Incoherent GFSK (Gaussian Frequency Shift Keying) demodulation method and device based on K-best algorithm

Abstract

The invention discloses an incoherent GFSK (Gaussian Frequency Shift Keying) demodulation method based on a K-best algorithm. Demodulation calculation is performed on a GFSK modulation signal based on a multi-symbol maximum likelihood detection method; if the number of branches of the stage of current demodulation calculation is larger than K, the demodulation value of the k stage is calculated, the conjugate relation value is calculated according to the demodulation values of the k-2 stage, the k-1 stage and the k stage and the received symbol value of the (k-1) th stage, K is the preset number of reserved branches of each stage, k is the stage of current demodulation calculation, and k is the number of the received symbols of the (k-1) th stage; the conjugate relation value is a relation value between a (k-1) th level receiving symbol value and a (k-1) th level demodulation value; calculating a similarity value between the (k-1) th level receiving symbol value and the (k-1) th level demodulation value according to the conjugate relation value and the corresponding branch integral plate value; k branch demodulation values with the maximum similarity value are reserved; the demodulation values of the K branches of the next level are calculated according to the reserved demodulation values of the K branches until T-level demodulation calculation is completed, the maximum value is selected from the demodulation values of the K branches of the T level to serve as the final demodulation value, and T is the observation length of the multi-symbol maximum likelihood detection method.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a non-coherent GFSK demodulation method and device based on a K-best algorithm. Background Art

[0002] GFSK (Gaussian Frequency Shift Keying) is a widely used modulation method. Its advantage is constant envelope transmission, which can effectively resist the nonlinear distortion of the RF power amplifier. At the same time, the Gaussian filter brings the effect of high frequency bandwidth utilization. It is widely used in low data rate and low cost personal wireless audio / video transmission systems and modern communication equipment (such as Bluetooth).

[0003] For the demodulation of GFSK modulated signals, traditional technical solutions include coherent demodulation and incoherent demodulation, but these demodulation technologies have some problems, such as:

[0004] Coherent demodulation: The demodulation performance can reach the theoretical limit, but carrier recovery is required, which is more complex and may require preamble code overhead to solve.

[0005] Incoherent demodulation does not require carrier recovery and can achieve optimal performance. It is usually based on the maximum likelihood detection method of multiple symbols, but it is extremely complex, requires a lot of calculations, has poor anti-interference ability, and does not provide LLR (log-likelihood ratio) to the subsequent error correction decoder. Summary of the invention

[0006] Based on the above situation, the main purpose of the present invention is to provide a non-coherent GFSK demodulation method based on the K-best algorithm, which reduces the complexity of the demodulation algorithm, greatly reduces the computing resources, enhances the anti-interference ability of the demodulation algorithm, improves the demodulation performance, and increases the wireless distance while achieving the ultimate performance.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0008] A non-coherent GFSK demodulation method based on K-best algorithm, the method comprising the steps of:

[0009] S10: performing demodulation calculation on the received GFSK modulated signal based on a multi-symbol maximum likelihood detection method;

[0010] S20: If the number of branches of the level where the current demodulation calculation is located is greater than K, the demodulation value of the k-th level is calculated based on the multi-symbol maximum likelihood detection method, and the conjugate relationship value of each branch is calculated according to the demodulation values ​​of each branch of the k-2th level, the k-1th level and the k-th level, and the k-1th level received symbol value, where K is the preset number of branches reserved at each level, k is the level where the current demodulation calculation is located, and the conjugate relationship value is the conjugate relationship value between the k-1th level received symbol value and the k-1th level demodulation value;

[0011] S30: Calculate the similarity value between the k-1th level received symbol value and the k-1th level demodulated value according to the conjugate relationship value and the integral plate value of the corresponding branch;

[0012] S40: retaining the demodulated values ​​of the K branches with the largest similarity values, and deleting the demodulated values ​​of the remaining branches;

[0013] S50: Continue to execute step S20 according to the demodulation values ​​of the retained K branches to calculate the demodulation values ​​of the K branches of the next level until the T-level demodulation calculation is completed, and select the maximum value among the demodulation values ​​of the K branches of the T level as the final demodulation value output, where T is the observation length of the multi-symbol maximum likelihood detection method.

[0014] Preferably, the S20 comprises the steps of:

[0015] S201: performing a convolution operation on the demodulated signals of each branch of the k-2th level, the k-1th level and the kth level with a Gaussian filter;

[0016] S202: Select the intermediate value after convolution, accumulate it and generate a complex form matched filtering candidate value;

[0017] S203: multiply the complex matched filtering candidate value by the phase accumulation complex value of the previous branch, and then conjugate multiply it by the k-1th level received symbol value to obtain the conjugate relationship value.

[0018] Preferably, the S30 is specifically:

[0019] The conjugate relationship value and the integral board value of the corresponding branch are accumulated, and then the absolute value square calculation is performed on the accumulated integral board value to obtain the similarity value.

[0020] Preferably, the method further comprises step S60:

[0021] Demodulation bit probability information is calculated according to the similarity value of the T-th stage reserved branch, and the demodulation bit probability information is provided to the error correction decoder for decoding and error correction.

[0022] Preferably, the K value is a positive integer greater than or equal to 2 and less than or equal to 16.

[0023] The present invention also provides a computer device, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 5.

[0024] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the method according to any one of claims 1 to 5 when executed by a processor.

[0025] The present invention also provides a computer program product, comprising a computer program / instruction, wherein the computer program / instruction implements the steps of the method according to any one of claims 1 to 5 when executed by a processor.

[0026] The present invention also provides a non-coherent GFSK demodulation device based on the K-best algorithm, comprising a demodulation module, a conjugate relationship calculation module and a distance calculation module:

[0027] The demodulation module is used to perform demodulation calculation on the received GFSK modulated signal based on a multi-symbol maximum likelihood detection method;

[0028] The conjugate relationship calculation module is used to calculate the demodulation value of level k based on the multi-symbol maximum likelihood detection method when the number of branches of the level where the current demodulation calculation is located is greater than K, and calculate the conjugate relationship value of each branch according to the demodulation values ​​of each branch of the k-2th level, the k-1th level and the kth level, and the k-1th level received symbol value, wherein K is the preset number of reserved branches at each level, k is the level where the current demodulation calculation is located, and the conjugate relationship value is the conjugate relationship value between the k-1th level received symbol value and the k-1th level demodulation value;

[0029] The distance calculation module is used to calculate the similarity value between the k-1th level received symbol value and the k-1th level demodulation value according to the conjugate relationship value and the integral plate value of the corresponding branch, and retain the demodulation values ​​of the K branches with the largest similarity values, and delete the demodulation values ​​of the remaining branches;

[0030] When the conjugate relationship calculation module calculates to level T, the distance calculation module selects the maximum value from the demodulation values ​​of the K branches at level T as the final demodulation value output, where T is the observation length of the multi-symbol maximum likelihood detection method.

[0031] Preferably, the conjugate relationship calculation module includes a convolution calculation unit, a candidate value calculation unit and a conjugate relationship value calculation unit:

[0032] The convolution calculation unit is used to perform convolution operation on the demodulated signal of each branch of the k-2th level, the k-1th level and the kth level with the Gaussian filter;

[0033] The candidate value calculation unit is used to select the intermediate value after convolution for accumulation and generate a complex form of matched filtering candidate value;

[0034] The conjugate relationship value calculation unit is used to multiply the complex form matched filtering candidate value with the phase accumulation complex value of the previous branch, and then conjugate multiply it with the k-1th level received symbol value to obtain the conjugate relationship value.

[0035] Preferably, the distance calculation module calculates the distance value between the k-1th level received symbol value and the k-1th level demodulated value according to the conjugate relationship value and the integral plate value of the corresponding branch, and specifically:

[0036] The conjugate relationship value and the integral board value of the corresponding branch are accumulated, and then the absolute value square calculation is performed on the accumulated integral board value to obtain the similarity value.

[0037] Preferably, the device further comprises an LLR calculation module:

[0038] The LLR calculation module is used to calculate demodulation bit probability information according to the similarity value of the T-th level reserved branch, and the demodulation bit probability information is used to provide to the error correction decoder for decoding and error correction.

[0039] This scheme adopts K-best algorithm to prune branches, retain K branches with the largest distance values, that is, the most likely branches, and then perform fission on these K retained branches, and continue to prune branches after fission, retaining K most likely branches, and so on, until the demodulation of the observation length T is calculated, and the demodulation value with the largest distance value in the T-level demodulation is selected as the final demodulation value. The incoherent GFSK demodulation method based on K-best algorithm of the present invention reduces the complexity of the demodulation algorithm, greatly reduces the computing resources, enhances the anti-interference ability of the demodulation algorithm, improves the demodulation performance, and increases the wireless distance while achieving the ultimate performance.

[0040] Other beneficial effects of the present invention will be explained in the specific implementation manner through the introduction of specific technical features and technical solutions. Through the introduction of these technical features and technical solutions, those skilled in the art should be able to understand the beneficial technical effects brought about by the technical features and technical solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The preferred implementation of the non-coherent GFSK demodulation method and device based on the K-best algorithm of the present invention will be described below with reference to the accompanying drawings.

[0042] Figure 1 A tree diagram of output signals of a GFSK demodulation method for multi-symbol differential detection in the prior art;

[0043] Figure 2 A flow chart of a non-coherent GFSK demodulation method based on a K-best algorithm according to a preferred embodiment of the present invention;

[0044] Figure 3 A tree diagram of an output signal of the GFSK demodulation method according to the present invention;

[0045] Figure 4 A flow chart of a demodulation process according to a preferred embodiment of the present invention;

[0046] Figure 5 and Figure 6 When K is 4, the flow chart of the GFSK demodulation method of the present invention is as follows;

[0047] Figure 7 The comparison chart of demodulation performance and limit performance for different K values;

[0048] Figure 8 The figure is a block diagram of a non-coherent GFSK demodulation device based on a K-best algorithm according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0049] The present invention is described below based on embodiments, but the present invention is not limited to these embodiments. In the following detailed description of the present invention, some specific details are described in detail. In order to avoid confusing the essence of the present invention, known methods, processes, procedures, and components are not described in detail.

[0050] In addition, persons of ordinary skill in the art will appreciate that the drawings provided herein are for illustration purposes and are not necessarily drawn to scale.

[0051] Unless the context clearly requires otherwise, throughout the specification and claims, the words "include", "comprising" and similar words should be interpreted in an inclusive sense rather than an exclusive or exhaustive sense; that is, in the sense of "including but not limited to".

[0052] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0053] This technical solution proposes a GFSK non-coherent demodulation method based on the K-best algorithm, which can achieve optimal performance (close to coherent demodulation performance) at a very low complexity cost.

[0054] A simple GFSK modulator, the input bit stream can be expressed as (p1, p2, ...), the value is ± 1, each bit is sampled M times, the sampled bit signal stream is mixed with the Gaussian filter g υ After convolution, the instantaneous phase accumulation of the Gaussian filter output is expressed as:

[0055]

[0056] Among them, β is the modulation index and υ is the sampling point.

[0057] Finally, the baseband complex modulated signal is expressed as:

[0058]

[0059] Among them, j is expressed as n is a symbol sequence.

[0060] Assuming that the modulated signal is interfered by additive noise, the signal received by the baseband can be expressed as:

[0061] r k =hs k +z k

[0062] Where h is the channel gain, z k is additive white Gaussian noise, and k is the symbol.

[0063] Generally speaking, the GFSK demodulator based on multi-symbol differential detection in the prior art uses a group of symbols for maximum likelihood detection, and the demodulator output is expressed as:

[0064]

[0065] in,

[0066] (1) T is the observation length, that is, the number of symbols used;

[0067] (2)Θ is the set of all possible matched filters;

[0068] (3) is one possible matched filter signal, which is composed of ±1 of each branch in the figure. For example, in the figure, the first branch s1, s2, s3 are +1, +1, +1 respectively, the second branch s1, s2, s3 are +1, +1, -1, ..., the eighth branch s1, s2, s3 are -1, -1, -1 respectively. These 8 modulation signals will be conjugate multiplied with r1, r2, r3 respectively to calculate the demodulation value of each branch according to the above formula;

[0069] (4) argmax represents a function whose output value is such that Maximum In the above formula, the output value of argmax is is a T×1 vector, which is the demodulated output of length T. For example, The demodulated values ​​of may be the first branches s1, s2, s3.

[0070] Theoretically, as T increases, the performance of the GFSK demodulator with multi-symbol differential detection will improve, but Figure 1 As shown in the tree diagram, this will also cause the computing resources to increase exponentially, which leads to the problems of extremely high complexity, huge amount of calculation, and poor anti-interference ability of the existing incoherent demodulation.

[0071] Figure 2 A non-coherent GFSK demodulation method based on a K-best algorithm according to a preferred embodiment of the present invention comprises the following steps:

[0072] S10: Demodulate the received GFSK modulated signal based on the multi-symbol maximum likelihood detection method, that is, use the algorithm of the GFSK demodulator with multi-symbol differential detection to demodulate the received GFSK modulated signal.

[0073] S20: If the number of branches of the level where the current demodulation calculation is located is greater than K, the demodulation value of the k-th level is calculated based on the multi-symbol maximum likelihood detection method, and the conjugate relationship value of each branch is calculated according to the demodulation values ​​of each branch of the k-2th level, the k-1th level and the k-th level, and the k-1th level received symbol value, where K is the preset number of branches reserved at each level, k is the level where the current demodulation calculation is located, and the conjugate relationship value is the conjugate relationship value between the k-1th level received symbol value and the k-1th level demodulation value; wherein the k-th level received symbol value refers to the received baseband signal, that is, r in the above formula k .

[0074] S30: Calculate the similarity value between the received symbol value of the k-1th level and the demodulated value of the k-1th level according to the conjugate relationship value and the integral plate value of the corresponding branch;

[0075] S40: retaining the demodulated values ​​of the K branches with the largest similarity values, and deleting the demodulated values ​​of the remaining branches;

[0076] S50: Continue to execute step S20 according to the demodulation values ​​of the retained K branches to calculate the demodulation values ​​of the K branches of the next level until the T-level demodulation calculation is completed, and select the maximum value among the demodulation values ​​of the K branches of the T level as the final demodulation value output, where T is the observation length of the multi-symbol maximum likelihood detection method.

[0077] The non-coherent GFSK demodulation method based on the K-best algorithm of the present invention reduces the complexity of the demodulation algorithm, greatly reduces the computing resources, enhances the anti-interference ability of the demodulation algorithm, improves the demodulation performance, and increases the wireless distance while achieving the ultimate performance. Figure 3 As shown, if K in the K-best algorithm is 4, that is, the number of branches retained at each level is 4, then when the tree diagram reaches the third level (i.e., the third symbol), 8 branches have been calculated. This scheme will use the K-best algorithm to prune branches, retaining K branches with the largest similarity values, that is, the most likely branches, and then perform fission on these K retained branches. After fission, continue to prune branches, retain K most likely branches, and so on until the demodulation of the observation length T is calculated, and the demodulation value with the largest similarity value in the T-level demodulation is selected as the final demodulation value. For example, the final demodulation value may be the first branch s1, s2, s3...sK.

[0078] In a preferred embodiment, Figure 4 As shown, step S20 may include the steps of:

[0079] S201: performing a convolution operation on the demodulated signals of each branch of the k-2th level, the k-1th level and the kth level with a Gaussian filter;

[0080] S202: Select the intermediate value after convolution, accumulate it and generate a complex form matched filtering candidate value;

[0081] S203: multiply the complex matched filtering candidate value by the phase accumulation complex value of the previous branch, and then conjugate multiply it by the k-1th level received symbol value to obtain the conjugate relationship value.

[0082] In a preferred implementation, S30 may specifically be: accumulating the conjugate relationship value and the integral board value of the corresponding branch, and then performing absolute value square calculation on the accumulated integral board value to obtain the similarity value.

[0083] Next, taking K being 4 as an example, the technical solution of the present invention is exemplarily described.

[0084] like Figure 5 and 6 As shown, Figure 5 In the tree diagram, select the demodulation values ​​of the 2nd, 3rd, and 4th levels. Convolve with a Gaussian filter and then select the middle value after convolution Accumulate (Accum) and generate the complex form matched filter candidate value Exp(j*), which is then multiplied by the phase accumulation complex value of the previous branch (Previousbranch phase) and then multiplied by the corresponding intermediate received symbol r k-1 Perform conjugate multiplication and finally accumulate with the branch scoreboard.

[0085] like Figure 6 As shown, the absolute value of the scoreboard value of each current branch is squared, and after the similarity value is obtained, a judgment will be made, and the K branches with the largest similarity value (distance value) will be retained, and the rest will be pruned. Among them, since K=4, each node has two scoreboard values, so there are 8 branch scoreboards.

[0086] In a preferred embodiment, the K value may be a positive integer greater than or equal to 2 and less than or equal to 16. Figure 7 This is a comparison chart of demodulation performance and limit performance for different K values. It can be seen from the figure that as K increases, the performance will be infinitely close to the performance limit. When K = 16, the false alarm rate is at 10%, and the gap between the solution and the limit performance is only 0.5dB, which is very close. In the figure, PER is the false alarm rate and SNR is the signal-to-noise ratio.

[0087] In a preferred embodiment, the GFSK demodulation method of the present invention may further include step S60: calculating demodulation bit probability information according to the distance value of the T-th level reserved branch, wherein the demodulation bit probability information is provided to the error correction decoder for decoding and error correction.

[0088] In a specific implementation, the LLR value may be calculated according to the algorithm of the following formula:

[0089]

[0090] Among them, E T It is expressed as the Tth column (the last column) of the matrix E, which is used to record the cumulative distance values ​​in the dendrogram.

[0091] The present invention also discloses a computer device, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to perform the steps of the method of the present invention.

[0092] The present invention also discloses a computer program product, comprising a computer program / instruction, wherein the computer program / instruction implements the steps of the method of the present invention when executed by a processor.

[0093] In addition, the present invention also provides a computer-readable storage medium, such as a chip, a CD, etc., on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method of the present invention are implemented.

[0094] It should be noted that the computer-readable storage medium described in the embodiments of the present disclosure is not limited to the above-mentioned embodiments, and may also be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device.

[0095] The present invention also discloses a non-coherent GFSK demodulation device based on K-best algorithm, such as Figure 8 As shown, it includes a demodulation module, a conjugate relationship calculation module and a distance calculation module.

[0096] The demodulation module is used to perform demodulation calculation on the received GFSK modulated signal based on the multi-symbol maximum likelihood detection method; the conjugate relationship calculation module is used to calculate the k-level demodulation value based on the multi-symbol maximum likelihood detection method when the number of branches in the level of the current demodulation calculation is greater than K, and calculate the conjugate relationship value of each branch according to the demodulation values ​​of each branch of the k-2 level, k-1 level and k level, and the k-1 level received symbol value, wherein K is the preset number of branches reserved at each level, k is the level of the current demodulation calculation, and the conjugate relationship value is the k-2 level. The conjugate relationship value between the received symbol value of the k-1th level and the demodulation value of the k-1th level; the distance calculation module is used to calculate the distance value between the received symbol value of the k-1th level and the demodulation value of the k-1th level according to the conjugate relationship value and the integral plate value of the corresponding branch, and retain the demodulation values ​​of the K branches with the largest distance values, and delete the demodulation values ​​of the remaining branches; when the conjugate relationship calculation module calculates to the Tth level, the distance calculation module selects the maximum value from the demodulation values ​​of the K branches of the Tth level as the final demodulation value output, and T is the observation length of the multi-symbol maximum likelihood detection method.

[0097] In a preferred embodiment, the conjugate relationship calculation module 20 includes a convolution calculation unit, a candidate value calculation unit and a conjugate relationship value calculation unit. The convolution calculation unit is used to perform a convolution operation on the demodulated signal of each branch of the k-2th level, k-1th level and kth level with a Gaussian filter; the candidate value calculation unit is used to select the intermediate value after the convolution for accumulation and generate a complex form matched filter candidate value; the conjugate relationship value calculation unit is used to multiply the complex form matched filter candidate value with the phase accumulation complex value of the previous branch, and then conjugate multiply it with the k-1th level received symbol value to obtain the conjugate relationship value.

[0098] In a preferred embodiment, the distance calculation module calculates the distance value between the k-1th level received symbol value and the k-1th level demodulated value based on the conjugate relationship value and the integral board value of the corresponding branch. Specifically, the conjugate relationship value and the integral board value of the corresponding branch are accumulated, and then the absolute value square calculation is performed on the accumulated integral board value to obtain the similarity value.

[0099] In a preferred embodiment, the device further comprises an LLR calculation module, which is used to calculate demodulation bit probability information according to the similarity value of the T-th level reserved branch, and the demodulation bit probability information is used to provide to the error correction decoder for decoding and error correction.

[0100] It will be appreciated by those skilled in the art that, under the premise of no conflict, the above-mentioned preferred solutions can be freely combined and superimposed. Among them, the flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings, for example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions. The numbering of each step in this article is only for the convenience of description and reference, and is not used to limit the order of execution. The specific execution order is determined by the technology itself, and those skilled in the art can determine various allowable and reasonable orders based on the technology itself.

[0101] It should be noted that the use of step numbers (letters or numbers) to refer to certain specific method steps in the present invention is only for the purpose of convenience and brevity of description, and is by no means intended to limit the order of these method steps by letters or numbers. Those skilled in the art will understand that the order of the relevant method steps should be determined by the technology itself and should not be inappropriately limited by the existence of step numbers. Those skilled in the art can determine various permissible and reasonable step orders based on the technology itself.

[0102] Those skilled in the art will appreciate that, without conflict, the above-mentioned preferred solutions can be freely combined and superimposed.

[0103] It should be understood that the above-mentioned embodiments are merely illustrative and not restrictive. Without departing from the basic principles of the present invention, various obvious or equivalent modifications or substitutions that can be made by those skilled in the art to the above-mentioned details will all be included in the scope of the claims of the present invention.

Claims

1. A non-coherent GFSK demodulation method based on K-best algorithm, characterized in that: The method comprises the steps of: S10: performing demodulation calculation on the received GFSK modulated signal based on a multi-symbol maximum likelihood detection method; S20: If the number of branches of the level where the current demodulation calculation is located is greater than K, the demodulation value of the k-th level is calculated based on the multi-symbol maximum likelihood detection method, and the conjugate relationship value of each branch is calculated according to the demodulation values ​​of each branch of the k-2th level, the k-1th level and the k-th level, and the k-1th level received symbol value, where K is the preset number of branches reserved at each level, k is the level where the current demodulation calculation is located, and the conjugate relationship value is the conjugate relationship value between the k-1th level received symbol value and the k-1th level demodulation value; S30: Calculate the similarity value between the k-1th level received symbol value and the k-1th level demodulated value according to the conjugate relationship value and the integral plate value of the corresponding branch; S40: retaining the demodulated values ​​of the K branches with the largest similarity values, and deleting the demodulated values ​​of the remaining branches; S50: Continue to execute step S20 according to the demodulation values ​​of the retained K branches to calculate the demodulation values ​​of the K branches of the next level until the T-level demodulation calculation is completed, and select the maximum value among the demodulation values ​​of the K branches of the T level as the final demodulation value output, where T is the observation length of the multi-symbol maximum likelihood detection method.

2. The non-coherent GFSK demodulation method based on the K-best algorithm according to claim 1, characterized in that: The S20 comprises the steps of: S201: performing a convolution operation on the demodulated signals of each branch of the k-2th level, the k-1th level and the kth level with a Gaussian filter; S202: Select the intermediate value after convolution, accumulate it and generate a complex form matched filtering candidate value; S203: multiply the complex matched filtering candidate value by the phase accumulation complex value of the previous branch, and then conjugate multiply it by the k-1th level received symbol value to obtain the conjugate relationship value.

3. The non-coherent GFSK demodulation method based on the K-best algorithm according to claim 1, characterized in that: The S30 is specifically: The conjugate relationship value and the integral board value of the corresponding branch are accumulated, and then the absolute value square calculation is performed on the accumulated integral board value to obtain the similarity value.

4. The non-coherent GFSK demodulation method based on the K-best algorithm according to claim 1, characterized in that: The step S60 is also included: Demodulation bit probability information is calculated according to the similarity value of the T-th stage reserved branch, and the demodulation bit probability information is provided to the error correction decoder for decoding and error correction.

5. The non-coherent GFSK demodulation method based on K-best algorithm according to claim 1, characterized in that: The K value is a positive integer greater than or equal to 2 and less than or equal to 16.

6. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

9. A non-coherent GFSK demodulation device based on K-best algorithm, characterized in that: Including demodulation module, conjugate relationship calculation module and distance calculation module: The demodulation module is used to perform demodulation calculation on the received GFSK modulated signal based on a multi-symbol maximum likelihood detection method; The conjugate relationship calculation module is used to calculate the demodulation value of level k based on the multi-symbol maximum likelihood detection method when the number of branches of the level where the current demodulation calculation is located is greater than K, and calculate the conjugate relationship value of each branch according to the demodulation values ​​of each branch of the k-2th level, the k-1th level and the kth level, and the k-1th level received symbol value, wherein K is the preset number of reserved branches at each level, k is the level where the current demodulation calculation is located, and the conjugate relationship value is the conjugate relationship value between the k-1th level received symbol value and the k-1th level demodulation value; The distance calculation module is used to calculate the similarity value between the k-1th level received symbol value and the k-1th level demodulation value according to the conjugate relationship value and the integral plate value of the corresponding branch, and retain the demodulation values ​​of the K branches with the largest similarity values, and delete the demodulation values ​​of the remaining branches; When the conjugate relationship calculation module calculates to level T, the distance calculation module selects the maximum value from the demodulation values ​​of the K branches at level T as the final demodulation value output, where T is the observation length of the multi-symbol maximum likelihood detection method.

10. The non-coherent GFSK demodulation device based on K-best algorithm according to claim 9, characterized in that: The conjugate relationship calculation module includes a convolution calculation unit, a candidate value calculation unit and a conjugate relationship value calculation unit: The convolution calculation unit is used to perform convolution operation on the demodulated signal of each branch of the k-2th level, the k-1th level and the kth level with the Gaussian filter; The candidate value calculation unit is used to select the intermediate value after convolution for accumulation and generate a complex form of matched filtering candidate value; The conjugate relationship value calculation unit is used to multiply the complex form matched filtering candidate value with the phase accumulation complex value of the previous branch, and then conjugate multiply it with the k-1th level received symbol value to obtain the conjugate relationship value.

11. The non-coherent GFSK demodulation device based on K-best algorithm according to claim 9, characterized in that: The distance calculation module calculates the distance value between the k-1th level received symbol value and the k-1th level demodulated value according to the conjugate relationship value and the integral plate value of the corresponding branch, which is specifically: The conjugate relationship value and the integral board value of the corresponding branch are accumulated, and then the absolute value square calculation is performed on the accumulated integral board value to obtain the similarity value.

12. The non-coherent GFSK demodulation device based on K-best algorithm according to claim 9, characterized in that: The device also includes an LLR calculation module: The LLR calculation module is used to calculate demodulation bit probability information according to the similarity value of the T-th level reserved branch, and the demodulation bit probability information is used to provide to the error correction decoder for decoding and error correction.

Images