WO2010041295A1

WO2010041295A1 - Radio relay device and reproduction method in relay station

Info

Publication number: WO2010041295A1
Application number: PCT/JP2008/002844
Authority: WO
Inventors: 大渕一央; 河▲崎▼義博; 田島喜晴; 太田好明; 田中良紀
Original assignee: 富士通株式会社
Priority date: 2008-10-08
Filing date: 2008-10-08
Publication date: 2010-04-15

Abstract

A relay station (2) transfers to a reception station (3), a packet transmitted from a transmission station (1). The relay station (2) stores in a buffer, a reproduction packet obtained by demodulating and decoding the received packet data. The relay station (2) measures a quality of the received packet. If the received packet has a quality lower than a threshold value level, the relay station (2) transmits the reproduction packet to a reception station (3) even if a NACK message is not received from the reception station (3).

Description

Radio relay apparatus and retransmission method in relay station

The present invention relates to an apparatus for relaying a radio signal and a retransmission method in a relay station. The present invention can also be applied to, for example, a wireless packet relay system.

A wireless packet relay system that transfers a packet transmitted from a transmitting station to a receiving station via a relay station is known. In the example shown in FIG. 1, the packet a transmitted from the base station (transmitting station) 1 is transferred to the terminal device (receiving station) 3 a via the relay station 2. Thus, by providing the relay station 2, the base station 1 can transmit a packet to the terminal device 3a located outside the cell range.

Since the terminal device 3b is located in the cell range, the terminal device 3b receives the direct wave transmitted from the base station 1 and the relay wave relayed by the relay station 2. In this case, the terminal device 3b can improve the reception quality of the packet b by performing diversity reception of the direct wave signal and the relay wave signal.

As the relay method of the wireless packet relay system, the AF (Amplitude and Forward) method and the DF (Decoded and Forward) method have been proposed. In the AF method, the relay station amplifies the signal received from the transmission station as necessary and transfers the signal to the reception station without demodulating the signal transmitted from the transmission station. Therefore, the AF method has a small delay at the relay station, but the noise component is also amplified, so that the reception quality may be deteriorated. On the other hand, in the DF scheme, the relay station demodulates / decodes the signal transmitted from the transmitting station and then transfers it to the receiving station. Therefore, the reception quality is improved in the DF method, but the delay in the relay station becomes large. Furthermore, the DF method requires more radio resources than the AF method.

A method for solving this problem is proposed in Non-Patent Document 1. A time chart of the method proposed in Non-Patent Document 1 is shown in FIG. In this method, at the time of the first packet transmission, the relay station transfers the packet to the destination station by the AF method (non-regenerative relay). At this time, the relay station demodulates the radio packet and stores it in the memory. When the destination station detects an error, it returns a NACK message. Then, the relay station reproduces the packet from the data stored in the memory and transmits it to the destination station. That is, the relay station retransmits the packet by the DF method (regenerative relay).

Thus, in the proposed method, the packet is first transferred by the AF method. For this reason, when there is no error, the packet is transferred without delay. Further, when an error occurs, high-quality communication is realized by retransmitting the packet by the DF method.

However, in the proposed method, since retransmission control is started in response to a NACK message, the delay time during packet retransmission is large. That is, a predetermined time is required for error detection processing at the destination station. Also, it takes a predetermined time for the relay station to analyze and recognize the NACK message. Therefore, when the communication environment is not good, retransmission procedures are frequently executed, and the throughput is greatly reduced. This problem does not only occur in packet communication, but can occur widely in wireless relay systems.

Related techniques are described in

Patent Documents

1 and 2, for example.
A Study on Wireless Packet Transfer Method in Cooperative Transmission, 2007 IEICE Communication Society Conference, 445 pages, B-5-123 Japanese Patent Laid-Open No. 2001-156692 JP 2005-143006 A

In one aspect, an object of the present invention is to improve a delay time related to retransmission.
In the first plan, retransmission is performed between the relay station and the receiving station. The relay station receives a radio signal from the transmission station and transmits a radio signal to the reception station based on the received radio signal, thereby performing relay processing of transmission data from the transmission station to the reception station. The disclosed retransmission method performs a first transfer process to the receiving station for the wireless signal without demodulating the wireless signal received from the transmitting station, and based on the quality of the received wireless signal, A second transfer process is executed based on data obtained by demodulating or error correcting decoding the radio signal. The receiving station performs reception processing of the radio signal transmitted by the first transfer process and the radio signal transmitted by the second transfer process.

In the second plan, the relay station executes the second transfer process based on the quality of the radio signal received from the transmitting station. That is, the relay station can start retransmission control without waiting for a response from the receiving station. Thereby, the delay time related to retransmission in the wireless relay system is improved.

In the third plan, the radio relay device includes a transfer unit that transfers a radio signal transmitted from the transmission station to the reception station without demodulating, a measurement unit that measures reception quality of the radio signal, and the measurement unit. A retransmission unit that transmits reproduction data obtained by demodulating and decoding the radio signal to the receiving station when the measured reception quality is lower than a threshold level;

It is a figure which shows the structure of a wireless packet relay system. 6 is a time chart of the method proposed in Non-Patent Document 1. It is a figure which shows the structure of a transmission station. It is a figure which shows the structure of a relay station. It is a figure which shows the structure of a receiving station. It is a figure explaining a wireless packet relay method (case 1). It is a figure explaining a wireless packet relay method (case 2). It is a figure explaining a wireless packet relay method (case 3). It is a figure explaining a wireless packet relay method (case 4). It is a figure which compares the delay time of AF system, DF system, and the system of embodiment. It is a flowchart which shows operation | movement of a transmission station. It is a flowchart (the 1) which shows operation | movement of a relay station. It is a flowchart (the 2) which shows operation | movement of a relay station. It is a flowchart which shows operation | movement of a receiving station. It is a figure explaining the structure and operation | movement of an error correction encoding part. It is a figure explaining partial resending.

FIG. 3 is a diagram illustrating a configuration of the transmission station according to the embodiment. In FIG. 3, a function for transmitting and receiving packet data is depicted. The transmitting station 10 corresponds to the base station 1 in the example shown in FIG. However, in FIG. 1, the

terminal devices

3a and 3b can also operate as transmitting stations.

The error detection encoding unit 11 adds an error detection code to the given packet data. The packet data includes, for example, header data and payload data, and is transmitted by a wireless packet. The error detection code is not particularly limited. For example, a CRC (Cyclic Redundancy Check) code can be used. The error correction encoding unit 12 adds an error correction code to the packet data output from the error detection encoding unit 11. The error correction code is not particularly limited, but for example, a turbo code can be used. The error correction code may be, for example, an FEC, BCH code, Huffman code, Hamming code, or the like. The buffer 13 stores the packet data encoded by the error detection encoding unit 11 and the error correction encoding unit 12. The modulation unit 14 generates modulation data based on the packet data stored in the buffer 13. The wireless transmission unit 15 transmits modulated data. That is, a radio signal for transmitting packet data is transmitted.

The radio receiving unit 16 receives a radio signal from the relay station 20. The demodulator 17 demodulates the received radio signal. The ACK / NACK determination unit 18 analyzes data obtained by demodulation and detects an acknowledgment (ACK) message or a negative acknowledgment (NACK: Negative ACK) message. When an ACK message is detected, the packet data stored in the buffer 13 is discarded. On the other hand, when the NACK message is detected, the packet data stored in the buffer 13 is retransmitted.

FIG. 4 is a diagram illustrating a configuration of the relay station according to the embodiment. Note that the relay station 20 corresponds to the base station 2 in the example shown in FIG.
The wireless reception unit 21 receives a wireless signal from the transmission station 10. The demodulator 22 demodulates the radio signal received from the transmission station 10. The synthesizer 23 synthesizes the packet data obtained by the demodulator 22 and the packet data stored in the buffer 24. The buffer 24 stores the combined packet data obtained by the combining unit 23. Note that the buffer 24 is cleared at the first transmission. Therefore, the combining unit 23 performs a packet data combining operation when a packet is retransmitted.

The error correction decoding unit 25 performs error correction on the packet data output from the synthesis unit 23. Thereby, when there is an error, the error is corrected. However, the number of correctable errors is determined by the coding rate and the like. In other words, when there are many errors, it may not be possible to correct all the errors. The error detection unit 26 detects an error in the packet data output from the error correction decoding unit 25. That is, the error correction decoding unit 25 detects the presence or absence of an error that could not be corrected.

The ACK / NACK generation unit 27 generates an ACK message if no error is detected by the error detection unit 26, and generates an NACK message when an error is detected. The modulation unit 28 generates a modulated signal based on the generated ACK message data or NACK message data. The wireless transmission unit 29 transmits the generated modulated signal (ACK message or NACK message) to the transmission station 10.

The signal received by the wireless receiver 21 is input to the demodulator 22 and guided to the wireless transmitter 35 without being demodulated. At this time, the wireless transmission unit 35 amplifies the signal and transmits it to the receiving station 40.

The quality measuring unit 30 measures the quality of the radio signal transmitted from the transmitting station 10. At this time, the quality measurement unit 30 is not particularly limited, but detects, for example, the reception power of a radio signal, a signal-to-interference ratio (SIR), and the number of error corrections. The received power and SIR are calculated using, for example, digital data representing the received signal. The number of error corrections is the number of errors corrected by the error correction decoding unit 25. Furthermore, when turbo decoding is used in the wireless packet relay system, the quality measurement unit 30 may detect the number of repetitions in turbo decoding. Then, the quality measurement unit 30 compares the measured quality level with a predetermined threshold level.

The functions of the error detection encoding unit 31, the error detection encoding unit 32, the buffer 33, and the modulation unit 34 are basically the error detection encoding unit 11, the error detection encoding unit 12, and the buffer 13 provided in the transmitting station 10. This is the same as the modulation unit 14. However, the packet data output from the error correction decoding unit 25 is input to the error detection encoding unit 31. The buffer 33 stores reproduction packet data. In this embodiment, the reproduction packet data is obtained by demodulating and decoding (including error correction) a radio signal transmitted from the transmission station 10 and executing encoding again.

When the quality measured by the quality measuring unit 30 is worse than the threshold level, the packet data stored in the buffer 33 is transmitted to the receiving station 40. That is, retransmission control is executed. Here, since the packet data stored in the buffer 33 is subjected to error correction processing, the reliability is high. That is, highly reliable packet data is transmitted at the time of retransmission.

The functions of the radio reception unit 36, the demodulation unit 37, and the ACK / NACK determination unit 38 are basically the same as those of the radio reception unit 16, the demodulation unit 17, and the ACK / NACK determination unit 18 included in the transmission station 10. However, the wireless receiving unit 36 receives a wireless signal from the receiving station 30. When a NACK message from the receiving station 40 is detected, the packet data stored in the buffer 33 is transmitted to the receiving station 40.

FIG. 5 is a diagram illustrating a configuration of the receiving station according to the embodiment. In FIG. 5, a function for transmitting and receiving packet data is depicted. The receiving station 40 corresponds to the

terminal devices

3a and 3b in the example illustrated in FIG. However, in FIG. 1, the base station 1 can also operate as a receiving station.

The functions of the radio reception unit 41, demodulation unit 42, synthesis unit 43, buffer 44, error correction decoding unit 45, error detection unit 46, ACK / NACK generation unit 47, modulation unit 48, and radio transmission unit 49 are basically the same. Includes a radio reception unit 21, a demodulation unit 22, a synthesis unit 23, a buffer 24, an error correction decoding unit 25, an error detection unit 26, an ACK / NACK generation unit 27, a modulation unit 28, and a radio transmission unit 29 included in the relay station 20. Is the same. That is, the receiving station 40 regenerates the packet data by demodulating the radio signal transmitted from the relay station 20 and performing error correction. Further, an ACK / NACK message is generated according to the presence / absence of an error and returned to the relay station 20.

Next, the wireless packet relay method according to the embodiment will be described with reference to FIGS. Here, the base station 1 shown in FIG. 1 is a transmitting station, and the terminal device 3 (3a or 3b) is a receiving station. Then, the base station 1 transmits packet data # 0 to the terminal device 3.

<Case 1>
In Case 1, the communication environment between the base station 1 and the relay station 2 and between the relay station 2 and the terminal device 3 is good. Hereinafter, the sequence in case 1 will be described with reference to FIG.

The packet transmitted from the base station 1 is received by the relay station 2. The receiving station 2 transfers the packet transmitted from the base station 1 to the terminal device 3 by the AF method. That is, the radio signal transmitted from the base station 1 is simply amplified and transferred to the terminal device 3 without being demodulated / decoded in the relay station 2.

The terminal device 3 receives a radio signal from the relay station 2. The terminal device 3 demodulates the received radio signal, performs error correction, and detects an error. These processes are executed by the demodulator 42, the error correction decoder 45, and the error detector 46. Here, in case 1, no error is detected. “No error is detected” includes a state in which all errors are corrected using an error correction code. In this case, the terminal device 3 generates an ACK message and returns it to the relay station 2. This ACK message is generated by the ACK / NACK generator 47. When the relay station 2 receives this ACK message, the communication between the relay station 2 and the terminal device 3 ends.

The relay station 2 transfers the signal received from the base station 1 to the terminal device 3, demodulates the received signal, executes error correction, and detects an error. Thereby, packet data is reproduced. These processes are executed by the demodulator 22, the error correction decoder 25, and the error detector 26. Here, in case 1, no error is detected. Then, the reproduced packet data is encoded again by the error detection encoding unit 31 and the error correction encoding unit 32 and stored in the buffer 33. Further, the relay station 2 generates an ACK message and returns it to the base station 1. The ACK message is generated by the ACK / NACK generator 27. When the base station 1 receives this ACK message, the communication between the base station 1 and the relay station 2 ends.

Note that the relay station 2 measures the quality of the radio signal received from the base station 1. The quality of the radio signal is measured by the quality measuring unit 30. Here, in case 1, the quality of the radio signal is better than the threshold level. In this case, the relay station 2 does not execute retransmission control based on reception quality.

In addition, when the terminal device 3 is located within the cell range of the base station 1, the terminal device 3 receives radio signals from both the base station 1 and the relay station 2. At this time, the signal arriving at the terminal device 3 via the relay station 2 is delayed with respect to the signal directly propagated from the base station 1 to the terminal device 3. However, the AF method has a small delay. Therefore, for example, in a communication system using OFDM, if this delay is smaller than the guard interval, the terminal device 3 can demodulate the signal by diversity reception (or RAKE combining). In this case, improvement in reception quality is expected.

A packet transmitted to the terminal device 3 is stored in the buffer 13 of the base station 1. The buffer 13 is cleared when the ACK / NACK determination unit 18 detects an ACK message. On the other hand, a reproduction packet is stored in the buffer 33 of the relay station 2. The buffer 33 is cleared when the ACK / NACK determination unit 38 detects an ACK message.

In this embodiment, it is assumed that the time required to process a packet received at the relay station 2 and the time required to process a packet received at the terminal device 3 are the same in this embodiment. This time is 3 time slots in the example shown in FIG.

As described above, in the wireless packet relay method of the embodiment, the first transmitted packet is relayed by the AF method. Therefore, virtually no delay occurs.

<Case 2>
In Case 2, the communication environment between the base station 1 and the relay station 2 is good, but the communication environment between the relay station 2 and the terminal device 3 is not good. Hereinafter, the sequence in case 2 will be described with reference to FIG.

The packet transmitted from the base station 1 is transferred to the terminal device 3 by the relay station 2 as in the case 1. At this time, the relay station 2 transfers the packet by the AF method. Even in case 2, the relay station 2 does not detect an error in the received packet. Therefore, the reproduction packet is stored in the buffer 33 and an ACK message is returned to the base station 1. Also, the quality of the radio signal is better than the threshold level. Therefore, the relay station 2 does not execute retransmission control based on reception quality.

The terminal device 3 receives a radio signal from the relay station 2. However, in case 2, since the communication environment between the relay station 2 and the terminal device 3 is poor, an error in the radio signal is detected. In this example, “error is detected” means that all errors cannot be corrected even if an error correction code is used. In this case, the terminal device 3 generates a NACK message and returns it to the relay station 2. This NACK message is generated by the ACK / NACK generation unit 47.

When the relay station 2 receives the NACK message from the terminal device 3, the relay station 2 transmits the reproduction packet stored in the buffer 33 to the terminal device 3. That is, the packet is retransmitted from the relay station 2 to the terminal device 3 by the DF method. At this time, it takes a three-slot period from the terminal device 3 receiving the packet until the NACK message is returned, and a three-slot time from the relay station 3 receiving the NACK message to retransmitting the packet. I need it. That is, in

case

2, 6 slot periods are required for retransmission processing.

Note that the terminal device 3 stores the first received packet in the buffer 44. The buffer 44 is cleared if no error is detected in the received packet. However, in case 2, an error in the received packet is detected. Therefore, the terminal device 3 returns a NACK message and waits for a retransmission packet without clearing the buffer 44. When the packet is retransmitted from the relay station 2 in response to the NACK message, the combining unit 43 combines the packet data stored in the buffer 44 and the retransmitted packet data. The packet data obtained by this combination is decoded and the transmission data is reproduced.

If the error is still detected in the combined packet data, retransmission control is repeated between the terminal device 3 and the relay station 2. In this case, since the packet data are cumulatively combined, the reception characteristics are enhanced by repeating retransmission.

<Case 3>
In Case 3, the communication environment between the base station 1 and the relay station 2 and between the relay station 2 and the terminal device 3 is not good. However, it is assumed that the relay station 2 does not detect an error in the received packet. Hereinafter, the sequence in case 3 will be described with reference to FIG.

The packet transmitted from the base station 1 is transferred to the terminal device 3 by the relay station 2 as in the

cases

1 and 2. At this time, the relay station 2 transfers the packet by the AF method. Even in case 3, the relay station 2 does not detect an error in the received packet. Therefore, the reproduction packet is stored in the buffer 33 and an ACK message is returned to the base station 1.

However, in case 3, the quality of the received radio signal is worse than the threshold level. In this case, the relay station 2 predicts that an error will be detected in the terminal device 3, and executes retransmission control. That is, the relay station 2 executes retransmission control without waiting for a NACK message from the terminal device 3. Specifically, when the quality measurement unit 30 determines that the quality of the received radio signal is worse than the threshold level, the reproduction packet stored in the buffer 33 is transmitted to the terminal device 3. That is, packet retransmission is performed by the DF method.

The terminal device 3 receives the first packet and the retransmission packet. Here, in the example shown in FIG. 8, the error of the first packet is detected. Then, the terminal device 3 generates a NACK message and sends it back to the relay station 2 in order to request packet retransmission. However, as described above, the relay station 2 performs packet retransmission regardless of the presence or absence of the NACK message. Therefore, the terminal device 3 can receive a retransmission packet earlier than when a retransmission is requested with a NACK message. In the example illustrated in FIG. 8, the terminal device 3 receives a retransmission packet after a three-slot period from when the first packet is received. This shortens the delay time related to retransmission.

Note that the relay station 2 receives the NACK message from the terminal device 3. Therefore, normally, the relay station 2 executes retransmission control according to the NACK message. However, in case 3, the relay station 2 performs packet retransmission (predictive retransmission) based on the reception quality before receiving the NACK message. Therefore, in this case, the relay station 2 does not perform retransmission control according to the NACK message.

Also, the NACK message transmitted from the terminal device 3 may arrive at the base station 1. When the base station 1 receives this NACK message, it performs retransmission control. In this case, retransmission packets are duplicated. Therefore, the relay station 2 transmits an ACK message to the base station 1 at a timing when the terminal device 3 transmits a NACK message. Then, in the base station 1, the radio wave of the ACK message from the relay station 2 is stronger than the radio wave of the NACK message from the terminal device 3, and thus the base station 1 ends the process according to the ACK message. .

Further, similarly to Case 2, the terminal device 3 combines the first packet and the retransmission packet to reproduce the transmission data.
If the terminal device 3 does not detect an error in the first packet, the terminal device 3 returns an ACK message to the relay station 2. However, in case 3, the relay station 2 performs packet retransmission regardless of the error detection result in the terminal device 3. That is, the terminal device 3 receives an unnecessary retransmission packet. In this case, the terminal device 3 discards the received retransmission packet.

As described above, in the radio packet relay method of the embodiment, when the reception quality at the relay station 2 is poor, retransmission control is executed without waiting for an ACK / NACK message from the terminal device 3. Therefore, the delay time related to retransmission is shortened.

<Case 4>
In Case 4, the communication environment between the base station 1 and the relay station 2 is bad. Hereinafter, the sequence in case 4 will be described with reference to FIG.

The packet transmitted from the base station 1 is transferred to the terminal device 3 by the relay station 2 as in the cases 1 to 3. However, in Case 4, since the communication environment between the base station 1 and the relay station 2 is poor, the relay station 2 detects an error in the received packet. Therefore, a NACK message is returned from the relay station 2 to the base station 1. At this time, the relay station 2 does not store the reproduction packet in the buffer 33. In the example shown in FIG. 9, an error in the received packet is detected also in the terminal device 3, and a NACK message is returned from the terminal device 3 to the relay station 2.

When receiving the NACK message, the base station 1 retransmits the packet. That is, retransmission control by the base station 1 is executed.

FIG. 10 is a diagram comparing the delay times of the AF method, the DF method, and the embodiment. The horizontal axis of the graph shown in FIG. 10 indicates the probability that a packet error is detected at the receiving station when relayed by the AF method. The vertical axis represents the delay time required for relaying in units of time slots shown in FIGS. However, the following conditions are introduced in FIG.
(1) Packets transferred by the DF method have zero error rate at the receiving station. (2) When two packets transferred by the AF method are combined, the error rate is zero at the receiving station. (3) The AF method is relayed. The delay is zero (4) The time required for packet processing at the relay station and the receiving station is 3 slot period (5) The prediction accuracy is 100% in the system of the embodiment

When the error rate is low, the AF method delay time is very small. However, the AF method requires 6 slot periods for retransmission control. Therefore, if the communication environment is bad and the error rate is high, the delay time increases rapidly. In the DF method, the delay time is constant under the condition (1). In this example, the delay time of the DF method is 3 slot periods. That is, in the DF method, the delay time is not reduced even when the communication environment is good.

In the method of the embodiment, when the error rate is low, the AF method becomes dominant, so the delay time is small. Also, as the error rate increases, the frequency of packet retransmission increases accordingly. However, in the method of the embodiment, the time required for packet retransmission is a three-slot period as shown in FIG. Therefore, even if the error rate increases, the delay time does not increase so much.

In the method described in Non-Patent Document 1, as shown in FIG. 2, when an error is detected at the destination station, a NACK message is returned, and retransmission control is executed according to the NACK message. Therefore, the delay time of the method described in Non-Patent Document 1 is the same as that of the AF method under the above conditions (1) to (3).

FIG. 11 is a flowchart showing the operation of the transmitting station (base station 1, transmitting station 10). In step S1, input of packet data to be transmitted is awaited. When packet data to be transmitted is input, the packet is transmitted to the relay station in step S2. At this time, the transmitted packet is stored in the buffer 13. In step S3, reception of an ACK / NACK message is awaited. In step S4, it is checked whether the received message is ACK or NACK. If an ACK message is received, the process returns to step S1 to wait for input of the next packet data. On the other hand, when the NACK message is received, the packet stored in the buffer 13 is transmitted to the relay station in step S5. That is, packet retransmission is executed.

In cases 1 to 3 described above, an ACK message is returned from the relay station, so steps S1 to S4 are executed. On the other hand, in case 4, since the NACK message is returned from the relay station, steps S1 to S5 are executed.

FIG. 12 is a flowchart showing the operation of the relay station (relay stations 2 and 20). In step S11, it waits for a packet from the transmitting station. When a packet from the transmitting station is received, in step S12, the packet received from the transmitting station is transferred to the receiving station without being demodulated / decoded. That is, the packet is relayed by the AF method. In step S13, the received signal is demodulated / decoded. In step S14, the quality of the received radio signal is measured. In step S15, it is checked whether the received packet is a retransmission packet.

If the received packet is not a retransmission packet, steps S16 and S17 are skipped and the process proceeds to step S18. In step S18, it is checked whether there is an error in the received packet. If an error is detected, the received packet is stored in the buffer 24 in step S19. In step S20, a NACK message is returned to the transmitting station.

If no error is detected in step S18, the packet data after error correction is encoded again in step S21 and stored in the buffer 33. That is, the reproduction packet is stored in the buffer 33. In step S22, an ACK message is returned to the transmitting station. In step S23, it is checked whether an ACK message corresponding to the packet transferred in step S12 has already been received from the receiving station.

If the corresponding ACK message has been received, the process ends and the process returns to step S11. On the other hand, if the corresponding ACK message has not been received, it is checked in step S24 whether the measured reception quality is worse than the threshold level. The reception quality is measured in step S14 in this flowchart. If the reception quality is better than the threshold level, the process ends and the process returns to step S11. On the other hand, if the reception quality is worse than the threshold level, the reproduction packet stored in the buffer 33 is transmitted to the receiving station in step S25. That is, the packet is retransmitted by the DF method.

When the retransmission packet is received from the transmitting station (step S15: Yes), it is checked in step S16 whether the corresponding packet transmitted earlier by the transmitting station is correctly received. If the previous packet has been correctly received, the process proceeds to step S22. On the other hand, if the previous packet has not been received correctly, the previous packet stored in the buffer 24 and the retransmission packet are combined in step S17. When the process of this flowchart is executed for the previous packet, the previous packet is stored in the buffer 24 in step S19.

As described above, in the relay station of the embodiment, when an error of the received packet is not detected (step S18: no error) and the reception quality is worse than the threshold level (step S24: Yes), packet retransmission is executed in step S25. Is done. This packet retransmission is executed regardless of the ACK / NACK message from the receiving station as shown in the flowchart of FIG.

FIG. 13 is a flowchart showing the operation of the relay station corresponding to the ACK / NACK message. The process of this flowchart is executed in parallel with the process shown in FIG.

In step S31, an ACK / NACK message returned from the receiving station is awaited. When a message is received from the receiving station, it is checked in step S32 whether the message is ACK or NACK. If an ACK message is received, the process ends and the process returns to step S31. On the other hand, if a NACK message is received, the process proceeds to step S33. In step S33, it is checked whether the received message is a NACK message corresponding to the packet stored in the buffer 33. When the corresponding NACK message is received, the packet stored in the buffer 33 is transmitted to the receiving station in step S34. On the other hand, when another NACK message is received, the process returns to step S31.

As described above, in the relay station of the embodiment, when a NACK message is received from the receiving station, general retransmission control according to the NACK message is executed.

Here, the operation of the relay station shown in the flowchart will be described with reference to the cases 1 to 4 described above. In Case 1, after the packet is transferred to the receiving station by the AF method in Step S12, it is determined that there is no error in Step S18. Further, in step S24, it is determined that “reception quality is good”. Therefore, the retransmission control in step S25 is not executed. Furthermore, since the ACK message is returned from the receiving station, the retransmission control in step S34 is not executed.

Also in case 2, as in case 1, after the packet is transferred to the receiving station by the AF method in step S12, the retransmission control in step S25 is not executed. However, since the NACK message is returned from the receiving station, the retransmission control in step S34 is executed.

In case 3, as in case 1, after the packet is transferred to the receiving station by the AF method in step S12, it is determined that there is no error in step S18. Therefore, the reproduction packet is stored in the buffer 33 in step S21. However, since it is determined in step S24 that "reception quality is bad", retransmission control in step S25 is executed. That is, the reproduction packet stored in the buffer 33 is transmitted to the receiving station.

In Case 4, after the packet is transferred to the receiving station by the AF method in Step S12, it is determined that “there is an error” in Step S18. Therefore, a NACK message is returned to the transmitting station in step S20, and retransmission control by the transmitting station (step S5 in FIG. 11) is executed.

FIG. 14 is a flowchart showing the operation of the receiving station (

terminal devices

3, 3a, 3b, receiving station 40). The operation of the receiving station is basically the same as that of the relay station (steps S11 to S22 shown in FIG. 12). However, the receiving station does not execute the transfer process in step S12 and the quality measurement in step S14. However, the terminal device may have a function of measuring the quality of the radio signal.

In case 1, “No” is determined in step S43, “No error” is determined in step S46, and then an ACK message is returned to the relay station in step S50.

In cases 2 to 4, it is determined in step S46 that there is an error for the first packet, and a NACK message is returned to the relay station in step S48. Thereafter, in case 2, the relay station retransmits the packet according to the NACK message. In case 3, the relay station retransmits the packet autonomously (that is, regardless of the NACK message). Further, in case 4, the transmitting station retransmits the packet. In any case, in step S45, the receiving station combines the first packet received earlier and the retransmission packet.

Next, the reception quality measured at the relay station will be described. The reception quality is not particularly limited. For example, the following parameters are measured.

(1) Error correction number The demodulated data obtained by the demodulator 22 is held in the relay station. Further, reproduced data obtained by performing error correction on the demodulated data and further executing re-encoding is held. Then, the number of error corrections is detected by comparing the demodulated data and the reproduced data for each bit. The smaller the number of error corrections, the higher the quality.

(2) Number of iterations of turbo decoding In turbo decoding, the same operation is repeatedly executed while monitoring the presence of errors. Then, when the error disappears, the calculation ends. Therefore, the number of repetitions executed until there is no error represents the reception quality. That is, if the number of repetitions is small, the reception quality is good. In the turbo code, the number of repetitions executed until the number of errors becomes smaller than a predetermined value may be counted.

(3) SIR
The ratio of the received radio signal to noise is measured.
(4) Received power The power of the received radio signal is measured.

An example is shown. Here, it is assumed that the number of error corrections is used as the reception quality. Further, a 10-bit error is allowed between the base station and the relay station, and a 10-bit error is allowed between the relay station and the terminal device. That is, it is assumed that each relay station and terminal device can correct errors up to 10 bits. In this case, the relay station transfers the radio signal to the terminal device by the AF method, and demodulates the radio signal to count the number of error corrections. At this time, if the number of error corrections exceeds 10 bits, the relay station transmits a reproduction packet to the terminal device without waiting for an ACK / NACK message from the terminal device. On the other hand, if the number of error corrections is 10 bits or less, the relay station waits for an ACK / NACK message from the terminal device. When the relay station receives the NACK message, the relay station transmits a retransmission packet to the terminal device simultaneously with the retransmission timing of the base station. When the ACK message is received, the relay station does not perform retransmission control.

In step S24 in FIG. 12, the measured quality is compared with the threshold level. This threshold level is a value for predicting “whether or not an error that cannot be corrected at the receiving station occurs when the packet is transferred to the receiving station by the AF method”. That is, “if the quality is better than this threshold level, even if the packet is transferred to the receiving station by the AF method, no error occurs at the receiving station, or all errors can be corrected at the receiving station” or “this If the quality is lower than the threshold level, the threshold level is determined based on the criterion that if the packet is transferred to the receiving station by the AF method, an error that cannot be corrected occurs at the receiving station. Such threshold levels can be determined based on simulations or actual measurements.

Also, the threshold level may be determined in consideration of the packet coding rate. That is, when the coding rate is small, the number of errors that can be corrected at the receiving station is large, so that the threshold level can be lowered. On the other hand, when the coding rate is large, the threshold level may be increased because the number of errors that can be corrected at the receiving station is small.

FIG. 15 is a diagram for explaining the configuration and operation of the

error correction encoders

12 and 32. In this example, it is assumed that a turbo code is used as an error correction code. Note that the configurations and operations of the error correction coding units provided in the transmitting station and the relay station are basically the same.

As shown in FIG. 15A, the error correction encoding unit includes an interleaver 51,

encoders

52 and 53, a puncturing unit 54, and a multiplexing unit (MUX) 55, and encodes from an input information sequence X. Data Z is generated. The interleaver 51 performs an interleaving process on the information sequence X to generate a data sequence Y. In the interleaving process, bit rearrangement is performed according to a predetermined algorithm. The encoder 52 encodes the information sequence X to generate a parity sequence P (1), and the encoder 53 encodes the data sequence Y to generate a parity sequence P (2). Then, the puncturing unit 54 generates a parity sequence P (3) by selecting one or more bits from the parity sequences P (1) and P (2). In the example shown in FIG. 15B, 500 bits are selected from the parity sequence P (1), and 500 bits are selected from the parity sequence P (2). Then, multiplexing section 55 multiplexes information sequence X and parity sequence P (3) to generate encoded data Z.

Here, the error correction encoding unit 12 included in the transmission station and the error correction encoding unit 32 included in the relay station may generate the encoded data Z with the same puncturing pattern, or with different puncturing patterns. The encoded data Z may be generated. When encoding with puncturing patterns different from each other, there is no particular limitation. For example, the error correction encoding unit 12 included in the transmission station may include the odd number bits of the parity sequence P (1) and the P (2). The even bit is selected, and the error correction coding unit 32 included in the relay station selects the even bit of the parity sequence P (1) and the odd bit of P (2).

In a configuration in which encoded data is generated with different puncturing patterns in the transmitting station and the relay station, when the packet retransmission is executed by the relay station, the receiving station can use different error correction codes. Therefore, error correction capability is improved.

Also, when the quality of the received signal is not good, the relay station may retransmit a part of the received signal. For example, as shown in FIG. 16, when the SIR is deteriorated below the threshold level in a partial area of the received packet signal, only the data in that area may be retransmitted.

Claims

A relay station that receives a radio signal from the transmitting station and transmits a radio signal to the receiving station based on the received radio signal, and performs relay processing of transmission data from the transmitting station to the receiving station; and In the retransmission method performed between
Without demodulating the radio signal received from the transmitting station, the first transfer process to the receiving station is performed on the radio signal,
Based on the quality of the received radio signal, based on data obtained by demodulating or error correction decoding the radio signal, performing a second transfer process,
The receiving station performs reception processing of the radio signal transmitted by the first transfer processing and the radio signal transmitted by the second transfer processing.
A retransmission method in a relay station.
The relay station includes an error detection unit,
The second transfer process is executed when it is determined that there is no error in the data obtained by demodulating or error correcting decoding the radio signal.
The retransmission method according to claim 1.
The second transfer process includes an error correction encoding process and a puncture process after the error correction encoding process,
The puncturing process is executed with a puncture pattern different from a puncture pattern performed by the transmitting station on the radio signal.
The retransmission method according to claim 1.
The quality is the number of error corrections for the received radio signal, the number of iterations of turbo decoding performed until no error is detected, or the received power, or the SIR,
The second transfer process is executed when the quality falls below a predetermined standard.
The retransmission method according to claim 1.
In the second conversion process, data in a region where the quality falls below a predetermined standard is transmitted to the receiving station.
The retransmission method according to claim 1.
When performing the second transfer process, the relay station transmits a positive reception response for the received radio signal,
The retransmission method according to claim 1.
In a radio relay apparatus that performs a relay process of transmission data from the transmission station to the reception station by receiving a radio signal from the transmission station and transmitting the radio signal to the reception station based on the received radio signal.
A first transfer unit that performs a first transfer process on the wireless signal to the receiving station without demodulating the wireless signal received from the transmitting station;
A second transfer unit that executes a second transfer process based on data obtained by demodulating or error correcting decoding the wireless signal based on the quality of the received wireless signal;
A wireless relay device comprising:
A transfer unit that transfers the radio signal transmitted from the transmitting station to the receiving station without demodulating;
A measurement unit for measuring the reception quality of the radio signal;
When the reception quality measured by the measurement unit is lower than a threshold level, a retransmission unit that transmits reproduction data obtained by demodulating and decoding the radio signal to the receiving station;
A wireless relay device.
An error detector for detecting an error in the radio signal;
The retransmission section transmits the reproduction data to the receiving station when the quality is lower than a threshold level and an error is not detected by the error detection section. Wireless relay device.
An error correction unit for correcting an error of the radio signal;
An error detection unit for detecting an error that could not be corrected by the error correction unit;
The retransmission section transmits the reproduction data to the receiving station when the quality is lower than a threshold level and an error is not detected by the error detection section. Wireless packet relay device.