WO2012171291A1

WO2012171291A1 - Method and device for data retransmission and merging based on harq

Info

Publication number: WO2012171291A1
Application number: PCT/CN2011/080991
Authority: WO
Inventors: 杨芸霞
Original assignee: 中兴通讯股份有限公司
Priority date: 2011-06-14
Filing date: 2011-10-19
Publication date: 2012-12-20
Also published as: CN102833054A

Abstract

Disclosed are a method and device for data retransmission and merging based on a hybrid automatic repeat request (HARQ). The method includes: when receiving the data of the Nth transmission of a data packet, merging the data of the Nth transmission of the data packet, and performing the following operations on each coding block of the data packet: determining a weighted merging weight value of the coding blocks to be merged when merging data in the data of each transmission according to the signal-to-interference and noise ratio of the coding blocks to be merged in the data of the Nth transmission; performing data merging on the coding blocks to be merged in the data of the Nth transmission according to the determined weighted merging weight value; and obtaining the merged data of the data packet until the data of all coding blocks in the data packet has finished merging. The influence of channel state is considered when merging data, and the data merging is performed according to the weight, increasing the correct rate of merging data, and improving system performance.

Description

HARQ-based data retransmission combining method and device

The present invention relates to communication technologies, and more particularly to a data retransmission combining method and apparatus based on Hybrid Automatic Repeat ReQuest ( HARQ). Background technique

Under the application of long-distance global interoperability for Microwave Access (WIMAX)/Long Term Evolution (LTE) technology, the fading phenomenon of the wireless channel is very obvious. Therefore, WiMAX/LTE The technology adds a HARQ mechanism at the link layer to reduce information errors that reach the network layer. The HARQ technology is based on the link layer retransmission combining technology, which combines an automatic retransmission request (ARQ, Automatic Repeat request) and a forward error correction coding (FEC, Feed-forward Error Correction) data retransmission mechanism. Overcoming the shortcomings of ARQ and FEC, it can improve spectrum efficiency, greatly improve system service throughput, use retransmission to bring combined gain, and indirectly expand system coverage. The following is a detailed analysis of the difference in throughput between traditional ARQ and HARQ:

The implementation process of the ARQ technology is as follows: The sender sends a data packet containing bit information and bit check information. If the receiver detects the packet error, the receiver discards the data packet and requests the sender to retransmit. The average number of transmissions received by the receiving end of a data packet 7> can be expressed as:

7 = 1 + £ corpse

7=1 where, assuming that the undetected error can be ignored, the probability of merging the error for the previous packet, and the probability of the error occurring in the first retransmission is numerically estimated, which can be used The formula below indicates:

Where β =(Η ··,ίή, =(h ,-,h[) ^ is the channel response of the /ant antenna in the _ /th transmission, the number of antennas transmitted or received; ^/( ^) is the different antenna The gain function, Ρ( ), is the Conditioned Bit Error Rate (CBER) under channel decoding conditions.

Among them, if the channel is not related to retransmission in different times, Bay' J:

Wherein, if the channel retransmission of different times is related, the channel response corresponding to each antenna for each transmission is consistent, and:

The above ^ is the average power of each symbol data.

According to the above calculated 7>, it can be determined that the throughput efficiency of the ARQ is:

J k_

Tr k + n where z + is the factor of throughput loss for increased risk bits.

The implementation process of HARQ technology is: Performing Cyclic Redundancy Check (CRC) on a data packet at the transmitting end to make it have certain error detection capability, and then performing FEC encoding on the data after CRC, and obtaining a certain Data for error correction capabilities. After receiving the data packet, the receiving end first performs decoding error correction, and then performs error detection through the CRC. If a data error is detected, a retransmission request is initiated to the transmitting end. However, during the implementation of the HARQ technology, the erroneous data packets are not discarded, but when the retransmitted data packets arrive, the new data packets are obtained through the packet merging. That is to say, the principle of Chase HARQ packet merging is to save the wrong old data packet and combine it with the retransmitted new data packet. Assume that its FEC is the code rate = ^{/ v} , the memory is the convolutional code, and the receiver uses the soft input Viterbi decoding. If the channel is independent for different retransmissions, the probability of packet error in the first transmission for:

Wherein, the event that the first j data packets are combined to detect the error is the upper bound of the error event probability, and satisfies:

Where, for the free distance, the distance is not the correct number of paths.

Then the average number of transmissions received by the receiving end of a data packet 7 can be expressed as

7 = 1 + £ corpse (13⁄4)

7=1

Thr -

Tr k + n _p + um where / + is drawn to increase the check digit and the loss factor caused by the return to zero. It can be seen from the throughput calculation formulas of the above two ARQ and HARQ technologies that the throughput of the HARQ technology has been greatly improved.

However, in the traditional Chase HARQ, in the process of storing the wrong old data packet and performing equal gain combining and averaging with the retransmitted new data packet, the influence of the channel condition on the quality of the transmitted data packet is not considered, and the mean value combination of the same weight is directly performed. . In the actual wireless environment, the channel conditions of each retransmission are completely different. In this way, the channel conditions of a retransmission packet are better, and the channel conditions of a retransmission packet are poor, even a certain In the case where the secondary retransmission will be seriously interfered, the method of direct mean merging is adopted, and the combined weights of the received data packets are the same, and the letter cannot be reflected. The impact of channel quality conditions on the quality of data packet transmission, resulting in a stronger interference retransmission packet. When merging, the probability of merging data detection error increases, and the gain of retransmission and merging is greatly reduced, resulting in an increase in the number of data retransmissions. . When retransmissions have strong interference, it may even cause

Chase HARQ system performance is lower than non-Chase HARQ system performance. Summary of the invention

The invention provides a HARQ-based data retransmission combining method and device, which is used to solve the problem that the combined data retransmission probability caused by the current retransmission data mean value combination is high, the retransmission combining gain is low, and the system performance is affected.

The present invention provides a HARQ-based data retransmission combining method, including:

When the Nth transmission data of a data packet is received, the N transmission data of the data packet is merged, and the following operations are performed on each coding block of the data packet:

Determining, according to a signal to interference and noise ratio of the coding block to be combined in the N times of transmission data, a weighted combining weight of the to-be-combined coding block in the data to be merged;

And performing data combining on the to-be-combined coded blocks in the N times of transmission data according to the determined weighted combining weights;

The merged packet data is obtained until the data of all the encoded blocks in the data packet are merged.

A HARQ-based data retransmission combining device includes: a receiving module, a merge control module, and a data merge module; wherein

The receiving module is configured to receive transmission data of a data packet;

The merging control module, when receiving the Nth transmission data of a data packet, merging the N transmission data of the data packet, and controlling the data merging module to perform for each coding block of the data packet Merging operation, until the data of all the coding blocks in the data packet is merged, and the combined data packet data is obtained;

The data merging module is configured to code the block to be merged, according to the data to be transmitted N times a signal to interference and noise ratio of the combined coded block, determining a weighted combining weight of the to-be-combined coded block in data transmission in each transmission data; and coding to be combined in the N times of transmission data according to the determined weighted combining weight The block performs data merging.

The beneficial effects of the present invention are as follows:

The HARQ-based data retransmission combining method and device provided by the embodiment of the present invention, when performing data combining on multiple retransmitted data packets, determining, according to the signal to interference and noise ratio, each coding block in each retransmission data The weighted combining weights at the time of data combination are combined according to the weighted combining weights, so that the weights of the transmission data of different transmission channel conditions are different at the time of combining, and the data with a good channel condition has a larger weight, and the data with poor channel conditions The weight of the station is small, so that the transmission data is more contributed to the combined data when the channel condition is good, and the data transmitted is merged when the data is merged. The contribution is also relatively small, so that the accuracy of the combined data is greatly improved, the correct rate of the merged data is improved, the merge gain is maximized, the system performance is greatly improved, and the number of data retransmissions is virtually reduced. The implementation of the signal-to-noise ratio estimation is less complex in the implementation of the system, and the system cache requirement is smaller. At the same time, the CINR estimation granularity is smaller than the entire HARQ retransmission data packet, and the accuracy is higher, which contributes significantly to the improvement of the HARQ combining gain. . DRAWINGS

The drawings are intended to provide a further understanding of the invention, and are intended to be a part of the invention. In the drawing:

FIG. 1 is a flowchart of a HARQ-based data retransmission combining method according to an embodiment of the present invention; FIG. 1 is a schematic structural diagram of a HARQ-based data retransmission combining apparatus according to an embodiment of the present invention. detailed description

In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention clearer, It is to be understood that the present invention will be further described in detail below with reference to the drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

In order to improve the old data packet to be saved in the prior art and the retransmitted new data packet, the gain combining average does not reflect the difference in channel transmission conditions, thereby causing the problem that the combined data is not accurate. The HARQ-based data retransmission combining method provided by the example, when performing data merging, determining the weight of the merging according to the channel quality, and realizing the merging of the data packets. The method flow is as shown in FIG. 1 and includes the following steps:

Step S11: Receive transmission data of the Nth retransmission of a data packet;

Here, in the method of the present invention, the transmission data of the N retransmitted data packets received before the CRC check is passed is stored for use in subsequent merging.

Step S12: Start a merge process of the N times of data transmission of the above data packet.

Step S13: Determine a Signal to Interference plus Noise Ratio (CINR) of each coding block of the N times of transmission data.

Here, before the merging of the N transmission data of the above data packet, the CINR of each coding block for determining the N transmission data is further included. There are two ways to determine this:

Manner 1: When receiving the Nth transmission data of a data packet, the CINR of each coding block in the Nth transmission data is counted, and the CINR of each coding block in the previous N-1 transmission data that is previously counted and stored is obtained. .

Manner 2: When receiving the Nth transmission data of a data packet, the CINR of each coding block in the N transmission data is counted.

The mode 1 can only count the CINR of each coding block in the transmission data when receiving the Nth transmission data, and reduce the statistical workload when each data is merged. In the second mode, when the Nth transmission data is received, only the CINR of each coding block in the previous N times of transmission data including the current time can be counted, so that the system does not need to store the previous statistical data, thereby saving system storage space. For example: the signal to interference and noise ratio of the mth coding block in the transmission data of the Nth retransmission is CINR; where m represents the mth FEC coding block of the data packet, me {0, ..., M -1} , M is a positive integer, and N represents the transmission data of the Nth retransmitted packet.

Step S14: Acquire data of the coding block to be merged in the N times of transmission data.

Here, the code block data combining operation of step S15 and step S16 is performed for each code block of the above-described data packet.

Step S15: Determine, according to the CINR of the coding block to be merged in the N times of transmission data, a weighted combining weight of the coded block to be combined in the data transmission at the time of data combining.

Specifically, the CINR of the coding block to be combined in the N times of transmission data is normalized, and the weighted combining weights of the to-be-combined coding blocks in the data to be merged in each transmission data are determined according to the following formula:

Where ^w represents the weighted combining weight of the mth coded block in the nth transmission data at the time of data combination, me{0,..., M-1}, _M is a positive integer, ne {0,...,N } , _N is a positive integer;

^CINR represents the ^CINR of the mth coded block in the nth transmission data;

∑ ' ^N =. ^aNR represents the sum of the CINRs of the mth coded block in the N transmission data, ie {0, ······, N}. Step S16: Perform data combining on the to-be-combined coded blocks in the N times of transmission data according to the determined weighted combining weights to obtain the combined coded block data.

Here, for the data of each information bit in the code block to be merged, the data is merged by using the following formula: where bit ^ represents the combined data of the kth information bit in the mth code block, me {0, , M -1} , M is a positive integer, ke {0, ······ , K-1}, K is a positive integer; b ' ^m denotes the data of the kth information bit in the mth code block of the nth transmission, ne {0, ······, N} , N is a positive integer;

Representing a weighted merge weight of the mth code block in the nth transmission data at the time of data combining;

After merging the data of all the information bits in the code block to be merged, the combined code block data is obtained.

Step S17: Whether to merge the data of all the coding blocks in the above data packet, if yes, execute step S19; otherwise, execute step S18.

Step S18: Determine the next code block to be merged.

Here, after combining the data of one coding block, starting to merge the data of the next coding block, determining the next coding block to be merged, returning to step S14 to continue acquiring the data of the next coding block to be combined.

Step S19: Obtain the combined data packet data.

When the Nth transmission data of a data packet is received in step S11, the N transmission data of the data packet is merged, and the above data combining operation is performed for each coding block of the data packet until the data packet is merged. The data of all the encoded blocks in the data is obtained, and the combined packet data is obtained.

Preferably, the foregoing HARQ-based data retransmission combining method provided by the embodiment of the present invention further includes:

Step S20: Perform CRC check on the merged data packet, and if the check fails, continue to request the sender to retransmit the data of the data packet.

When the check passes, the data of the packet is no longer requested to be resent.

Based on the foregoing data retransmission combining method provided by the embodiment of the present invention, the embodiment of the present invention further provides a HARQ-based data retransmission combining device, where the device can be set in any network device having a data receiving function, and the device is The structure is as shown in FIG. 2, including: a receiving module 10, The control module 20 and the data merge module 30 are combined.

The receiving module 10 is configured to receive transmission data of the data packet.

The merging control module 20, when receiving the Nth transmission data of a data packet, merging the N transmission data of the data packet, and the control data merging module performs a merging operation for each coding block of the data packet until The data of all the coding blocks in the data packet is merged to obtain the combined data packet data.

a data merging module 30, configured to determine, according to a CINR of a coding block to be merged in the N times of transmission data, a weighted combining weight of the to-be-combined coded block in the data combination when the data is merged; And performing data combining on the to-be-combined coding blocks in the N times of transmission data according to the determined weighted combining weights to obtain the combined coded block data.

Preferably, the data combining module 30 further includes: a weight determining unit 301 and a data combining unit 302.

The weight determining unit 301 is configured to perform normalization calculation on the CINR of the coded block to be merged in the N times of transmission data, and determine, according to the following formula, the weighted combined weight of the to-be-combined coded block in the data combination. :

Wherein, the weighted merge weight of the mth coded block in the nth transmission data at the time of data merge, me {0, ······, M-1}, M is a positive integer, ne {0, · ····· , N} , N is a positive integer;

CINR represents the CINR of the mth coded block in the nth transmission data;

Γ^ΟΝΕ^ indicates the sum of the CINRs of the mth code block in the 传输 transmission data, ie {0, ······ , N}.

The data merging unit 302 is configured to perform data merging on the to-be-combined coded blocks in the N times of transmission data according to the determined weighted merging weights to obtain the merged coded block data, and specifically includes: The data of the information bits is combined by the following formula: Bit= _me =∑ _n ^N =. W bit

Where bit ^ represents the merged data of the kth information bit in the mth coded block, me {0, , M - 1} , M is a positive integer, ke {0, · · ··· · · , K -1 } , K is a positive integer;

Bit^ ^m represents the data of the kth information bit in the mth coded block of the nth transmission, ne {0, · · ··· · ·, N} , N is a positive integer;

After merging the data of all the information bits in the coded block to be merged, the combined coded block data is obtained.

Preferably, the foregoing HARQ-based data retransmission combining device further includes:

The statistic module 40 is configured to determine the CINR of each coding block of the N times of transmission data when the data of the N times of the data transmission is combined, and specifically includes: when receiving the Nth transmission data of a data packet, the statistics are N. Transmitting the CINR of each coding block in the data, and acquiring the CINR of each coding block in the previous N-1 transmission data that is previously counted and stored; or, when receiving the Nth transmission data of one data packet, counting N times The CINR of each code block in the transmitted data.

The verification module 50, after obtaining the combined data packet data, performs CRC check on the combined data packet data, and if the verification fails, continues to request the transmitting end to retransmit the data of the data packet.

When the baseband data is retransmitted and combined, the method determines the weight of each transmission data in the data combining according to the CINR reflecting the channel transmission condition, that is, the weighted combining weight, and the data with good channel conditions The weight of the data station is relatively large, and the weight of the data station with poor channel conditions is relatively small. Thus, when the channel condition is good, the transmission data has a relatively large contribution to the combined data at the time of merging, and the transmission is performed under severe interference. When the data is merged, the contribution to the merged data is also relatively small, so that the accuracy of the combined data is greatly improved, and the merger is increased. Maximizing benefits greatly improves system performance and virtually reduces the number of data retransmissions. In the above method, the CINR of the FEC coding block is used as the weighted combining weight. The CINR estimation has less implementation complexity in the implementation of the system, and requires less system buffering. At the same time, the CINR estimation granularity is smaller than the entire HARQ retransmission packet. The accuracy is higher, and the contribution to the improvement of the HARQ combining gain is more obvious.

The above description shows and describes a preferred embodiment of the present invention, but as described above, it should be understood that the present invention is not limited to the form disclosed herein, and should not be construed as Other combinations, modifications, and environments are possible and can be modified by the teachings or related art or knowledge within the scope of the inventive concept described herein. All changes and modifications made by those skilled in the art are intended to be within the scope of the appended claims.

Claims

Claim

A data retransmission combining method based on hybrid automatic repeat request HARQ, characterized in that: the method comprises:

When the Nth transmission data of a data packet is received, the N transmission data of the data packet is combined, and the following operations are performed on the coding block of the data packet:

The method according to claim 1, wherein the determining, according to the signal to interference and noise ratio of the coding block to be combined in the N times of transmission data, determining that the to-be-combined coding block in the data to be combined is in data combining Weighted consolidation weights, including:

The signal-to-noise ratio of the coded blocks to be combined in the N times of transmission data is normalized, and the weighted merge weights of the to-be-combined coded blocks in the data combination are determined by using the column formula:

w ^m ° = Z ^CINR - ^oCINR^

Wherein, the weighted merge weight of the mth coded block in the nth transmission data at the time of data merge, me {0, ······, M -1} , M is a positive integer, ne {0, · ····· , N} , N is a positive integer;

CIN1 represents the signal to interference and noise ratio of the mth coded block in the nth transmission data;

Z^CINR represents the sum of the signal-to-interference ratio of the mth coded block in the N-transmission data, ie {0, ······ , N}.

3. The method according to claim 1, wherein the determining according to the determined weighting And the weight value, the data combination of the to-be-combined coded blocks in the N times of transmission data, including: the data of each information bit in the coded block to be combined, the data is merged by using the following formula:

b : _me =∑ _n ^N =. W bit

Where bit ^ represents the merged data of the kth information bit in the mth coding block, me {0, , M-1} , M is a positive integer, ke {0, ······ , K-1 }, K is a positive integer;

Bit^ ^m represents the data of the kth information bit in the mth coded block of the nth transmission, ne {0, ······, N} , N is a positive integer;

The data of all the information bits in the coded block to be merged is merged to obtain the combined coded block data.

4. The method according to claim 1, wherein before the merging the N times of data of the data packet, the method further comprises:

When receiving the Nth transmission data of a data packet, the signal drying ratio of each coding block in the Nth transmission data is counted, and the signal to interference and noise ratio of each coding block in the previous N-1 transmission data is obtained; or

When the Nth transmission data of a data packet is received, the letter drying ratio of each coding block in the N transmission data is counted.

The method according to any one of claims 1 to 4, wherein after the combined data packet data is obtained, the method further comprises:

The CRC check is performed on the combined packet data, and if the check fails, the requesting end retransmits the data of the packet.

6. A HARQ-based data retransmission combining device, the device comprising: a receiving module, a merge control module, and a data merge module; The receiving module is configured to receive transmission data of a data packet;

The data merging module is configured to: for the coded block to be merged, determine a weighted combination of the to-be-combined coded block in data combination in each transmission data according to a signal to interference and noise ratio of the coded block to be merged in the N times of transmission data Weighting; performing data combining on the to-be-combined coding blocks in the N transmission data according to the determined weighted combining weights.

The device according to claim 6, wherein the data merging module comprises:

a weight determining unit, configured to perform normalization calculation on a signal to interference and noise ratio of the coded block to be merged in the N times of transmission data, and determine, according to the following formula, a weighted combination of the to-be-combined coded block in data combination in data transmission Weight:

Wherein, the weighted merge weight of the mth coded block in the nth transmission data at the time of data merge, me {0, · · ··· · ·, M - 1} , M is a positive integer, ne {0, · · · · · · , N} , N is a positive integer;

Γ^ΟΝΙ^ indicates the sum of the signal-to-noise ratio of the mth code block in the 传输 transmission data, i e {0, · · ··· · · , N}.

The device according to claim 6, wherein the data merging module comprises:

a data merging unit, configured to perform data merging on the to-be-combined coded blocks in the N times of transmission data according to the determined weighted merging weights, to obtain the merging coded block data; specifically: each information in the coded block to be combined Bit data, using the following formula for data combination Open:

Bit= _me =∑ _n ^N =. W bit

b ' ^m denotes the data of the kth information bit in the mth code block of the nth transmission, ne {0, ······, N} , N is a positive integer;

The device according to claim 6, wherein the device further comprises: a statistics module, configured to determine, when the N times of the data transmission of the data packet are combined, the letter of each coding block of the N times of transmission data The dry-to-noise ratio specifically includes: when receiving the Nth transmission data of a data packet, counting the signal drying ratio of each coding block in the Nth transmission data, and acquiring the coding blocks of the first N-1 transmission data. The signal to interference and noise ratio; or, when receiving the Nth transmission data of a data packet, the signal drying ratio of each coding block in the N transmission data is counted.

10. Apparatus according to any one of claims 6 to 9 wherein the apparatus further comprises:

The verification module is configured to perform CRC check on the merged data packet data, and if the verification fails, continue to request the transmitting end to retransmit the data of the data packet.