Short-wave communication equalization method and system based on decision feedback iteration

Abstract

The invention discloses a short-wave communication equalization method and system based on decision feedback iteration, and the method specifically comprises the steps: enabling a transmitting end to carry out the coding, interweaving and mapping of data information, generating a plurality of unknown data blocks, enabling known data sequences to be inserted between the unknown data blocks at intervals, forming a data segment sequence, and transmitting the data segment sequence to a receiving end through a channel; a receiving end extracts known data sub-segments and sends the known data sub-segments to a channel estimation module to estimate channel impulse response, the known data sub-segments and unknown data sub-segments extracted by the receiving end are sent to a first equalizer at the same time for segment-by-segment equalization, and then LLR calculation, de-interleaving and decoding are carried out; judging whether the decoding result meets an iteration ending condition or not, and if yes, ending iteration; and otherwise, coding, interleaving and mapping a decoding result, dividing a mapping result into mapping sub-segments, inputting the mapping sub-segments as initial values of iteration into a second equalizer for segment-by-segment equalization, and then performing LLR calculation, de-interleaving and decoding processes until iteration is finished. The short-wave communication receiving performance is improved, the operand is low, and the numerical calculation stability is good.

Description

Technical Field

[0001] The invention relates to the technical field of shortwave communication, and in particular to a shortwave communication equalization method and system based on decision feedback iteration. Background Art

[0002] The shortwave channel has multipath and time-varying characteristics, which can easily cause crosstalk between symbols and signal envelope changes over time, seriously affecting the communication quality. The use of equalization technology can reduce the impact of multipath and time-varying channels on signals to improve the quality of shortwave communications. With the increasing demand for broadband and high-speed data transmission, equalizers are not only required to better overcome the impact of channels, but also require the equalization algorithm to have low computational complexity.

[0003] Appendix D of the US military standard MIL-STD-188-110D defines shortwave broadband waveforms and specifies performance indicators. Existing literature has proposed a variety of equalization technologies for broadband high-speed waveforms, such as Turbo equalization technology, iterative bidirectional Kalman-DFE equalization technology, etc. The basic principle of Turbo equalization technology is similar to that of Turbo decoder, which is to repeatedly exchange soft information between equalization and decoding, and only make hard decisions on the information after the iteration. As the number of iterations increases, the performance improves, but it does not meet the performance indicator requirements, and the algorithm is complex. Iterative bidirectional Kalman-DFE equalization technology studies the equalization of high-speed waveforms under 3kHz bandwidth. By selecting the optimal equalization result that eliminates the crosstalk of the previous or next symbols, and using iterative technology, the performance is significantly improved, but it cannot meet the performance indicator requirements. Summary of the invention

[0004] The object of the present invention is to provide a shortwave communication equalization method and system based on decision feedback iteration with stable receiving performance, simple algorithm, low calculation amount and good numerical calculation stability.

[0005] The technical solution to achieve the purpose of the present invention is: a shortwave communication equalization method based on decision feedback iteration, comprising the following steps:

[0006] Step 1: The transmitter encodes, interleaves and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals to form a data segment sequence, and sends it to the receiver through the channel;

[0007] Step 2: The receiving end extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the first equalizer; extracts unknown data sub-segments segment by segment and sends them to the first equalizer for segment-by-segment equalization;

[0008] Step 3, saving the channel estimation value, unknown data sub-segment and equalization result output by the first equalizer segment by segment;

[0009] Step 4: The demapping module performs LLR demapping calculation on the equalization result, and deinterleaves and decodes the calculated LLR value. After the decoding is completed, the decoding result is sent to the iteration end judgment module;

[0010] Step 5: The iteration end determination module determines whether the decoding meets the iteration end condition according to the set criteria. If it meets the condition, the iteration ends and the decoding result is output; otherwise, the decoding result is sent to the encoding module of the iteration receiving unit;

[0011] Step 6: The decoding result is encoded, interleaved, and mapped in the iterative receiving unit. The mapping result is divided into mapping sub-segments. The mapping sub-segments are used as the initial values ​​of the iteration. The saved channel estimation values ​​and unknown data sub-segments and mapping sub-segments of the first equalizer input are input to the second equalizer segment by segment. Go to step 4 to perform the next LLR calculation and decoding process until the iteration end condition is reached.

[0012] Furthermore, the waveform frame of the data information is composed of three segments: a TLC segment, a synchronization segment and a data segment, wherein the TLC segment is used for level adjustment at the transmitting end and AGC at the receiving end; the synchronization segment is used for synchronization search, frequency offset compensation and waveform number transmission at the receiving end; and the data segment carries the data information to be transmitted;

[0013] The data segment is composed of known data sub-segments and unknown data sub-segments alternately. The known data sub-segments are used for estimating and tracking the channel state. The lengths of the known data sub-segments and the unknown data sub-segments vary with the bandwidth and the rate, and the modulation method of the unknown data varies with the rate.

[0014] Furthermore, the transmitting end in step 1 encodes, interleaves, and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals, forms a data segment sequence, and sends it to the receiving end through the channel, as follows:

[0015] Step 1.1: The sender encodes, interleaves, and maps the data information to generate an unknown data sequence. , divided into multiple unknown data blocks of length N according to the waveform parameters ;

[0016] Step 1.2, insert the known data sequence into the unknown data blocks to form a data segment sequence;

[0017] Step 1.3: Frame the TLC segment sequence, synchronization segment sequence and data segment sequence to form a transmission sequence , through the channel Send to the receiving end.

[0018] Furthermore, in step 1.3, the TLC segment sequence, the synchronization segment sequence and the data segment sequence are framed to form a transmission sequence. , through the channel Sent to the receiving end, as follows:

[0019] For an unknown data block , , through a time-varying channel , , the channel introduces additive white Gaussian noise , , received an unknown data block , , then the linear equations are expressed as:

[0020] , , , (1)

[0021] Will , , Written in vector form:

[0022]

[0023]

[0024]

[0025] Will Written The matrix form is:

[0026]

[0027] Then the equation system of formula (1) can be written as a matrix equation:

[0028] (2)

[0029] Therefore, the problem of balancing unknown data blocks is to find the optimal solution of the matrix equation ; In the presence of noise, the error The smallest sum of squares that is The best estimate of , so the least squares algorithm is used to find the best estimate , ignoring the noise vector , Constructed by the channel estimation value, the matrix equation of formula (2) is simplified to , The least squares solution of is:

[0030] (3)

[0031] in express The conjugate transpose of , express The inverse matrix of R is the autocorrelation matrix of the channel, which is a Hermitian matrix. , then the least squares solution is further expressed as .

[0032] Furthermore, the receiving end described in step 2 extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the first equalizer; extracts unknown data sub-segments segment by segment and sends them to the first equalizer for segment-by-segment equalization, as follows:

[0033] Step 2.1: Use Jacobi algorithm to solve the matrix equation and modify the algorithm to convert the matrix Rewrite it as follows:

[0034] (4)

[0035] in is a diagonal matrix; , , , , , ; Is The lower triangular matrix consists of the elements to the lower left of the main diagonal of ; Depend on The upper triangular matrix is ​​composed of the elements to the upper right of the main diagonal of The inverse of is directly achieved by finding the reciprocal of the main diagonal elements;

[0036] Step 2.2: , and get an estimate of X , perform hard decision element by element to get the initial estimate ,in is the value on the constellation diagram of the corresponding modulation mode;

[0037] Step 2.3, iterative solution, the initial value Substitute into the following formula to obtain the new sign estimate :

[0038] , (5)

[0039] in is the maximum number of iterations;

[0040] right Perform element-by-element judgment to obtain a new judgment value As the initial value for the next iteration;

[0041] Step 2.4: Compare the decision values ​​of two adjacent iterations , get the proportion of the number of different decision value elements, until the proportion is less than the threshold value Th, or the number of iterations reaches the set maximum number M, the iteration is terminated; otherwise, return to step 2.3 to start a new round of iteration process;

[0042] Step 2.5: Take the final estimate As the output of the first equalizer, it is sent to the subsequent demapping module.

[0043] Furthermore, in step 5, the iteration end determination module determines whether the decoding result meets the iteration end condition according to the set criteria, which is as follows:

[0044] The criterion for determining the end of iteration is as follows: the iteration does not end when the decoding result is received for the first time. Starting from the second time the decoding result is received, the difference ratio of the results of two adjacent decodings is compared with the threshold and the maximum number of iterations is set as the criterion for determining the end of iteration. When the maximum number of iterations is reached or the difference ratio of the results of two adjacent decodings is less than the set threshold, the iteration process is exited, the decoding result is output, and the reception of the sent data is completed.

[0045] Furthermore, the initial value of the second equalizer iteration in step 6 is The information after decoding is re-encoded, interleaved, and mapped. The maximum number of iterations is M = 1. The output value of the second equalizer is .

[0046] A shortwave communication equalization system based on decision feedback iteration, the system is used to implement the shortwave communication equalization method based on decision feedback iteration, the system includes a transmitting end module, a receiving end module, a storage module, a demapping module, an iteration result determination module and an iteration receiving unit, wherein:

[0047] The sending end module encodes, interleaves and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals to form a data segment sequence, and sends it to the receiving end through the channel;

[0048] The receiving end module extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the first equalizer; extracts unknown data sub-segments segment by segment and sends them to the first equalizer for segment-by-segment equalization;

[0049] A storage module, storing the channel estimation value input by the first equalizer, the unknown data sub-segment and the equalization result output by the first equalizer segment by segment;

[0050] The demapping module performs LLR demapping calculation on the equalization result, deinterleaves and decodes the calculated LLR value, and outputs the decoding result to the iteration end judgment module after the decoding is completed;

[0051] The iteration end judgment module judges whether the decoding meets the iteration end condition according to the set criteria. If it meets the condition, the iteration ends and the decoding result is output; otherwise, the decoding result is sent to the iteration receiving unit;

[0052] The decoding result is encoded, interleaved and mapped in the iterative receiving unit, and the mapping result is divided into mapping sub-segments. The mapping sub-segments are used as the initial values ​​of the iteration, and the channel estimation values ​​and unknown data sub-segments and mapping sub-segments of the first equalizer input are saved and input into the second equalizer segment by segment, and the mapping module is transferred to perform the next LLR calculation and decoding process until the iteration end condition is reached.

[0053] A mobile terminal comprises a memory, a processor and a computer program stored in the memory and operable on the processor. When the processor executes the program, the shortwave communication equalization method based on decision feedback iteration is implemented.

[0054] A computer-readable storage medium stores a computer program, which implements the steps of the shortwave communication equalization method based on decision feedback iteration when executed by a processor.

[0055] Compared with the prior art, the present invention has the following significant advantages: (1) by simultaneously eliminating the crosstalk between the front and rear symbols and combining iterative technology, shortwave communication equalization is achieved, the influence of multipath and channel fading on the shortwave broadband waveform is solved, and the shortwave communication receiving performance is improved; (2) there is no need to perform matrix inversion, the amount of calculation is low, and the numerical calculation stability is good. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 The invention is a flowchart of a shortwave communication equalization method based on decision feedback iteration.

[0057] Figure 2 It is a structural schematic diagram of the data information waveform frame in the present invention.

[0058] Figure 3It is a bit error rate curve diagram under the AWGN channel in an embodiment of the present invention.

[0059] Figure 4 4 is a bit error rate curve diagram under a Poor channel in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0061] In order to solve the problem that the receiving performance of shortwave broadband waveform is unstable due to the influence of multipath and channel fading, and most shortwave broadband high-speed waveform equalization technology algorithms are complex and have poor performance, the present invention is aimed at MIL-STD-188-110D Appendix D single-carrier continuous frequency band data waveform, and invents a decision feedback iterative equalization technology. By simultaneously eliminating the crosstalk of the front and rear symbols and combining the iterative technology, the performance is better than the index requirements specified in Appendix D.

[0062] MIL-STD-188-110D Appendix D defines a single-carrier continuous-band data waveform group, with bandwidths increasing in 3kHz increments from 3kHz to 24kHz. The main purpose of this waveform design is to provide high-speed, long-distance communications through shortwave bands under extreme conditions. The maximum communication rate is 120kpbs. The modulation modes are: Walsh, PSK, and QAM; 4 interleaving lengths are defined for each rate; the error correction code uses (2,1,7) or (2,1,9) convolutional codes. Figure 2 As shown in FIG. 1 , the waveform frame of the data information consists of three segments: a TLC segment, a synchronization segment, and a data segment. The TLC segment is used for level adjustment at the transmitting end and AGC at the receiving end; the synchronization segment is used for synchronization search, frequency offset compensation, and waveform number transmission at the receiving end; and the data segment carries the data information to be transmitted.

[0063] The data segment is composed of known data sub-segments and unknown data sub-segments alternately. The known data sub-segments are used for estimating and tracking the channel state. The lengths of the known data sub-segments and the unknown data sub-segments vary with the bandwidth and the rate, and the modulation method of the unknown data varies with the rate.

[0064] Combination Figure 1 The present invention provides a shortwave communication equalization method based on decision feedback iteration, comprising the following steps:

[0065] Step 1: The transmitter encodes, interleaves and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals to form a data segment sequence, and sends it to the receiver through the channel;

[0066] Step 2: The receiving end extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the equalizer 1; extracts unknown data sub-segments segment by segment and sends them to the equalizer 1 for segment-by-segment equalization;

[0067] Step 3, saving the channel estimation value, unknown data sub-segment and equalization result output by the equalizer 1 segment by segment;

[0068] Step 4: The LLR calculation module performs demapping calculation on the equalization result, and deinterleaves and decodes the calculated LLR value. After the decoding is completed, the decoding result is sent to the iteration end judgment module;

[0069] Step 5: The iteration end determination module determines whether the decoding result meets the iteration end condition according to the set criteria. If it meets the condition, the iteration ends and the decoding result is output; otherwise, the decoding result is sent to the encoding module of the iteration receiving unit;

[0070] Step 6: The decoding result is encoded, interleaved, and mapped in the iterative receiving unit. The mapping result is divided into mapping sub-segments. The mapping sub-segments are used as the initial values ​​of the iteration. The saved channel estimation values ​​and unknown data sub-segments and mapping sub-segments of the equalizer 1 input are input to the equalizer 2 segment by segment. Go to step 4 to calculate the LLR and the decoding process for the next time until the iteration end condition is reached.

[0071] As a specific example, the transmitting end in step 1 encodes, interleaves, and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals, forms a data segment sequence, and sends it to the receiving end through the channel, as follows:

[0072] Step 1.1: The sender encodes, interleaves, and maps the data information to generate an unknown data sequence. , divided into multiple unknown data blocks of length N according to the waveform parameters ;

[0073] Step 1.2, insert the known data sequence into the unknown data blocks to form a data segment sequence;

[0074] Step 1.3: Frame the TLC segment sequence, synchronization segment sequence and data segment sequence to form a transmission sequence , through the channel Send to the receiving end.

[0075] As a specific example, the TLC segment sequence, synchronization segment sequence and data segment sequence are framed to form a transmission sequence as described in step 1.3. , through the channel Sent to the receiving end, as follows:

[0076] For an unknown data block , , through a time-varying channel , , the channel introduces additive white Gaussian noise , , received an unknown data block , , then the relationship between them can be expressed by a system of linear equations:

[0077] , , , (1)

[0078] Will , , Written in vector form:

[0079]

[0080]

[0081]

[0082] Will Written The matrix form is:

[0083]

[0084] Then the equation system of formula (1) can be written as a matrix equation:

[0085] (2)

[0086] Therefore, the problem of balancing unknown data blocks is to find the optimal solution of the matrix equation ; In the presence of noise, the error The smallest sum of squares that is The best estimate of , so the least squares algorithm is used to find the best estimate , ignoring the noise vector , Constructed by the channel estimation value, the matrix equation of formula (2) is simplified to , The least squares solution of is:

[0087] (3)

[0088] in express The conjugate transpose of , express The inverse matrix of R is the autocorrelation matrix of the channel, which is a Hermitian matrix. , then the least squares solution can be further expressed as ; This solution involves matrix inversion, which requires a lot of computation, so an iterative algorithm with low computation and stable numerical calculation is selected.

[0089] As a specific example, the receiving end described in step 2 extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the equalizer 1; extracts unknown data sub-segments segment by segment and sends them to the equalizer 1 for segment-by-segment equalization, as follows:

[0090] Step 2.1: Use Jacobi algorithm to solve the matrix equation and modify the algorithm to convert the matrix Rewrite it as follows:

[0091] (4)

[0092] in is a diagonal matrix; , , , , , ; Is The lower triangular matrix consists of the elements to the lower left of the main diagonal; Depend on The upper triangular matrix is ​​composed of the elements to the upper right of the main diagonal of The inverse of is directly achieved by finding the reciprocal of the main diagonal elements;

[0093] Step 2.2: , and get an estimate of X , perform hard decision element by element to get the initial estimate ,in is the value on the constellation diagram of the corresponding modulation mode;

[0094] Step 2.3, iterative solution, the initial value Substitute into the following formula to obtain the new sign estimate :

[0095] , (5)

[0096] in is the maximum number of iterations;

[0097] right Perform element-by-element judgment to obtain a new judgment value As the initial value for the next iteration;

[0098] Step 2.4: Compare the estimated values ​​of two adjacent iterations , get the proportion of the number of different estimated value elements, until the proportion is less than a certain threshold value Th, or the number of iterations reaches the set maximum number M, the iteration is terminated; otherwise, return to step 2.3 to start a new round of iteration process;

[0099] Step 2.5: Take the final estimate As the output of equalizer 1, it is sent to the subsequent demapping module.

[0100] As a specific example, the iteration end determination module in step 5 determines whether the decoding satisfies the iteration end condition according to a set criterion, which is as follows:

[0101] The criterion for determining the end of iteration is as follows: the iteration does not end when the decoding result is received for the first time. Starting from the second time the decoding result is received, the difference ratio of the results of two adjacent decodings is compared with the threshold and the maximum number of iterations is set as the criterion for determining the end of iteration. When the maximum number of iterations is reached or the difference ratio of the results of two adjacent decodings is less than the set threshold, the iteration process is exited, the decoding result is output, and the reception of the sent data is completed.

[0102] As a specific example, the initial value of the equalizer 2 iteration in step 6 is The information after decoding is re-encoded, interleaved, and mapped. The maximum number of iterations is M = 1. The output value of equalizer 2 is .

[0103] The present invention also provides a shortwave communication equalization system based on decision feedback iteration, which is used to implement the shortwave communication equalization method based on decision feedback iteration. The system includes a transmitting end module, a receiving end module, a storage module, a demapping module, an iteration result determination module and an iteration receiving unit, wherein:

[0104] The sending end module encodes, interleaves and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals to form a data segment sequence, and sends it to the receiving end through the channel;

[0105] The receiving end module extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the first equalizer; extracts unknown data sub-segments segment by segment and sends them to the first equalizer for segment-by-segment equalization;

[0106] A storage module, storing the channel estimation value input by the first equalizer, the unknown data sub-segment and the equalization result output by the first equalizer segment by segment;

[0107] The demapping module performs LLR demapping calculation on the equalization result, deinterleaves and decodes the calculated LLR value, and sends the decoding result to the iteration end judgment module after the decoding is completed;

[0108] The iteration end judgment module judges whether the decoding result meets the iteration end condition according to the set criteria. If it meets the condition, the iteration ends and the decoding result is output; otherwise, the decoding result is sent to the iteration receiving unit;

[0109] The decoding result is encoded, interleaved and mapped in the iterative receiving unit, and the mapping result is divided into mapping sub-segments. The mapping sub-segments are used as the initial values ​​of the iteration, and the channel estimation values ​​and unknown data sub-segments and mapping sub-segments of the first equalizer input are saved and input into the second equalizer segment by segment, and the mapping module is transferred to perform the next LLR calculation and decoding process until the iteration end condition is reached.

[0110] The present invention also provides a mobile terminal, comprising a memory, a processor and a computer program stored in the memory and operable on the processor, wherein the processor implements the shortwave communication equalization method based on decision feedback iteration when executing the program.

[0111] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the shortwave communication equalization method based on decision feedback iteration are implemented.

[0112] The present invention is further described in detail below with reference to specific embodiments.

[0113] Example

[0114] This embodiment uses the bandwidth BW=24kHz, waveform number wid=7, long interleaving, (2,1,7) convolutional code in Appendix D to simulate and test the performance of iterative reception. The modulation mode of this waveform is 8PSK, and the demapping uses an accurate algorithm to calculate LLR. The simulation channel uses the AWGN channel and the mid-latitude disturbance channel of ITU-R.F1487, referred to as the Poor channel. The channel parameters are: multipath delay 2ms, Doppler spread 1hz. The performance indicators of this waveform are: when the signal-to-noise ratio is 13dB under the AWGN channel and the signal-to-noise ratio is 19dB under the Poor channel, the bit error rate is not greater than .

[0115] Figure 3 The simulation results under AWGN channel are given. It can be seen that the bit error rate is less than 11dB. , 2dB better than the performance in Appendix D, and iteration did not improve the performance.

[0116] Figure 4 The simulation results under the Poor channel are given. Without iteration, when the signal-to-noise ratio is 19dB, the bit error rate is around 1E-4. After 1 iteration, when the signal-to-noise ratio is 16dB, the bit error rate is lower than , which is 3dB better than the performance in Appendix D, 2 iterations are 1.5dB better than 1 iteration, and the performance is basically not improved after 2 iterations or more. It can be seen that the maximum number of iterations can be set to 2.

[0117] This embodiment uses decision feedback iterative equalization technology to iteratively receive the waveform defined in Appendix D of MIL-STD-188-110D, effectively eliminating the inter-symbol interference caused by fading and multipath during the transmission of shortwave signals. This method does not require matrix inversion, has a low amount of calculation, and has good numerical calculation stability. The simulation results are better than the performance indicators specified in Appendix D, indicating that this technology has good reception performance in the equalization reception of shortwave waveforms.

[0118] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A shortwave communication equalization method based on decision feedback iteration, characterized in that: The following steps are involved: Step 1: The transmitter encodes, interleaves and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals to form a data segment sequence, and sends it to the receiver through the channel; Step 2: The receiving end extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the first equalizer; extracts unknown data sub-segments segment by segment and sends them to the first equalizer for segment-by-segment equalization; Step 3, saving the channel estimation value, unknown data sub-segment and equalization result output by the first equalizer segment by segment; Step 4: The LLR calculation module demaps the equalization result, deinterleaves and decodes the calculated LLR value, and sends the decoding result to the iteration end judgment module after the decoding is completed; Step 5: The iteration end determination module determines whether the decoding result meets the iteration end condition according to the set criteria. If it does, the iteration ends and the decoding result is output; Otherwise, the decoding result is sent to the encoding module of the iterative receiving unit; Step 6: The decoding result is encoded, interleaved, and mapped in the iterative receiving unit, and the mapping result is divided into mapping sub-segments. The mapping sub-segments are used as the initial values ​​of the iteration, and the saved channel estimation values ​​and unknown data sub-segments and mapping sub-segments of the first equalizer input are input to the second equalizer segment by segment. Go to step 4 to perform the next LLR calculation, deinterleaving, and decoding process until the iteration end condition is reached.

2. The shortwave communication equalization method based on decision feedback iteration according to claim 1 is characterized in that: The waveform frame of the data information consists of three segments: a TLC segment, a synchronization segment and a data segment, wherein the TLC segment is used for level adjustment at the transmitting end and AGC at the receiving end; the synchronization segment is used for synchronization search, frequency offset compensation and waveform number transmission at the receiving end; and the data segment carries the data information to be transmitted; The data segment is composed of known data sub-segments and unknown data sub-segments alternately. The known data sub-segments are used for estimating and tracking the channel state. The lengths of the known data sub-segments and the unknown data sub-segments vary with the bandwidth and the rate, and the modulation method of the unknown data varies with the rate.

3. The shortwave communication equalization method based on decision feedback iteration according to claim 1 is characterized in that: The transmitting end in step 1 encodes, interleaves and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals to form a data segment sequence, and sends it to the receiving end through the channel, as follows: Step 1.1: The sender encodes, interleaves, and maps the data information to generate an unknown data sequence. , divided into multiple unknown data blocks of length N according to the waveform parameters ; Step 1.2, insert the known data sequence into the unknown data blocks to form a data segment sequence; Step 1.3: Frame the TLC segment sequence, synchronization segment sequence and data segment sequence to form a transmission sequence , through the channel Send to the receiving end.

4. The shortwave communication equalization method based on decision feedback iteration according to claim 3 is characterized in that: Step 1.3: The TLC segment sequence, synchronization segment sequence and data segment sequence are framed to form a transmission sequence. , through the channel Sent to the receiving end, as follows: For an unknown data block , , through a time-varying channel , , the channel introduces additive white Gaussian noise , , received an unknown data block , , then the linear equations are expressed as: ， ， ， （1） Will , , Written in vector form: ； ； ； Will Written The matrix form is: ； Then the equation system of formula (1) can be written as a matrix equation: （2） Therefore, the problem of balancing unknown data blocks is to find the optimal solution of the matrix equation ; In the presence of noise, the error The smallest sum of squares that is The best estimate of , so the least squares algorithm is used to find the best estimate , ignoring the noise vector , Constructed by the channel estimation value, the matrix equation of formula (2) is simplified to , The least squares solution of is: （3） in express The conjugate transpose of , express The inverse matrix of R is the autocorrelation matrix of the channel, which is a Hermitian matrix. , then the least squares solution is further expressed as .

5. The shortwave communication equalization method based on decision feedback iteration according to claim 1 is characterized in that: The receiving end described in step 2 extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the first equalizer; extracts unknown data sub-segments segment by segment and sends them to the first equalizer for segment-by-segment equalization, as follows: Step 2.1: Use Jacobi algorithm to solve the matrix equation and modify the algorithm to convert the matrix Rewrite it as follows: （4） in is a diagonal matrix; , , , , , ; Is The lower triangular matrix consists of the elements to the lower left of the main diagonal; Depend on The upper triangular matrix is ​​composed of the elements to the upper right of the main diagonal of The inverse of is directly achieved by finding the reciprocal of the main diagonal elements; Step 2.2: , and get an estimate of X , perform hard decision element by element to get the initial estimate ,in is the value on the constellation diagram of the corresponding modulation mode; Step 2.3, iterative solution, the initial value Substitute into the following formula to obtain the new sign estimate : ， （5） in is the maximum number of iterations; right Perform element-by-element judgment to obtain a new judgment value As the initial value for the next iteration; Step 2.4: Compare the decision values ​​of two adjacent iterations , obtain the proportion of the number of different decision value elements, until the proportion is less than the threshold value Th, or the number of iterations reaches the set maximum number M, the iteration is terminated; Otherwise, return to step 2.3 to start a new round of iteration process; Step 2.5: Take the final estimate As the output of the first equalizer, it is sent to the subsequent demapping module.

6. The shortwave communication equalization method based on decision feedback iteration according to claim 1, characterized in that: In step 5, the iteration end determination module determines whether the decoding result meets the iteration end condition according to the set criteria, which is as follows: The criterion for determining the end of iteration is as follows: the iteration does not end when the decoding result is received for the first time. Starting from the second time the decoding result is received, the difference ratio of the results of two adjacent decodings is compared with the threshold and the maximum number of iterations is set as the criterion for determining the end of iteration. When the maximum number of iterations is reached or the difference ratio of the results of two adjacent decodings is less than the set threshold, the iteration process is exited, the decoding result is output, and the reception of the sent data is completed.

7. The shortwave communication equalization method based on decision feedback iteration according to claim 1 is characterized in that: Initial value for the second equalizer iteration in step 6 The information after decoding is re-encoded, interleaved, and mapped. The maximum number of iterations is M = 1. The output value of the second equalizer is .

8. A shortwave communication equalization system based on decision feedback iteration, characterized in that: The system is used to implement the shortwave communication equalization method based on decision feedback iteration according to any one of claims 1 to 7, and the system includes a transmitting end module, a receiving end module, a storage module, a demapping module, an iteration end determination module and an iteration receiving unit, wherein: The sending end module encodes, interleaves and maps the data information to generate multiple unknown data blocks, inserts the known data sequence into the unknown data blocks at intervals to form a data segment sequence, and sends it to the receiving end through the channel; The receiving end module extracts known data sub-segments from the received signal segment by segment and sends them to the channel estimation module, estimates the channel impulse response corresponding to the sub-segment, and sends it to the first equalizer; extracts unknown data sub-segments segment by segment and sends them to the first equalizer for segment-by-segment equalization; A storage module, storing the channel estimation value input by the first equalizer, the unknown data sub-segment and the equalization result output by the first equalizer segment by segment; The demapping module performs LLR demapping calculation on the equalization result, deinterleaves and decodes the calculated LLR value, and outputs the decoded value to the iteration end judgment module after the decoding is completed; The iteration end judgment module judges whether the decoding result meets the iteration end condition according to the set criteria. If it meets the condition, the iteration ends and the decoding result is output; otherwise, the decoding result is sent to the encoding module of the iteration receiving unit; The decoding result is encoded, interleaved and mapped in the iterative receiving unit, and the mapping result is divided into mapping sub-segments. The mapping sub-segments are used as the initial values ​​of the iteration, and the channel estimation values ​​and unknown data sub-segments and mapping sub-segments of the first equalizer input are saved and input into the second equalizer segment by segment, and the mapping module is transferred to perform the next LLR calculation and decoding process until the iteration end condition is reached.

9. A mobile terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the shortwave communication equalization method based on decision feedback iteration as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the shortwave communication equalization method based on decision feedback iteration as described in any one of claims 1 to 7 are implemented.

Images