer processing (demodulation, etc.) on the processed data are shown. However, as will be apparent to those skilled in the art, the physical layer means 11 and 21 can also provide services other than those described above, and therefore can include functional modules for implementing these services. (For simplicity of explanation, these modules are not shown).

  The MAC layer means 12 and 22 are means for implementing a MAC layer function in each of the transmitter 10 and the receiver 20. As will be apparent to those skilled in the art, the MAC layer provides an addressing mechanism and a channel access control mechanism to allow communication between terminals and network nodes in a multipoint network (usually a LAN or MAN).

  In the present invention, the physical layer means 11 of the transmitter 10 transmits the communication data transmitted for the first time without performing channel coding. Specifically, when the upper layer of the transmitter 10 transmits some communication data to the receiver, the MAC layer means 12 receives the communication data transmitted from the upper layer and passes the communication data to the physical layer means 11. The physical layer means 11 transmits it via a physical transmission medium. Unlike the existing FEC or HARQ scheme, the communication data is subjected to channel encoding such as addition of FEC code executed by the encoding / decoding unit 111 for executing channel encoding or decoding at the time of initial transmission. The transmission / reception unit 112 transmits the data directly without adding it.

  “Data” used in the present invention is data in units of bits, symbols, packets, frames, ARQ blocks, and the like unless otherwise specified. For example, as is well known to those skilled in the art, data handled by the physical layer is usually in units of bits, and data handled by the MAC layer is usually in units of packets or frames. In the present invention, these may be collectively referred to as “data”, but those skilled in the art will be able to assign an appropriate data unit depending on the implementation. In the present invention, “packet” and “frame” are used interchangeably.

  The transmission / reception unit 212 of the receiver 20 receives reception data corresponding to the communication data transmitted by the transmitter 10. As described above, the received data received by the receiver 20 due to noise or interference during communication may be different from the communication data that the transmitter 10 tried to transmit to the receiver 20 due to an error or the like. There is sex.

  In the physical layer means 21 of the receiver 20, communication data is not channel-encoded at the time of transmission from the transmitter 10, so that received data received by the transmission / reception unit 212 is decoded by the encoding / decoding unit 211. There is no need to be subject to processing. Therefore, this received data is directly passed from the SoftPHY interface 23 to the MAC layer means 22.

  The SoftPHY interface 23 enables the receiver 20 to determine which bits in the received data should be corrected using the hint information from the physical layer. Thus, receiver 20 need not obtain additional feedback or information from transmitter 10. The SoftPHY interface 23 passes information about how close each received symbol is to the symbol determined by the physical layer (ie, physical layer hint information) to the upper layer. Since the receiver 20 does not perform channel decoding of received data, the SoftPHY interface 23 may be placed immediately after the transmission / reception unit 212. For example, when the transmission / reception unit 212 includes a demodulation module, the SoftPHY interface 23 can directly receive data demodulated by the demodulation module. In this case, the hint information may be a distance in the signal space between the constellation point of the first received signal and the constellation point mapped after demodulation. The longer the distance, the lower the certainty regarding demodulation, and the shorter, the higher the certainty. In this way, the physical layer hint information can serve as a reliability indicator regarding the likelihood that each symbol / bit in the received data is an error.

  The SoftPHY interface 23 passes the received data processed (by demodulation or the like) by the transmission / reception unit 212 and physical layer hint information extracted from the demodulator of the transmission / reception unit 212 to the MAC layer unit 22. In one embodiment of the present invention, the SoftPHY interface 23 passes received data to the MAC layer means 22 in units of packets.

  The MAC layer means 22 refers to the physical layer hint information passed from the SoftPHY interface 23 and analyzes the acquired received data. Specifically, the MAC layer means 22 uses the physical layer hint information passed from the SoftPHY interface 23 to convert the received data (in units of packets or the like) into one or more proper bit sequences or one or more illegal bit sequences. Mark as a sequence of bit strings containing (or both). Each appropriate bit string includes a continuous bit string predicted as appropriate based on the physical layer hint information, and each incorrect bit string includes a continuous bit string predicted as incorrect based on the physical layer hint information. As will be understood by those skilled in the art, the “appropriate bit string” and “illegal bit string” used in the present invention mean a bit string regarded as appropriate or incorrect based on physical layer hint information, but is not necessarily correct or actually It does not have to be an invalid bit string. Further, as apparent from the definitions of the proper bit string and the illegal bit string, the proper bit string and the illegal bit string are alternately generated.

  The MAC layer means 22 then constructs a feedback based on the sequence of marked bit strings. This feedback can include, for example, position information (such as the start position and length of each illegal bit string) regarding the illegal bit string, and a checksum of each appropriate bit string, as will be described later. In one embodiment of the present invention, the MAC layer means 22 includes an error detection code (eg, a cyclic redundancy check code (CRC) added before normal transmission by the physical layer of the transmitter, for the entire received data (eg, all packets). )), If it is determined that the entire received data is proper, the marking of the bit string is ignored, and an acknowledgment message (ACK) indicating that the current received data has been successfully received is transmitted to the transmitter. Note that 10 is fed back.

  As shown in FIG. 2, the determination made by the MAC layer means 22 based on the physical layer hint information may lead to a misjudgment (determination of fraud as appropriate or determination of appropriate as improper). In order to prevent the retransmission bit from being deficient due to an erroneous determination appropriate bit string (that is, the MAC layer means 22 determined to be appropriate based on the physical layer hint information but actually includes an error), each appropriate bit string checksum (Such as CRC) needs to be sent to the transmitter 10 so that the transmitter 10 can check whether the bit string is truly proper. “PHY hint information” in FIG. 2 is an exemplary case in which the distance in the signal space between the arrangement point of the received signal and the arrangement point of the demodulated symbol is used as the physical layer hint information. Therefore, in this example, as described above, the shorter the distance, the higher the reliability, and the longer the distance, the lower the reliability.

  The MAC layer unit 22 passes the constructed feedback data to the encoding / decoding unit 211 of the physical layer unit 21. Subsequently, the encoding / decoding unit 211 performs channel coding on the feedback data to be transmitted, and passes the channel-coded feedback to the transmission / reception unit 212. The transmission / reception unit 212 performs some processing (such as modulation) on the encoded feedback and transmits it to the transmitter 10.

  Thereafter, the transmission / reception unit 112 of the transmitter 10 receives the feedback transmitted from the receiver 20, performs processing such as demodulation, and then passes it to the encoding / decoding unit 111 for decoding. . The encoding / decoding unit 111 passes the decoded feedback to the MAC layer unit 12 which is an upper layer.

  The MAC layer means 12 checks the checksum of each proper bit sequence included in the decoded feedback, calculates the checksum for the portion corresponding to each proper bit sequence in the communication data, and makes a pair between the two checksum sets. 1. Compare 1 If the checksum of each proper bit sequence included in the feedback matches the checksum of the portion corresponding to the proper bit sequence in the communication data, those proper bit sequences determined to be proper by the receiver 20 are truly proper. . However, if the checksum of each proper bit sequence included in the feedback does not match the checksum of the portion corresponding to the proper bit sequence in the communication data, the proper bit sequence determined to be proper by the receiver 20 includes an error. That is, a proper bit string erroneously determined to be proper exists in the communication data.

  Subsequently, the MAC layer means 12 constructs retransmission data. The retransmission data includes a portion corresponding to the bit string reported as illegal by the receiver 20 in the original communication data and a portion corresponding to the erroneous determination appropriate bit string detected in the above in the original communication data. It is done.

  The receiver 20 can assign a sequence number to each appropriate bit string in the feedback data, for example, as a method for identifying an actually invalid bit string from among the appropriate bit strings indicated by the feedback. Therefore, the retransmission data constructed by the transmitter 10 can include the sequence number of the misjudgment proper bit string in addition to the illegal bit string and the misjudgment proper bit string.

  The MAC layer means 12 passes the constructed retransmission data to the encoding / decoding unit 111. The encoding / decoding unit 111 encodes the retransmission data and passes the encoded retransmission data to the transmission / reception unit 112. The transmission / reception unit 112 performs some processing (such as modulation) on the encoded retransmission data and transmits it to the receiver 20.

  Thereafter, the transmission / reception unit 212 of the receiver 20 receives the retransmission data transmitted from the transmitter, executes a process such as demodulation on the data, and then encodes / decodes the data for decoding. 211. The encoding / decoding unit 211 passes the decoded retransmission data to the upper layer MAC layer means 22.

  The MAC layer means 22 includes communication data (ie, an error that the transmitter 10 was intended to transmit to the receiver 20) based on the decoded retransmission data and the first received data. No communication data).

  The error recovery scheme (executed by the wireless communication system 1) used in the above wireless communication can be referred to as “modified partial packet recovery—automatic repeat request” (modified PP-ARQ).

  As described above, most current systems require channel coding as a countermeasure against bit errors. Increased overhead due to redundancy is a major drawback of channel coding. This redundancy cannot be reduced arbitrarily even with adaptive modulation and coding. On the other hand, according to the modified PP-ARQ of the present invention, channel coding is not executed at the first transmission of communication data. On the receiver side, if it is determined that there is an error bit using the hint information presented by the physical layer, channel-coded feedback is provided to the transmitter. The transmitter constructs retransmission data based on this feedback, and transmits the channel-coded retransmission data to the receiver. As a result, the receiver can construct appropriate communication data based on the retransmission data and the first received data.

  In the physical layer, the first transmitted data is not subjected to channel coding (that is, redundant FEC information is not added). Thus, overhead is greatly reduced and the amount of data transmitted is also reduced. Error recovery is similarly performed by retransmitting only those bits that are presumed to contain errors. Since the amount of feedback and retransmission data is generally small, these data are channel coded. This reduces the situation in which retransmission is performed multiple times.

  Therefore, according to the present invention, overhead is reduced, the quality of error recovery is maintained at the same level, and the throughput of the system is improved.

  Below, the process of the radio | wireless communication method performed by the radio | wireless communications system 1 shown in FIG. 1 is demonstrated, referring FIG.

  As shown in FIG. 3, in step S101, the physical layer means 11 of the transmitter 10 transmits communication data without channel coding.

  Subsequently, in step S <b> 102, the transmission / reception unit 212 of the receiver 20 receives reception data corresponding to the communication data transmitted by the transmitter 10.

  Thereafter, in step S103, the SoftPHY interface 23 receives the received data processed by the transmission / reception unit 212 (by demodulation or the like) for the MAC layer means 22 (for example, in units of packets) and the physical data extracted from the physical layer means 21. Pass layer hint information.

  Further, in step S104, the MAC layer means 22 analyzes the received data with reference to the physical layer hint information passed from the SoftPHY interface 23. Specifically, the MAC layer means 22 marks the received data obtained by using the physical layer hint information as a sequence of bit strings including one or more proper bit strings or one or more illegal bit strings (or both). .

  Next, in step S105, the MAC layer means 22 constructs feedback based on the sequence of marked bit strings.

  In step S <b> 106, the encoding / decoding unit 211 performs channel encoding on the feedback data transferred from the MAC layer unit 22, and passes the channel encoded feedback to the transmission / reception unit 212. The transmission / reception unit 212 performs some processing (such as modulation) on the encoded feedback and passes it to the transmitter 10.

  Thereafter, in step S107, the transmission / reception unit 112 of the transmitter 10 receives the feedback transmitted from the receiver 20, performs some processing such as demodulation on the feedback, and then encodes it for decoding. To the encoding / decoding unit 111. The encoding / decoding unit 111 passes the decoded feedback to the upper layer MAC layer means 12.

  Next, in step S108, the MAC layer means 12 analyzes the feedback. Specifically, the MAC layer means 12 performs a test for determining the presence / absence of a misjudgment appropriate bit string based on the decoded feedback.

  Subsequently, the MAC layer means 12 constructs retransmission data in step S109. As described above, this retransmission data corresponds to the portion of the original communication data corresponding to the bit string reported as illegal by the receiver 20 and the misjudgment appropriate bit string detected above in the original communication data. Part to be included. The retransmission data may further include an index (for example, a sequence number) of an appropriate bit string for erroneous determination.

  Then, the MAC layer means 12 passes the constructed retransmission data to the encoding / decoding unit 111 in step S110. The encoding / decoding unit 111 encodes the retransmission data and passes the encoded retransmission data to the transmission / reception unit 112. The transmission / reception unit 112 performs some processing (such as modulation) on the encoded retransmission data and transmits it to the receiver 20.

  In step S111, the transmission / reception unit 212 of the receiver 20 receives the retransmission data transmitted from the transmitter, performs a process such as demodulation on the data, and then encodes / decodes the data for decoding. The data is passed to the decryption unit 211. The encoding / decoding unit 211 passes the decoded retransmission data to the upper layer MAC layer means 22.

  In step S112, the MAC layer means 22 reconstructs communication data based on the decoded retransmission data and the first received data. This ends the process.

  Next, a radio communication system 1a according to a second embodiment of the present invention will be described with reference to FIG. Of the constituent elements of the wireless communication system 1a, the same reference numerals are assigned to the same constituent elements as those of the wireless communication system 1 shown in FIG.

  The difference between the wireless communication system 1a according to the second embodiment and the wireless communication system 1 according to the first embodiment is that the transmitter 10a of the wireless communication system 1a includes a MAC layer means 12a different from the MAC layer means 12. That is.

  The MAC layer means 12a differs from the MAC layer means 12 in that it further includes an adaptive decision function. That is, the MAC layer means 12a implements the modified PP-ARQ according to the present invention, based on the current channel state information, whether the physical layer means 11 transmits communication data without executing channel coding. Determine whether or not.

  As this current channel state information, for example, the signal-to-noise ratio (SNR) value of the channel reported by the receiver can be used. For example, in WiMAX, the SNR is included in the channel quality indicator (CQI) by the mobile station and reported via the feedback channel. Those skilled in the art will appreciate that current channel state information is not limited to SNR values obtained in the specific manner described above.

  The MAC layer means 12a instructs the physical layer means 11 to transmit the communication data without performing channel coding when the SNR value exceeds the threshold value, thereby modifying the modified PP according to the present invention. -Implement ARQ.

  Conversely, if the SNR value is less than or equal to the threshold, the MAC layer means 12a performs conventional FEC (ie, performs channel coding to transmit communication data) or a combination of FEC and ARQ ( That is, it is determined to execute HARQ).

  Such a threshold for SNR can be determined based on simulations and field experimental results.

  In the wireless communication system 1a according to the second embodiment, the MAC layer means 12a of the transmitter 10a makes an adaptive decision based on the channel state. When the channel condition is good, it is decided to use the modified PP-ARQ scheme according to the present invention, and when the channel condition is not good, it is decided to use the conventional FEC or HARQ scheme. The Such an adaptive error recovery scheme, also called “smart ARQ”, enables the wireless communication system 1a to find the best trade-off between overhead reduction and error recovery guarantees in various channel conditions. To do.

  Below, the process of the radio | wireless communication method performed by the radio | wireless communications system 1a shown in FIG. 4 is demonstrated, referring FIG.

  As shown in FIG. 5, the MAC layer means 12a determines the current channel state in step S201. When the SNR is used as an indicator of the channel state, the MAC layer means 12a decides to adopt the modified PP-ARQ if the SNR exceeds a predetermined threshold value, and sends the channel code to the physical layer means 11. Instructs to send communication data without conversion. Subsequent steps S202 to S213 are the same as steps S101 to S112 in FIG.

  If the SNR is below the threshold, the MAC layer means 12a decides to adopt a conventional FEC or HARQ scheme. FIG. 5 shows a process when HARQ is used.

  As shown in FIG. 5, if the system uses a conventional scheme, the physical layer means 11 performs channel coding and other necessary processing on the communication data to be transmitted in step S214, Transmit to the receiver 20.

  In step S215, the physical layer means 21 of the receiver 20 performs channel decoding and other necessary processing on the received data, and passes it to the MAC layer means 22.

  In step S216, the MAC layer means 22 performs error detection and error correction on the received data based on the error detection code (CRC), error correction code (FEC), etc. included in the received data. If the error correction fails, the system performs a conventional HARQ procedure. In FIG. 5, some steps of this procedure are not shown in detail. How to implement these steps is a fact known to those skilled in the art.

  Next, a radio communication system 1b according to a third embodiment of the present invention will be described with reference to FIG. Among the components of the wireless communication system 1b, the same components as those of the wireless communication system 1a shown in FIG.

  The difference between the wireless communication system 1b according to the third embodiment and the wireless communication system 1a according to the second embodiment is that the receiver 20b of the wireless communication system 1b includes a MAC layer means 22b different from the MAC layer means 22. That is.

  The MAC layer means 22b uses threshold rules when marking the received data as a sequence of proper bit sequences and illegal bit sequences. Specifically, when the length of one appropriate bit string sandwiched between two illegal bit strings is sufficiently short (for example, shorter than the threshold value G), it is better to consider that the appropriate bit string is also an illegal bit string. And retransmission can be performed more efficiently. In other words, as shown in FIG. 7, the MAC layer means 22b combines two illegal bit strings and one appropriate bit string shorter than the threshold value between them, and marks them as a single illegal bit string. .

  The threshold value is determined by the following method.

λ_ｇのしきい値Ｇは以下の式によって得られる。

ここで

は天井関数、
Ｓは適正ビット列または不正ビット列の可能な最大の開始位置、
Ｂは適正ビット列または不正ビット列の可能な最大長、
λ_ｇは現在の適正ビット列の長さ、
λ_ｃはフィードバックに含まれる各適正ビット列のチェックサムの長さである。 If you want to

Threshold G of lambda _g is obtained by the following equation.

here

Is the ceiling function,
S is the maximum possible start position of a proper bit string or illegal bit string,
B is the maximum possible length of a proper bit string or illegal bit string,
λ _g is the length of the current proper bit string,
λ _c is a checksum length of each appropriate bit string included in the feedback.

  S and B are usually the same length as the total length of received data currently handled by the MAC layer of the receiver. That is, S corresponds to a situation in which a proper bit string or illegal bit string is at the end of the received data, and B corresponds to a situation in which the proper bit string or illegal bit string starts from the first bit of the received data and continues to the last bit. For example, the SoftPHY interface 23 passes the bits of received data supplied from the physical layer means 21 to the MAC layer in units of N-bit packets, so that the MAC layer handles the received data in units of packets, and each N-bit packet Is marked as a proper bit string or illegal bit string, each of the above S and B is considered to be equal to the packet length (ie, N).

  The complexity of this algorithm is O (L), where L is the number of incorrect bit strings before the threshold rule is applied. As described above, the threshold value G for the length of the appropriate bit string does not require complicated calculation and can be determined by a simple method, so that the efficiency when the receiver constructs the feedback further increases.

  As described above, in the wireless communication system 1b according to the third embodiment, the MAC layer means 22b of the receiver 20b marks the appropriate bit sequence and the illegal bit sequence using the threshold rule, thereby reducing calculation complexity. This improves the efficiency of feedback construction.

  Next, processing of the wireless communication method executed by the wireless communication system 1b shown in FIG. 6 will be described with reference to FIG.

  Steps S301 to S304, S306 to S313, and S314 to S316 shown in FIG. 8 are substantially the same as those in FIG.

  In step S305, the MAC layer means 22b of the receiver 20 marks the received data as a sequence of a proper bit string and an illegal bit string, and one shorter one than the threshold G sandwiched between the two illegal bit strings. The threshold rule is used to mark the proper bit string as a single illegal bit string.

The foregoing has described several exemplary embodiments of the invention. Although the error recovery scheme of the present invention is a system-dependent scheme, when the smart ARQ scheme of the present invention (that is, the adaptive error recovery scheme) is performed on a link level WiMAX set set, the simulation result of FIG. Obtained. FIG. 9 compares the throughput of the wireless communication system of the present invention with the performance of FEC schemes with different coding rates. From this comparison, it is clear that the performance of the scheme of the present invention exceeds that of the scheme used in the WiMAX standard. Table 1 below shows the OFDM parameters used in the simulation.

  While the invention has been described above with reference to a few specific embodiments thereof, it will be understood that the invention is susceptible to various modifications, combinations, and alterations, and these modifications, combinations, and alterations are claimed by the appended claims. It will be understood by those skilled in the art that the present invention is included in the scope of the present invention.

11: Physical layer 12: MAC layer 111: Encoding or decoding (feedback, retransmission data only)
112: Transmission / reception (modulation / demodulation, etc.)
21: Physical layer 211: Encoding or decoding (feedback, retransmission data only)
212: Transmission / reception (modulation / demodulation, etc.)
22: MAC layer 23: SoftPHY interface 12a: MAC layer (adaptive decision)
22b: MAC layer (threshold rule)

Claims (10)

Equipped with transmitter and receiver,
The transmitter is
Transmitter physical layer means for transmitting communication data without channel coding;
The receiver
Receiver physical layer means for receiving received data corresponding to communication data transmitted from the transmitter physical layer means and presenting hint information as to whether or not each bit in the received data is appropriate;
Extracting hint information from the receiver physical layer means, and passing the hint information and received data to a receiver medium access control (MAC) layer means, the receiver physical layer means and the receiver MAC layer means Between the interface and
Based on the hint information passed from the interface , the received data is marked as a sequence of one or more proper bit sequences and one or more illegal bit sequences, each position of the one or more illegal bit sequences, and one or more Receiver MAC layer means for constructing feedback indicating a checksum of each of the proper bitstreams ;
Before Receiver physical layer unit, a feedback encodes and transmits the feedback obtained by encoding, receives and decodes the retransmitted data encoded,
The transmitter physical layer means receives and decodes encoded feedback, encodes retransmission data, transmits the encoded retransmission data;
The transmitter further includes transmitter MAC layer means for determining retransmission data based on the decoded feedback;
The transmitter MAC layer means may determine one or more frauds in the communication data based on the position of each of the one or more malformed bitstreams and the checksum of each of the one or more proper bitstreams indicated by feedback. Determining retransmission data including a portion corresponding to a bit string and a portion corresponding to a misjudgment appropriate bit string that is an illegal bit string determined to be appropriate in communication data;
The receiver MAC layer means marks two illegal bit sequences and one proper bit sequence sandwiched between them and shorter than a threshold as a single illegal bit sequence,
The threshold is calculated by the following formula:

(here,
G is a threshold value,

Is the ceiling function,
S is the maximum possible start position of a proper bit string or illegal bit string,
B is the maximum possible length of a proper bit string or illegal bit string,
λc is the checksum length)
A wireless communication system.   2. The transmitter MAC layer means instructs the transmitter physical layer means whether or not to transmit communication data without executing channel coding based on channel state information. The wireless communication system according to 1.   The channel state information includes a signal to noise ratio, and the transmitter MAC layer means does not perform channel coding on the transmitter physical layer means when the signal to noise ratio exceeds a threshold value. The wireless communication system according to claim 2, wherein the wireless communication system instructs to transmit communication data. The wireless communication system according to any one of claims 1 to 3, wherein S and B are equal to the total number of bits included in the received data . Transmitting communication data without performing channel encoding;
Receiving received data corresponding to the communication data;
Presenting physical layer hint information about whether each bit in the received data is appropriate;
Analyzing the received data with reference to physical layer hint information;
Building feedback on the received data based on the analysis results;
Encoding the feedback and transmitting the encoded feedback;
Receiving and decoding the encoded feedback; and
Determining retransmission data based on the decoded feedback;
Encoding the retransmission data and transmitting the encoded retransmission data;
Receiving and decoding the encoded retransmission data; and
Reconstructing the communication data based on the received data and the decoded retransmission data,
In the analysis step, based on the physical layer hint information, the received data is marked as a sequence of one or more proper bit sequences and one or more illegal bit sequences,
Constructing feedback indicating the position of each of the one or more incorrect bit sequences and the checksum of each of the one or more proper bit sequences in the construction step;
Corresponding to one or more illegal bit sequences in the communication data based on the position of each of the one or more illegal bit sequences and the checksum of each of the one or more proper bit sequences indicated by the feedback in the determining step And retransmission data including a portion corresponding to a misjudgment appropriate bit sequence that is an illegal bit sequence determined to be appropriate in communication data,
In the marking step, two illegal bit strings and one appropriate bit string shorter than the threshold value sandwiched between them are marked as a single illegal bit string,
The threshold is calculated by the following formula:

(here,
G is a threshold value,

Is the ceiling function,
S is the maximum possible start position of a proper bit string or illegal bit string,
B is the maximum possible length of a proper bit string or illegal bit string,
λc is the checksum length)
A wireless communication method . 6. The wireless communication method according to claim 5, further comprising the step of determining whether communication data is transmitted without performing channel coding based on channel state information . Deciding to transmit communication data without performing channel coding if the channel state information includes a signal-to-noise ratio and the determining step has a signal-to-noise ratio exceeding a threshold value; The wireless communication method according to claim 6, wherein: The wireless communication method according to any one of claims 5 to 7, wherein S and B are equal to a total number of bits included in the received data . Receives received data corresponding to the communication data transmitted from the transmitter, presents hint information about whether each bit in the received data is appropriate, encodes feedback, and transmits the encoded feedback And physical layer means for receiving and decoding the encoded retransmission data;
An interface between the physical layer means and the MAC layer means that extracts hint information from the physical layer means and passes the hint information and the received data to the MAC layer means;
Based on the hint information passed from the interface, the received data is marked as a sequence of one or more proper bit sequences and one or more illegal bit sequences, each position of the one or more illegal bit sequences, and one or more MAC layer means for constructing feedback indicating each checksum of the proper bitstream,
The MAC layer means marks two illegal bit strings and one appropriate bit string shorter than the threshold value sandwiched between them as a single illegal bit string,
The threshold is calculated by the following formula:

(here,
G is a threshold value,

Is the ceiling function,
S is the maximum possible start position of a proper bit string or illegal bit string,
B is the maximum possible length of a proper bit string or illegal bit string,
λc is the checksum length)
A receiver characterized by that . The receiver according to claim 9, wherein S and B are equal to the total number of bits included in the received data .

Images