An adaptive modulation and coding method based on LDPC code

Abstract

The present invention discloses an adaptive modulation and coding method based on LDPC code, which includes the steps of receiving end signal-to-noise ratio estimation, LDPC decoding, calculation of failure ratio of check equation, dynamic threshold adjustment, mode decision, mode feedback, and selecting LDPC coding parameters and symbol mapping mode according to feedback mode at the transmitting end. The present invention can overcome the threshold mismatch problem in the actual channel of the traditional method using a fixed signal-to-noise ratio threshold, and adaptively adjust the decision threshold of each modulation and coding mode to the real threshold corresponding to the current channel, thereby improving the robustness of the algorithm in practical applications. The present invention has low computational complexity, is simple to implement, is easy to implement in engineering, and can be applied to various communication systems using LDPC coding.

Description

Technical Field

[0001] The invention relates to an adaptive modulation and coding method based on LDPC code, which is particularly suitable for various communication systems that need to adaptively adjust modulation and coding modes according to changes in channel conditions. Background Art

[0002] Adaptive modulation and coding technology is a technology that allows the communication system to automatically use the most appropriate modulation and coding methods to transmit information according to the channel conditions. Its purpose is to maximize the transmission rate under a given bit error rate requirement.

[0003] The traditional adaptive modulation and coding method uses the signal-to-noise ratio estimation at the receiving end as the decision factor for mode switching. First, a computer is used to simulate the bit error rate versus signal-to-noise ratio curve for each modulation and coding mode, and then a signal-to-noise ratio threshold is set for each modulation and coding mode according to the expected bit error rate. It is assumed that the bit error rate requirement can be met when the signal-to-noise ratio is higher than this threshold. In the actual operation of the system, the signal-to-noise ratio is first estimated at the receiving end, and then the mode with the highest transmission rate is selected from all modes whose threshold is lower than the signal-to-noise ratio estimation at the receiving end for use. The flowchart of this traditional adaptive modulation and coding method is shown in the figure below. Figure 1 shown.

[0004] However, since the SNR threshold of each mode is not consistent under different channel multipath structures, when the multipath structures of the actual channel and the simulated channel are different, the SNR threshold obtained through simulation will have a large error, resulting in errors in the selection of adaptive modulation and coding modes. The following simulation intuitively demonstrates this problem.

[0005] In the simulation, the fixed symbol mapping constellation is BPSK, and the LDPC code with code rate 1 / 4 and code length 64800 is used. The number of channel multipaths is set to 1, 2, and 3 respectively, and the bit error rate of the receiving end is simulated as the signal-to-noise ratio changes. In the simulation, the channel estimation method uses the pilot-based LS channel estimation, and the channel equalization method uses MMSE equalization. The bit error rate is obtained by averaging the simulation results of 10,000 times under each signal-to-noise ratio. In each simulation, the gain of each channel tap is randomly generated. The simulation results are shown in Figure 2 shown.

[0006] It can be seen that the signal-to-noise ratio thresholds for achieving the same bit error rate under different channel multipath structures for the same modulation and coding method are not the same. Moreover, even if the channel multipath structure itself remains unchanged, the baseband channel multipath number will change when the symbol rate is switched. For example, assuming that the baseband channel multipath number is 1 at a 1MHz symbol rate, after switching the symbol rate to 4MHz, the baseband channel multipath number will become 4. At this time, this type of method will also have a large error.

[0007] In order to overcome the problem of inconsistent signal-to-noise ratio thresholds in different channels, some methods use the network / service layer packet loss rate over a period of time, or the CRC check failure rate of the data link layer over a period of time to assist in adaptive modulation and coding. This type of method counts the packet loss rate / CRC check failure rate of the receiving end over a period of time, and forces the speed to be reduced if it exceeds a certain threshold.

[0008] This type of decision quantity is indeed independent of the channel and is applicable to various channels, but it has the following disadvantages: first, it is necessary to count the reception results over a period of time to obtain the decision quantity, and the algorithm has a long response time to channel changes; second, in units of frames / packets, each transmission can only obtain a binary hard information representing bit errors / no bit errors, but has no idea how many bit errors occurred in this transmission. The amount of information obtained is small and the use is not flexible; third, cross-layer information interaction (physical layer and link layer / network layer / application layer) is required, and the processing delay is long. Summary of the invention

[0009] The technical problem to be solved by the present invention is to avoid the shortcomings in the above background, and provide an adaptive modulation and coding method based on LDPC code, which can dynamically adjust the signal-to-noise ratio threshold of each modulation and coding mode during the communication process, so that the signal-to-noise ratio threshold of each mode is automatically adapted to the channel, improve the robustness of the algorithm in practical applications, and obtain better communication quality. Different from the above technology, the present invention uses the soft information of the failure ratio of the check equation in the output of the LDPC decoder, which is unrelated to the multipath structure of the channel, to dynamically adjust the signal-to-noise ratio threshold of each modulation and coding mode during the communication process, and then use the adjusted threshold to make a judgment, thereby realizing highly robust and fast-response adaptive modulation and coding that can automatically adapt to the channel.

[0010] The technical solution adopted by the present invention is:

[0011] An adaptive modulation and coding method based on LDPC code includes the following processes:

[0012] (1) Set N modulation coding modes as needed, denoted by Ω 1 ,Ω 2 ,…,Ω N , as the selection range of the adaptive modulation and coding algorithm, each modulation and coding mode Ω n Corresponding to a set of parameters {LDPC code length L n , LDPC code rate R n , symbol mapping constellation diagram Γ n}, n = 1, 2, ..., N;

[0013] (2) Modulation coding mode Ω n ,n=1,2,…,N, set the signal-to-noise ratio threshold And the LDPC decoding verification equation failure ratio threshold And the modulation coding modes are arranged in order from small to large according to the signal-to-noise ratio threshold;

[0014] (3) The modulation and coding mode currently used by the system is recorded as Ω u ,u∈{1,2,…,N}, and the corresponding parameters are recorded as When the system starts, u=1 is set, that is, the transmitter uses the mode with the lowest signal-to-noise ratio threshold Ω 1 Carry out data transfer;

[0015] (4) The receiving end estimates the signal-to-noise ratio based on the received signal, and the estimated result is recorded as SNR est , and perform demodulation and LDPC decoding, and record the decoded bit vector as For L u ×1-dimensional column vector;

[0016] (5) Based on Calculate the failure rate r of the verification equation:

[0017]

[0018] Among them, H u is the modulation coding mode Ω u The corresponding LDPC code supervision matrix, H u For L u (1-R u )×L u dimensional matrix, ||·|| 0 represents the zero norm of the orientation quantity, and calculates All addition operations are modulo 2 additions;

[0019] (6) Estimation result based on signal-to-noise ratio SNR est The sum of the failure ratio r of the check equation, for the modulation coding mode Ω u The signal-to-noise ratio threshold Make adjustments:

[0020]

[0021] Among them, δ up It is the signal-to-noise margin when the threshold is raised;

[0022] (7) Estimation result based on signal-to-noise ratio SNR est Make mode judgment and update mode number u:

[0023]

[0024] (8) The determined mode sequence number u is transmitted back to the transmitting end through the feedback channel;

[0025] (9) The sender selects the corresponding parameter {L according to the received mode sequence number u u , R u , Γ u} to send data.

[0026] The present invention utilizes the soft information of the failure ratio of the check equation in the output of the LDPC decoder, which is irrelevant to the channel multipath structure, to dynamically adjust the signal-to-noise ratio threshold of each modulation coding mode during the communication process, and then uses the adjusted threshold to make mode decisions, which has the following advantages:

[0027] First, the method of the present invention can make the signal-to-noise ratio threshold of each mode adaptively converge to the real threshold corresponding to the current channel, thereby ensuring the correctness and robustness of adaptive mode switching;

[0028] Second, the failure ratio of the check equation used in the method of the present invention is soft information, and its threshold can be flexibly set according to the expected bit error rate, and the judgment can be made in one transmission, and the algorithm has a short response time to channel changes;

[0029] Third, the method of the present invention utilizes the inherent properties of LDPC codes and decoded output information to guide adaptive modulation coding, and does not require modification of the existing communication system, which is convenient for promotion and transplantation;

[0030] Fourthly, the method of the present invention does not require cross-layer information interaction, is easy to implement, and has a small processing delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flow chart of the traditional adaptive modulation and coding method;

[0032] Figure 2 It is the relationship curve between bit error rate and signal-to-noise ratio of the same modulation and coding mode under different channels;

[0033] Figure 3 is a flow chart of dynamic threshold adjustment in the method of the present invention;

[0034] Figure 4 is a flow chart of mode determination in the method of the present invention;

[0035] Figure 5 It is a flow chart of the adaptive modulation and coding method proposed by the present invention;

[0036] Figure 6 is the curve of mode 1 bit error rate changing with signal-to-noise ratio under Gaussian channel;

[0037] Figure 7 is the curve of mode 2 bit error rate changing with signal-to-noise ratio under Gaussian channel;

[0038] Figure 8is the curve of bit error rate of mode 3 changing with signal-to-noise ratio under Gaussian channel;

[0039] Fig. 9 is the relationship curve between the bit error rate of mode 1 and the failure ratio of the check equation under Gaussian channel;

[0040] Fig.10 is the relationship curve between the bit error rate of mode 2 and the failure ratio of the check equation under Gaussian channel;

[0041] Fig.11 is the relationship curve between the bit error rate of mode 3 and the failure ratio of the check equation under Gaussian channel;

[0042] Fig.12 It is the relationship curve between the bit error rate of mode 1 and the failure ratio of the check equation in 1, 2, and 3 path channels;

[0043] Fig.13 It is a schematic diagram of the interaction process between the sending and receiving ends in the implementation case;

[0044] Fig.14 is the curve of bit error rate of each mode changing with signal-to-noise ratio in the implementation case under 2-path channel;

[0045] Fig.15 is the simulation result of the adaptive mode selection process under the condition of linear variation of signal-to-noise ratio;

[0046] Fig.16 is the simulation result of the convergence process of threshold values ​​of each mode under the condition of linear variation of signal-to-noise ratio;

[0047] Fig.17 is the simulation result of the adaptive mode selection process under the condition of random variation of signal-to-noise ratio;

[0048] Fig.18 It is the simulation result of the convergence process of thresholds of each mode under the condition of random variation of signal-to-noise ratio. DETAILED DESCRIPTION

[0049] The method proposed in the present invention is further described below in conjunction with specific cases.

[0050] An adaptive modulation and coding method based on LDPC code, the flow chart is as follows Figure 5 As shown, the following process is included:

[0051] (1) Candidate modulation and coding mode settings

[0052] Set N candidate modulation and coding modes as needed, denoted as Ω 1 ,Ω 2 ,…,Ω N , as the selection range of the adaptive modulation and coding algorithm, each modulation and coding mode Ω n ,n=1,2,…,N corresponds to a set of parameters {LDPC code length Ln , LDPC code rate R n , symbol mapping constellation diagram Γ n}.

[0053] (2) Determine the signal-to-noise ratio threshold of each mode and the LDPC decoding verification equation failure ratio threshold

[0054] Through computer simulation, the modulation coding mode Ω is determined n ,n=1,2,…,N signal-to-noise ratio threshold And LDPC decoding check equation failure ratio threshold

[0055] For the signal-to-noise ratio threshold, since this value will be corrected in the actual communication process later, here we only set an initial value, so you can choose any channel for simulation, such as a Gaussian channel. Simulate the bit error rate and signal-to-noise ratio curve of each mode under the selected channel, and select the signal-to-noise ratio corresponding to the expected bit error rate as the signal-to-noise ratio threshold.

[0056] As for the LDPC decoding check equation failure ratio threshold, since it is independent of the specific channel selection, any channel can be selected for simulation. The relationship curve between the bit error rate of each mode and the LDPC decoding check equation failure ratio is simulated under the selected channel, and the check equation failure ratio corresponding to the expected bit error rate is selected as the threshold.

[0057] Here, the modulation coding modes are arranged in order from small to large according to the signal-to-noise ratio threshold, that is, n=1,2,…,N-1.

[0058] (3) System initialization

[0059] The modulation coding mode currently used by the system is recorded as Ω u ,u∈{1,2,…,N}, and its corresponding parameters are recorded as When the system starts, u is initialized to 1, that is, the transmitter uses the mode with the lowest signal-to-noise ratio threshold Ω 1 Perform data transmission.

[0060] (4) Signal-to-noise ratio estimation

[0061] Each time the transmitter transmits a signal, the receiver estimates the signal-to-noise ratio based on the received signal, and the estimated result is recorded as SNR. est (This method does not limit the method of estimating the signal-to-noise ratio).

[0062] (5) Calculation of failure ratio of verification equation

[0063] The receiving end demodulates and LDPC decodes the received signal, and records the decoded bit vector as ( For L u ×1-dimensional column vector).

[0064] according to Calculate the failure rate r of the calibration equation. r can be calculated by the following formula

[0065]

[0066] Among them, H u Is the mode Ω u The corresponding LDPC code supervision matrix (H u For L u (1-R u )×L u dimensional matrix), ||·|| 0 represents the zero norm of the orientation quantity. Note that the calculation All addition operations are modulo 2.

[0067] (6) Dynamic Threshold Adjustment

[0068] According to SNR est and r, for mode Ω u The signal-to-noise ratio threshold Make adjustments:

[0069]

[0070] Among them, δ up It is the signal-to-noise ratio margin when the threshold is raised, which can be set according to needs. The flowchart of dynamic threshold adjustment is as follows Figure 3 shown.

[0071] (7) Mode judgment

[0072] According to SNR est Make mode judgment and update mode number u:

[0073]

[0074] The flow chart of mode judgment is as follows Figure 4 shown.

[0075] (8) Mode feedback

[0076] The determined mode sequence number u is transmitted back to the transmitting end through the feedback channel. The present invention does not limit the specific feedback method.

[0077] (9) Update sending parameters

[0078] The sender selects the corresponding parameter {L according to the received mode sequence number u u , R u , Γ u} to send data.

[0079] In this example, three candidate modulation and coding schemes are set, as shown in Table 1.

[0080] Table 1 Candidate modulation and coding schemes in the implementation example

[0081] <![CDATA[LDPC code length L n > <![CDATA[LDPC code rate R n > <![CDATA[Modulation constellation diagram Γ n > Mode 1 (n=1) 64800 1 / 4 8PSK Mode 2 (n=2) 64800 1 / 3 QPSK Mode 3 (n=3) 64800 1 / 2 QPSK

[0082] In this example, it is assumed that the expected bit error rate is 1e-4. In order to determine the signal-to-noise ratio thresholds of the three modes, their bit error rate versus signal-to-noise ratio curves are simulated under a Gaussian channel. The results are as follows: Figure 6 , Figure 7 , Figure 8 As shown (the dotted line in the figure is the expected bit error rate).

[0083] According to the simulation results, the signal-to-noise ratio threshold of mode 1 is set to The signal-to-noise ratio threshold for mode 2 is set to The signal-to-noise ratio threshold for mode 3 is set to

[0084] In order to determine the failure ratio thresholds of the three modes of the check equation, the relationship curves between their bit error rate and the failure ratio of the check equation are simulated under the Gaussian channel. The results are as follows: Fig. 9 , Fig.10 , Fig.11 As shown (the dotted line in the figure is the expected bit error rate).

[0085] According to the simulation results, the failure ratio threshold of the verification equation of mode 1 is set to The failure ratio threshold of the verification equation for mode 2 is set to The failure ratio threshold of the verification equation for mode 3 is set to

[0086] In order to illustrate that the threshold of the failure ratio of the check equation is independent of the channel, that is, it is consistent under different channels, the relationship curve between the bit error rate and the failure ratio of the check equation is simulated for mode 1 in 1, 2, and 3 path channels. The simulation results are shown in Fig.12 shown.

[0087] It can be seen that the bit error rate of mode 1 in 1-, 2-, and 3-path channels is consistent with the failure ratio of the check equation, and is Fig. 9 The simulation results confirm that the failure ratio threshold of the checksum equation is consistent for different channels, so it is reliable to use it to determine whether the bit error rate reaches the expected bit error rate. In contrast, it is unreliable to use a fixed signal-to-noise ratio threshold to determine whether the bit error rate reaches the expected bit error rate, because the relationship between the bit error rate and the signal-to-noise ratio will change with the channel (see Figure 2 ), so the signal-to-noise ratio threshold in different channels is different.

[0088] In this example, it is assumed that the interaction process between the sending and receiving ends is as follows Fig.13 As shown in the figure. During the interaction, each time the receiving end selects an adaptive mode based on the received data, it sends the selected mode number in the signaling back to the transmitting end. After the transmitting end demodulates the sequence number contained in the signaling, the next transmission uses the mode corresponding to the sequence number for modulation and coding and data transmission.

[0089] Next, set the initial mode of the system to Ω u =Ω 1 , the adaptive mode selection process of this method and the convergence process of each mode threshold are simulated under a 2-path channel.

[0090] In order to determine the correctness of the convergence of each mode threshold, we first determine through simulation that the actual signal-to-noise ratio thresholds of mode 1, mode 2, and mode 3 in the 2-path channel are 13.8dB, 15.8dB, and 18.8dB respectively. The simulation results are shown in Fig.14 shown.

[0091] Regarding the channel signal-to-noise ratio variation process, two cases are considered.

[0092] In the first case, the signal-to-noise ratio changes linearly from 13dB to 21dB. The simulation results in this case are as follows: Fig.15 and Fig.16 Each sample point in the figure represents a data transmission from the sender to the receiver. Fig.15 The vertical axis in represents the estimated signal-to-noise ratio of the receiving end in this transmission. Fig.16 The vertical axis in represents the SNR threshold adjustment result of each mode at the receiving end after the transmission (the horizontal axis is 0, which represents the initial state). In the simulation, it is assumed that the SNR estimation is error-free and the relationship between the SNR and the BER is in accordance with Fig.14 In the simulation curve, the relationship between the bit error rate and the failure ratio of the verification equation is consistent with Figures 9 to 11 Simulation curve. The signal-to-noise ratio margin δ when the threshold is increased in the simulation up Set to 0.1dB.

[0093] Depend on Fig.16 It can be seen that although at the initial sampling point, the threshold settings of the three modes (8.4dB, 10.2dB, 12dB) do not match the actual thresholds (14.2dB, 15.8dB, 18.8dB) corresponding to the current channel, as the communication process proceeds, the thresholds of the three modes are gradually adjusted to near the actual thresholds. Fig.15 It can be seen that mode misjudgment occurs only in the threshold adjustment process at the beginning of the communication process. After the threshold converges to the true value (about after the 20th sample point), all mode decisions are correct, indicating that the algorithm can adapt to the channel and demonstrate the effectiveness and robustness of this method.

[0094] In the second case, the signal-to-noise ratio varies randomly between 10dB and 25dB. The simulation results in this case are as follows: Fig.17 and Fig.18 The signal-to-noise margin δ when the threshold is increased in the simulation up Set to 0.1dB.

[0095] Depend on Fig.18 It can be seen that although at the initial sampling point, the threshold settings of the three modes (8.4dB, 10.2dB, 12dB) do not match the actual thresholds (14.2dB, 15.8dB, 18.8dB) corresponding to the current channel, as the communication process proceeds, the thresholds of the three modes are gradually adjusted to near the actual thresholds. Fig.17 It can be seen that mode misjudgment only occurs during the threshold adjustment process at the beginning of the communication process. After the threshold converges to the vicinity of the true value (approximately after the 29th sample point), all mode decisions are correct, indicating that the algorithm can adapt to the channel and demonstrate the effectiveness and robustness of this method.

Claims

1. An adaptive modulation and coding method based on LDPC code, characterized in that: The process includes: (1) Set N modulation coding modes as needed, denoted as Ω1, Ω2, …, Ω N , as the selection range of the adaptive modulation and coding algorithm, each modulation and coding mode Ω n Corresponding to a set of parameters {LDPC code length L n , LDPC code rate R n , symbol mapping constellation diagram Γ n }, n = 1, 2, ..., N; (2) Modulation coding mode Ω n ,n=1,2,…,N, set the signal-to-noise ratio threshold And the LDPC decoding verification equation failure ratio threshold And the modulation coding modes are arranged in order from small to large according to the signal-to-noise ratio threshold; (3) The modulation and coding mode currently used by the system is recorded as Ω u ,u∈{1,2,…,N}, and the corresponding parameters are recorded as When the system starts, u=1 is set, that is, the sender uses the mode Ω1 with the lowest signal-to-noise ratio threshold for data transmission; (4) The receiving end estimates the signal-to-noise ratio based on the received signal, and the estimated result is recorded as SNR est , and perform demodulation and LDPC decoding, and record the decoded bit vector as For L u ×1-dimensional column vector; (5) Based on Calculate the failure rate r of the verification equation: Among them, H u is the modulation coding mode Ω u The corresponding LDPC code supervision matrix, H u For L u (1-R u )×L u dimensional matrix, ||·||0 represents the zero norm of the orientation quantity, and the calculation All addition operations are modulo 2 additions; (6) Estimation result based on signal-to-noise ratio SNR est The sum of the failure ratio r of the check equation is the modulation coding mode Ω u The signal-to-noise ratio threshold Make adjustments: Among them, δ up It is the signal-to-noise margin when the threshold is raised; (7) Estimation result based on signal-to-noise ratio SNR est Make mode judgment and update mode number u: (8) The determined mode sequence number u is transmitted back to the transmitting end through the feedback channel; (9) The sender selects the corresponding parameter {L according to the received mode sequence number u u , R u , Γ u } to send data.

Images