A windowed MMSE method combined with dilated convolution

Abstract

The invention discloses a windowed MMSE method combined with a hole convolution, and relates to the field of signal processing. The method comprises: performing a windowing process on the sequence data corresponding to the received first channel received signal and the second channel received signal, and obtaining the windowed data matrix of the two channels; determining the mutual correlation vector and the autocorrelation matrix according to the windowed data matrix of the two channels; determining the MMSE weighting coefficient according to the mutual correlation vector and the autocorrelation matrix; wherein the MMSE weighting coefficient is used to perform a hole convolution on the first channel received signal and the second channel received signal to obtain a synthetic signal after interference suppression. Compared with the prior art, the spatial resolution of the useful signal source and the interference signal source is effectively improved, and the anti-interference ability of the received signal in the weak signal space is improved.

Description

Technical Field

[0001] The present invention relates to the field of signal processing technology, and more specifically, to a windowed MMSE method combined with dilated convolution. Background Art

[0002] Anti-interference is an important function of wireless communication systems, and interference suppression is usually performed from multiple dimensions such as time domain, frequency domain, and spatial domain. For co-frequency and full-band interference, the spatial domain anti-interference method is often used. Its principle is to eliminate the interference signal by utilizing the different directions of the useful signal and the interference signal to achieve interference suppression.

[0003] In the prior art, the minimum mean square error (MMSE) method based on the reference sequence is often used. By adjusting the channel weighting coefficient of the receiving antenna array, the mean square error between the output signal and the reference sequence is minimized, and then the MMSE weighting coefficient of each received signal is obtained based on the least squares optimization algorithm. The weighted sum of each channel signal is obtained to obtain the predicted value, thereby achieving interference signal suppression. It is suitable for wireless communication systems that contain known reference sequences in the transmission data frame. It should be noted that the working principle of the MMSE method is equivalent to predicting the sample value at the current sampling position when suppressing the interference of the current interfered signal; however, in a complex electromagnetic environment system with high interference-to-signal ratio, close wave angle, and weak received signal, interference suppression cannot be performed using only the sample data at the current time or backward delay, thereby reducing the reliability of the system in a complex electromagnetic interference environment. Summary of the invention

[0004] In order to overcome the defect of insufficient reliability of the above-mentioned prior art, the present invention provides a windowed MMSE method combined with dilated convolution.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a windowed MMSE method combined with dilated convolution comprises:

[0007] The received sequence data corresponding to the first channel receiving signal and the second channel receiving signal are subjected to windowing processing to obtain a windowed data matrix of the two channels;

[0008] Determine the cross-correlation vector and the autocorrelation matrix according to the windowed data matrices of the two channels;

[0009] An MMSE weighting coefficient is determined according to the mutual correlation vector and the autocorrelation matrix; wherein the MMSE weighting coefficient is used to perform a dilated convolution on the first channel received signal and the second channel received signal to obtain a composite signal after interference suppression.

[0010] In a second aspect, a spatial domain anti-interference method is provided, applying the method described in the first aspect.

[0011] According to a third aspect, an electronic device includes:

[0012] A memory for storing computer executable instructions or computer programs;

[0013] The processor is used to implement the method described in the first aspect when executing the computer executable instructions or computer program stored in the memory.

[0014] In a fourth aspect, a computer-readable storage medium is provided, characterized in that at least one instruction, at least one program, code set or instruction set is stored on the storage medium, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement the method described in the first aspect.

[0015] In a fifth aspect, a computer program product comprises a computer program or a computer executable instruction, wherein when the computer program or the computer executable instruction is executed by a processor, the method described in the first aspect is implemented.

[0016] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0017] Based on the traditional MMSE method, the present invention uses a windowing method combined with a dilated convolution to generate more accurate MMSE weighting coefficients under the condition of fewer training samples, effectively improve the spatial resolution of useful signal sources and interference signal sources, and at the same time improve the anti-interference ability of the received signal in the weak signal (near the demodulation sensitivity) space, optimize the spatial domain anti-interference performance of the traditional MMSE method, and enhance reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of a windowed MMSE method combined with dilated convolution in Example 1 of the present application;

[0019] Figure 2 This is a schematic diagram of the process of windowing processing in Example 1 of the present application;

[0020] Figure 3 A comparison diagram of the weighting coefficients obtained and the sliding step lengths when filtering the received signal in Example 1 of the present application;

[0021] Figure 4 This is another flowchart of the MMSE method applied to the dual antenna array in Embodiment 1 of the present application;

[0022] Figure 5 This is a flow chart of the spatial domain anti-interference method applied to the dual antenna array in Example 2 of the present application. DETAILED DESCRIPTION

[0023] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged in appropriate cases, which is only to describe the distinction mode adopted by the objects of the same attributes in the embodiments of the present application when describing. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment containing a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment. The term "determine" widely covers various actions, and may include acquisition, calculation, calculation, processing, derivation, investigation, search (for example, search in a table, database or other data structure), ascertainment, and similar actions, and may also include reception (for example, receiving information), access (for example, accessing data in a memory) and similar actions, and may also include generation, creation, establishment and similar actions, as well as parsing, selection, selection and similar actions, etc. The relevant definitions of other terms will be given in the following description.

[0024] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0025] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0026] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0027] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0028] Example 1

[0029] As described in the background technology, when the traditional MMSE algorithm is used to suppress interference on the current interfered signal, it is equivalent to predicting the sample value at the current sampling position. In a complex electromagnetic environment system with high interference-to-signal ratio, close incoming wave angle, and weak received signal, interference suppression cannot be performed using only the sample data at the current moment or backward delay, thereby reducing the reliability of the system in a complex electromagnetic interference environment.

[0030] This embodiment provides a windowed MMSE method combined with dilated convolution, referring to FIG1 , including:

[0031] The received sequence data corresponding to the first channel receiving signal and the second channel receiving signal are subjected to windowing processing to obtain a windowed data matrix of the two channels;

[0032] Determine the cross-correlation vector and the autocorrelation matrix according to the windowed data matrices of the two channels;

[0033] An MMSE weighting coefficient is determined according to the mutual correlation vector and the inverse matrix of the autocorrelation matrix; wherein the MMSE weighting coefficient is used to perform a hole convolution on the first channel received signal and the second channel received signal to obtain a composite signal after interference suppression.

[0034] In this embodiment, a windowing method combined with a dilated convolution is used to generate a more accurate spatial anti-interference weighting coefficient (i.e., MMSE weighting coefficient) under the condition of fewer training samples, thereby effectively improving the spatial resolution of useful signal sources and interference signal sources, while improving the ability of the received signal to resist interference in the weak signal (near the demodulation sensitivity) space, thereby optimizing the spatial anti-interference performance of the traditional MMSE method.

[0035] Those skilled in the art should understand that the convolution described in this embodiment can be achieved by using a filter with a certain window size to perform sliding filtering on the antenna data received based on the antenna channel. Among them, the parameters of the filter include the MMSE weighting coefficient and the sliding step size (i.e., the sample interval between two consecutive filters), which can be obtained through training of training data, and the entire sliding filtering process is convolution. It should also be noted that compared with the ordinary convolution method, the hole convolution uses the method of adding holes to cover a reasonable range of time windows without increasing excessive computational complexity, balancing prediction performance and complexity. Based on this understanding, the hole convolution is to use a filter carrying the MMSE weighting coefficient to perform weighted summation on the first channel received signal and the second channel received signal.

[0036] In order to filter out the interference of each sample point, in some preferred embodiments, when the first channel received signal and the second channel received signal are subjected to sliding filtering using a filter carrying the MMSE weighting coefficient, the sliding step size of the sliding window is set to a single sample point, that is, sample point-by-sample weighted processing is performed to achieve prediction of all sample points.

[0037] In some specific implementations, determining the MMSE weighting coefficient according to the mutual correlation vector and the autocorrelation matrix includes:

[0038] Performing an inverse operation on the autocorrelation matrix to obtain an inverse autocorrelation matrix;

[0039] The inverse autocorrelation matrix is ​​pre-multiplied by the cross-correlation vector to obtain the MMSE weighting coefficient.

[0040] In some preferred embodiments, referring to FIG. 2 , the windowing process includes:

[0041] Define the expansion rate parameter to represent the sample point interval between adjacent sampling points;

[0042] Using a filter with a window size and an expansion rate, according to a preset sliding step, the window corresponding to the sequence data at the current moment is respectively shifted leftward and rightward on the time axis, and sliding filtering is performed to obtain the forward window data and the backward window data at the current moment respectively;

[0043] The forward window data, the backward window data and the current time data are spliced ​​to obtain the windowed data matrix, wherein the mathematical expression of the corresponding predicted value at each time in the windowed data matrix is:

[0044]

[0045] In the formula, Indicates the current moment, represents the current moment forecast value, Represents a splicing operation, represents the expansion rate; Indicates the window size; Indicates The sample value of each channel (i.e. antenna).

[0046] In this preferred embodiment, by setting a window, using the sample point values ​​at the previous and next moments to predict the sample point value at the current moment, and using the correlation between adjacent sample points, the utilization rate of data information is increased, thereby improving the prediction accuracy. At the same time, in this preferred embodiment, by introducing the expansion rate parameter To achieve the addition of holes (i.e., the sample point intervals between adjacent sampling points), that is, the MMSE weighting coefficient achieves the addition of holes by setting a window with an expansion rate parameter, so that the method can cover a reasonable range of time windows without increasing excessive computational complexity.

[0047] It can be understood that the windowed data matrix By several composition.

[0048] It should be emphasized that when setting the window size, those skilled in the art should not set it too small to ensure the accuracy of the prediction results.

[0049] In some specific implementations, for the sampling point at the initial moment, the data cannot be shifted left on the time axis, and 0 is used as the forward window data at that moment.

[0050] In some other specific implementations, when the sampling point is at the end time, the data cannot be shifted rightward on the time axis, and 0 is used as the backward window data at that time.

[0051] In order to make the weight coefficients obtained by the MMSE algorithm have higher generalization performance, the number of samples used in calculating the coefficients should contain richer sample information. In some optional embodiments, a cross-symbol training strategy is adopted to determine the optimal value of the MMSE weight coefficient, that is, each sample corresponds to one symbol during training.

[0052] Furthermore, the cross-symbol training strategy includes:

[0053] The sliding step size during windowing is set to be the same as the sampling multiple, so that samples predicted by different windows correspond to different symbols.

[0054] Those skilled in the art should understand that the sliding step size during training (i.e., the process of obtaining the MMSE weighting coefficient) is set to the number of sample points represented by the symbol, i.e., the sampling multiple, which is conducive to increasing the diversity of training data. If the sliding step size is not set to the sampling multiple, there may be some repeated information between training samples. This is because two or more samples may represent the same symbol. The cross-symbol training strategy is adopted to ensure that the amount of information of the training samples is maximized.

[0055] It should be noted that the sliding step size when obtaining the MMSE weighting coefficient (such as A in FIG. 3 ) is different from the sliding step size when performing sample-by-sample sliding filtering on the received signal (such as B in FIG. 3 )).

[0056] In some preferred embodiments, the sequence data is determined by searching for peaks associated with a reference sequence, comprising:

[0057] Based on the reference sequence correlation peak search, respectively determine a first correlation peak value for the first channel received signal and a second correlation peak value for the second channel received signal;

[0058] Determine a reference signal index position according to the first correlation peak value and the second correlation peak value ;

[0059] According to the reference signal index position , extracting first sequence data and second sequence data from the first channel received signal and the second channel received signal respectively.

[0060] In some preferred embodiments, determining the mutual correlation vector and the autocorrelation matrix comprises:

[0061] The windowed data matrices of the two channels are spliced ​​to obtain a spliced ​​matrix ;

[0062] For the stitching matrix Performing a cross-correlation operation with a preset local reference sequence to determine the cross-correlation vector;

[0063] For the stitching matrix An autocorrelation operation is performed to determine the autocorrelation matrix.

[0064] In some specific implementations, referring to FIG. 4 , the implementation steps of the MMSE method applied to a dual antenna array are as follows:

[0065] S10. Search for the peaks related to the reference sequence to obtain the index Confirm that antenna 1 receives the signal Receive signal with antenna 2 Sequence data corresponding to the position;

[0066] S20, obtaining windowed data of antenna 1 and antenna 2 through windowing processing;

[0067] S30, splicing the windowed data of the two antennas to obtain a data matrix (i.e., the splicing matrix);

[0068] S40, the data matrix Perform correlation operation with the local reference sequence Ref to obtain a cross-correlation vector;

[0069] S50, data matrix Perform an autocorrelation operation to obtain an autocorrelation matrix, and perform an inverse operation on the autocorrelation matrix to obtain an inverse autocorrelation matrix;

[0070] S60, multiply the inverse autocorrelation matrix by the cross-correlation vector to obtain the optimal MMSE weighting coefficient .

[0071] Example 2

[0072] This embodiment provides a spatial domain anti-interference method, which applies the method described in Embodiment 1.

[0073] More specifically, as a non-limiting example, the spatial domain anti-interference method is applied to a multi-antenna array, preferably a dual-antenna array in this embodiment, referring to FIG5 , including:

[0074] Based on the power inversion anti-interference method, the first receiving signal received by the first antenna module and a second receiving signal received by the second antenna module Calculate PI weighting coefficient , using the PI weighting coefficient Data , Perform weighted processing to obtain a PI composite signal;

[0075] Based on the reference sequence correlation peak search, determine the first correlation peak value according to the PI synthetic signal , determining a second correlation peak value according to the first received signal and the second received signal and the third correlation peak ;

[0076] Determine the reference signal index position according to the highest value among the first correlation peak value, the second correlation peak value and the third correlation peak value ;

[0077] Based on the windowed MMSE method described in Example 1, a second composite signal after interference suppression is determined according to the reference signal index position, the first received signal, and the second received signal.

[0078] It should be noted that, in the spatial domain anti-interference method, the power inversion anti-interference method is not necessary and is only used as a preferred example.

[0079] It can be understood that the options in the above-mentioned embodiment 1 are also applicable to this embodiment, so they will not be described again here.

[0080] Example 3

[0081] This embodiment provides a computer-readable storage medium, on which is stored at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor, so that the processor executes part or all of the steps of the method provided in Example 1 of the present application.

[0082] It is understood that the storage medium may be transient or non-transient. Exemplarily, the storage medium includes, but is not limited to, a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0083] Exemplarily, the processor may be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA).

[0084] Exemplarily, the read-only memory includes but is not limited to MASK ROM, PROM, EPROM, EEPROM, Flash, etc.

[0085] Exemplarily, the random access memory includes but is not limited to DRAM, SRAM, SDRAM, DDR SDRAM, etc.

[0086] In some examples, a computer program product is provided, which can be implemented in hardware, software or a combination thereof. As a non-limiting example, the computer program product can be embodied as the storage medium, or as a software product, such as an SDK (Software Development Kit).

[0087] As a non-limiting example, a computer program product is provided, the computer program product includes a computer program or a computer executable instruction, the computer program or the computer executable instruction is stored in a computer-readable storage medium. A processor of an electronic device reads the computer program or the computer executable instruction from the computer-readable storage medium, and the processor executes the computer executable instruction, so that the electronic device performs some or all steps of the method described in the embodiment of the present application.

[0088] In some examples, a computer program is provided, comprising a computer-readable code, and when the computer-readable code is run in a computer device, a processor in the computer device executes a part or all of the steps in the method.

[0089] This embodiment also proposes an electronic device, including a memory and a processor, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and when the processor executes the at least one instruction, at least one program, a code set or an instruction set, it implements part or all of the steps of the method described in Example 1.

[0090] In some examples, a hardware entity of the electronic device is provided, including: a processor, a memory and a communication interface; wherein the processor generally controls the overall operation of the electronic device; the communication interface is used to enable the electronic device to communicate with other terminals or servers through a network; the memory is configured to store instructions and applications executable by the processor, and can also cache data to be processed or processed by the processor and various modules in the electronic device (including but not limited to image data, audio data, voice communication data and video communication data), which can be implemented by flash memory (FLASH), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or random access memory (RAM).

[0091] The processor may include one or more processing elements. Thus, the processor may include one or more integrated circuits (ICs) configured to perform the functions of the processor. In addition, each integrated circuit may include circuits (e.g., a first circuit, a second circuit, and other circuits, etc.) configured to perform the functions of the processor.

[0092] Furthermore, data may be transmitted between the processor, the communication interface and the memory via a bus, which may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together.

[0093] It can be understood that the options in the above-mentioned embodiment 1 are also applicable to this embodiment, so they will not be described again here.

[0094] The same or similar reference numerals correspond to the same or similar components;

[0095] The terms used to describe the positional relationship in the drawings are only used for illustrative purposes and should not be construed as limiting the present application;

[0096] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application may be combined with each other.

[0097] In different specific implementations, the method or system described in the present application can be implemented in software, hardware or a combination thereof. In addition, the order of the steps of the method can be changed, and various elements can be added, reordered, combined, omitted, modified, etc.

[0098] Obviously, the above-mentioned embodiments of the present application are merely examples for clearly illustrating the present application, and are not intended to limit the implementation methods of the present application, and are not intended to limit the present application. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. Each discrete structure / function module or unit can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part, and the structure and function of the discrete components can be implemented as a combined structure or component. It is not necessary and impossible to enumerate all the implementation methods here. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the claims of the present application.

Claims

1. A windowed MMSE method combined with dilated convolution, applied to a dual antenna array, characterized in that: include: The received sequence data corresponding to the first channel receiving signal and the second channel receiving signal are subjected to windowing processing to obtain a windowed data matrix of the two channels; Determine the cross-correlation vector and the autocorrelation matrix according to the windowed data matrices of the two channels; Determine an MMSE weighting coefficient according to the mutual correlation vector and the autocorrelation matrix; wherein the MMSE weighting coefficient is used to perform a hole convolution on the first channel received signal and the second channel received signal to obtain a composite signal after interference suppression; The windowing process includes: Define the expansion rate parameter to represent the sample point interval between adjacent sampling points; Using a filter with a window size and an expansion rate, according to a preset sliding step, the window corresponding to the sequence data at the current moment is respectively shifted leftward and rightward on the time axis, and sliding filtering is performed to obtain the forward window data and the backward window data at the current moment respectively; The forward window data, the backward window data and the current time data are spliced ​​to obtain the windowed data matrix; wherein the mathematical expression of the corresponding predicted value at each moment in the windowed data matrix is: In the formula, Indicates the current moment, Indicates The current moment prediction value of each channel, Represents a splicing operation, represents the expansion rate; Indicates the window size; Indicates The sample values ​​of the channels; and, A cross-symbol training strategy is adopted to determine the optimal value of the MMSE weighting coefficient.

2. The windowed MMSE method combined with dilated convolution according to claim 1, characterized in that: The cross-symbol training strategy includes: The sliding step size during windowing is set to be the same as the sampling multiple, so that samples predicted by different windows correspond to different symbols.

3. The windowed MMSE method combined with dilated convolution according to claim 1, characterized in that: The sequence data is determined by searching for peaks related to the reference sequence, including: Based on the reference sequence correlation peak search, respectively determine a first correlation peak value for the first channel received signal and a second correlation peak value for the second channel received signal; Determine a reference signal index position according to the first correlation peak value and the second correlation peak value ; According to the reference signal index position , extracting first sequence data and second sequence data from the first channel received signal and the second channel received signal respectively.

4. The windowed MMSE method combined with dilated convolution according to any one of claims 1 to 3, characterized in that: The determining of the mutual correlation vector and the autocorrelation matrix comprises: The windowed data matrices of the two channels are spliced ​​to obtain a spliced ​​matrix ; For the stitching matrix Performing a cross-correlation operation with a preset local reference sequence to determine the cross-correlation vector; For the stitching matrix An autocorrelation operation is performed to determine the autocorrelation matrix.

5. A spatial interference prevention method, characterized in that: Apply the method according to any one of claims 1 to 4.

6. An electronic device, characterized in that: include: A memory for storing computer executable instructions or computer programs; The processor is used to implement the method according to any one of claims 1 to 5 when executing the computer executable instructions or computer program stored in the memory.

7. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the method as described in any one of claims 1-5.

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

Images