Filter coefficient optimization method, device and equipment
By constructing the filter coefficient matrix and optimizing the filter coefficient storage, the problem of high complexity in the MMSE filter coefficient calculation in the OFDM system is solved, reducing the occupation of hardware storage resources, and achieving efficient filtering performance.
Patent Information
- Application Number
- CN202311003557.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-09
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2043-08-09
AI Technical Summary
In orthogonal frequency division multiplexing (OFDM) system, when MMSE filter coefficient calculation, the calculation complexity increases due to the high complexity of matrix inversion, and a large number of signal-to-noise ratios (SNR) and root mean squares are required to preset a large number of signal-to-noise ratios (SNR) and root mean squares. Delay parameters and occupy a large amount of hardware storage resources.
By obtaining the preset combination of N group signal-to-noise ratio and root mean square delay, the filter coefficients when PCI mode 6 is equal to 0 and PCI mode 6 is equal to 5 are determined, and a filter coefficient matrix is constructed and stored to optimize the filtering reduces the number of stored filter coefficients.
Without reducing filtering performance, the consumption of hardware storage resources is effectively reduced, the number of stored filter coefficients is reduced, and the storage amount of hardware resources is reduced.
Smart Images

Figure CN117134742B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a filter coefficient optimization method, device and equipment. Background Art
[0002] In the Orthogonal Frequency Division Multiplexing (OFDM) system, the commonly used channel estimation algorithms are least squares LS and root mean square error MMSE. MMSE channel estimation is mainly based on channel parameters, filter coefficients are designed according to certain criteria, and the LS channel estimation results are filtered. MMSE estimation can obtain better estimation accuracy than LS estimation. At the same MSE value, the signal-to-noise ratio of MMSE estimation is 10-15d B higher than that of LS.
[0003] In the related art, the two main parameters used in the calculation of MMSE filter coefficients are the signal-to-noise ratio (SNR) and the root mean square delay of the channel. Since the calculation of MMSE filter coefficients involves matrix inversion, when the matrix dimension is high, the calculation complexity will be very large. Therefore, the industry generally pre-sets a set of root mean square delay and SNR, calculates the filter coefficients according to the preset parameters, and saves them in a table, and performs filtering by table lookup during implementation. However, in order to better adapt to the air interface channel environment, a large number of SNR and root mean square delay parameters need to be preset, so a large number of filter coefficients will be generated, occupying a large amount of hardware storage resources. Summary of the invention
[0004] In view of the problems in the prior art, embodiments of the present invention provide a filter coefficient optimization method, device and equipment.
[0005] Specifically, the embodiment of the present invention provides the following technical solutions:
[0006] In a first aspect, an embodiment of the present invention provides a filter coefficient optimization method, comprising:
[0007] Obtaining N preset combinations of signal-to-noise ratio and root mean square delay;
[0008] Under each combination of the signal-to-noise ratio and the root mean square delay, respectively determine the filter coefficient when PCI mode 6 is equal to 0 and the filter coefficient when PCI mode 6 is equal to 5;
[0009] According to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, the filter coefficient matrices corresponding to each combination of the signal-to-noise ratio and the root mean square delay are obtained; the filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value.
[0010] Further, the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay are obtained according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, including:
[0011] The 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 are added to the tail of the filter coefficients when PCI mode 6 is equal to 0, to obtain the filter coefficient matrices corresponding to each combination of the signal-to-noise ratio and the root mean square delay.
[0012] Further, after adding the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 to the tail of the filter coefficients when PCI mode 6 is equal to 0 to obtain the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay, the method further includes:
[0013] When the filter coefficient matrix has N columns, the first 0.5N columns of the filter coefficient matrix are used as the filter coefficient matrix.
[0014] Furthermore, after obtaining the target filter coefficient matrix, it also includes:
[0015] The filter coefficient matrix is stored in a storage unit of the filter.
[0016] Further, after obtaining the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay, the method further includes:
[0017] According to the signal-to-noise ratio and root mean square delay of the target multipath channel, a target filter coefficient matrix is determined from the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay; the target multipath channel is a multipath channel between the terminal and the target cell;
[0018] The filter coefficient corresponding to the target cell is determined according to the value of PCI mode 6 of the target cell and the target filter coefficient matrix.
[0019] Further, determining the filter coefficient corresponding to the target cell according to the value of PCI mode 6 of the target cell and the target filter coefficient matrix includes:
[0020] According to the Xth to Yth rows in the target filter coefficient matrix, the filter coefficient corresponding to the target cell is determined; wherein X and Y are determined based on the following formula:
[0021] Take 5 as the minuend, the value of the PCI modulo 6 of the target cell as the subtrahend, and the difference between the minuend and the subtrahend is taken as the target result;
[0022] The sum of the target result and 1 is taken as X;
[0023] The sum of the number of all subcarriers in the filtering unit of the filter and the target result is taken as Y.
[0024] Furthermore, after determining the filter coefficient corresponding to the target cell, the method further includes:
[0025] According to the filter coefficient corresponding to the target cell, the least squares LS channel estimation result is filtered to obtain the minimum mean square error MMSE channel estimation result.
[0026] In a second aspect, an embodiment of the present invention further provides a filter coefficient optimization device, comprising:
[0027] An acquisition module, used to acquire N preset groups of signal-to-noise ratio and root mean square delay combinations;
[0028] A processing module, used for determining the filter coefficient when PCI mode 6 is equal to 0 and the filter coefficient when PCI mode 6 is equal to 5 respectively under each combination of the signal-to-noise ratio and the root mean square delay;
[0029] The optimization module is used to obtain the filter coefficient matrices corresponding to each combination of the signal-to-noise ratio and the root mean square delay according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0; the filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value.
[0030] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the filter coefficient optimization method as described in the first aspect is implemented.
[0031] In a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the filter coefficient optimization method as described in the first aspect is implemented.
[0032] In a fifth aspect, an embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the filter coefficient optimization method as described in the first aspect.
[0033] The filter coefficient optimization method, device and equipment provided by the embodiments of the present invention obtain N preset groups of signal-to-noise ratio and root mean square delay combinations, and determine the filter coefficients of each signal-to-noise ratio and root mean square delay combination when PCI mode 6 is equal to 0 and PCI mode 6 is equal to 5, and then obtain the filter coefficient matrices corresponding to each signal-to-noise ratio and root mean square delay combination according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, and the filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value, thereby optimizing the filter coefficients; compared with the prior art, when the number of preset SNR and root mean square delay parameters remains unchanged, the filter coefficient optimization method in the present application can effectively reduce the number of filter coefficients to be stored without reducing the filtering performance, so that the required hardware storage resources are greatly reduced, and the hardware resource storage amount is effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0035] Figure 1 is a flowchart of a filter coefficient optimization method provided by an embodiment of the present invention;
[0036] Figure 2 is a schematic diagram of generating a filter coefficient matrix provided by an embodiment of the present invention;
[0037] Figure 3 is a schematic diagram of filter coefficient optimization provided by an embodiment of the present invention;
[0038] Figure 4 is a schematic diagram of determining a filter coefficient corresponding to a target cell provided by an embodiment of the present invention;
[0039] Figure 5 is a structural schematic diagram of a filter coefficient optimization device provided by an embodiment of the present invention;
[0040] Figure 6 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0041] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0042] The method of the embodiment of the present invention can be applied to the filter coefficient optimization scenario. When the preset SNR and the number of root mean square delay parameters remain unchanged, the filter coefficient optimization method in the present application can effectively reduce the number of filter coefficients to be stored without reducing the filtering performance, thereby greatly reducing the required hardware storage resources and effectively reducing the storage capacity of hardware resources.
[0043] In the related art, the two main parameters used in the calculation of MMSE filter coefficients are the signal-to-noise ratio (SNR) and the root mean square delay of the channel. Since the calculation of MMSE filter coefficients involves matrix inversion, when the matrix dimension is high, the calculation complexity will be very large. Therefore, the industry generally pre-sets a set of root mean square delay and SNR, calculates the filter coefficients according to the preset parameters, and saves them in a table, and performs filtering by table lookup during implementation. However, in order to better adapt to the air interface channel environment, a large number of SNR and root mean square delay parameters need to be preset, so a large number of filter coefficients will be generated, occupying a large amount of hardware storage resources.
[0044] The filter coefficient optimization method of the embodiment of the present invention obtains N preset groups of signal-to-noise ratio and root mean square delay combinations, and determines the filter coefficients of each signal-to-noise ratio and root mean square delay combination when PCI mode 6 is equal to 0 and PCI mode 6 is equal to 5, and then obtains the filter coefficient matrices corresponding to each signal-to-noise ratio and root mean square delay combination according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, and the filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value, thereby optimizing the filter coefficients; compared with the prior art, when the number of preset SNR and root mean square delay parameters remains unchanged, the filter coefficient optimization method in the present application can effectively reduce the number of filter coefficients to be stored without reducing the filtering performance, so that the required hardware storage resources are greatly reduced, and the hardware resource storage amount is effectively reduced.
[0045] In order to facilitate a clearer understanding of the technical solutions of the embodiments of the present application, some technical contents related to the embodiments of the present application are first introduced.
[0046] In OFDM systems, commonly used channel estimation algorithms include LS and MMSE. LS estimation only requires one divider to estimate the pilot point channel response. Its computational complexity increases only linearly with the number of pilot points. It has low complexity and fast computation, making it suitable for high-speed transmission communication systems such as OFDM. However, the accuracy of LS estimation is easily affected by Gaussian noise and inter-subcarrier interference. Since the channel response of the data subcarrier is completely obtained by interpolating and filtering the channel estimate of the pilot subcarrier, the performance of OFDM system equalization depends largely on the accuracy of the pilot point channel estimation.
[0047] MMSE channel estimation is mainly based on channel parameters, filter coefficients are designed according to certain criteria, and the LS channel estimation results are filtered. MMSE estimation can obtain better estimation accuracy than LS estimation. At the same MSE value, the signal-to-noise ratio of MMSE estimation is 10-15d B higher than that of LS. However, the main disadvantage of MMSE estimation is its high algorithm complexity. As the number of operation points increases, its algorithm complexity increases exponentially.
[0048] In engineering, wireless channels have very complex propagation paths, which can cause large variations in multipath delays. If a fixed RMS delay is used, the resulting correlation function mismatch will cause a certain loss in channel estimation performance. If the real-time RMS delay can be obtained according to changes in channel conditions during communication, and the MMSE channel estimation parameters are adjusted accordingly, the best MMSE channel estimation performance can be obtained. Based on this, the industry has proposed an adaptive parameter MMSE channel estimation coefficient adjustment algorithm, which estimates the channel RMS delay spread and signal-to-noise ratio, adaptively adjusts the channel estimation parameters, and generates the best MMSE channel estimation coefficients to filter the LS estimated channel response, which has better channel estimation performance than the fixed coefficient MMSE channel estimation algorithm.
[0049] In order to facilitate engineering applications, a simplified version of the MMSE algorithm, the LMMSE algorithm, was proposed to convert the complex two-dimensional time-frequency filtering of MMSE into two one-dimensional filtering. However, the inverse matrix needs to be calculated in the process of calculating the coefficients of the two filters of the LMMSE algorithm. When the matrix dimension is relatively high, the complexity of the matrix inversion will be greatly increased. The two parameters SNR (signal-to-noise ratio) and the root mean square delay of the channel are mainly used in the calculation of the LMMSE filter coefficients.
[0050] Since the calculation process of LMMSE filter coefficients involves matrix inversion, when the matrix dimension is high, the calculation complexity will be very large. Therefore, the industry generally pre-sets a set of root mean square delay and SNR, calculates the filter coefficients according to the preset parameters, and saves them in a table, and performs filtering by looking up the table during implementation.
[0051] In actual applications, the air interface environment channel changes in real time, so the filter coefficients need to be updated in real time. If the filter coefficients are calculated continuously according to the channel environment, this will also bring great pressure to the hardware calculation and storage. Although the filter coefficients are stored in a table according to the preset parameters in the prior art, the complex inverse matrix operation is removed during implementation, but in order to better adapt to the air interface channel environment, a large number of SNR and RMS delay parameters need to be preset, so a large number of filter coefficients will be generated, occupying a large amount of hardware storage resources.
[0052] Combine the following Figure 1-Figure 6 The technical solution of the present invention is described in detail with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.
[0053] Figure 1 FIG. 1 is a flow chart of an embodiment of a filter coefficient optimization method provided by an embodiment of the present invention. Figure 1 As shown, the method provided in this embodiment includes:
[0054] Step 101: Obtain N preset groups of signal-to-noise ratio and root mean square delay combinations;
[0055] Specifically, the two main parameters used in the calculation of MMSE filter coefficients are the signal-to-noise ratio (SNR) and the root mean square delay of the channel. Since the calculation process of MMSE filter coefficients involves matrix inversion, when the matrix dimension is high, the calculation complexity will be very large. Therefore, the industry generally pre-sets a set of root mean square delay and SNR, calculates the filter coefficients according to the preset parameters, and saves them in a table, and performs table lookup for filtering when it is implemented. However, in order to better adapt to the air interface channel environment, a large number of SNR and root mean square delay parameters need to be preset, so a large number of filter coefficients will be generated, occupying a large amount of hardware storage resources.
[0056] In order to solve the above problem, in the embodiment of the present application, N preset groups of signal-to-noise ratio and root mean square delay combinations are first obtained; for example, 10 preset SNR values and 6 root mean square delay values are obtained, and a total of 10*6 groups of signal-to-noise ratio and root mean square delay combinations are combined. Optionally, the number of signal-to-noise ratio and root mean square delay combinations and the SNR values and root mean square delay values in the signal-to-noise ratio and root mean square delay combinations can also be determined according to actual conditions.
[0057] Step 102: Under each combination of signal-to-noise ratio and root mean square delay, determine the filter coefficient when PCI mode 6 is equal to 0 and the filter coefficient when PCI mode 6 is equal to 5 respectively;
[0058] Specifically, after obtaining the preset N groups of signal-to-noise ratio and root mean square delay combinations, the filter coefficients when PCI mode 6 is equal to 0 and the filter coefficients when PCI mode 6 is equal to 5 are determined in the embodiment of the present application for each signal-to-noise ratio and root mean square delay combination. That is, based on the research findings that in the LTE system, the position of CRS in the RB is related to PCI, specifically, to PCI mode 6, when there is a set of SNR and root mean square delay parameters, there is no need to store all the coefficients in the 6 groups of filter coefficients corresponding to PCI mode 6. In fact, only the filter coefficients in the two cases of PCI mode 6 equal to 5 and PCI mode 6 equal to 0 need to be calculated; wherein each filter coefficient is a complex number, including two parts, a real number I and an imaginary number Q, each of which is 16 bits. Optionally, the filter coefficients when PCI mode 6 is equal to 0 and the filter coefficients when PCI mode 6 is equal to 5 for each signal-to-noise ratio and root mean square delay combination can be calculated based on the calculation method in the prior art, which will not be repeated in the embodiment of the present application.
[0059] Step 103, based on the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, obtain the filter coefficient matrices corresponding to each combination of signal-to-noise ratio and root mean square delay; the filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value.
[0060] Specifically, after determining the filter coefficients when PCI mode 6 is equal to 0 and the filter coefficients when PCI mode 6 is equal to 5 respectively under various combinations of signal-to-noise ratios and root mean square delays, in the embodiment of the present application, the filter coefficient matrices corresponding to the combinations of signal-to-noise ratios and root mean square delays are obtained according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0; the filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value. Optionally, the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 can be added to the tail of the filter coefficients when PCI mode 6 is equal to 0 to obtain the filter coefficient matrices corresponding to the combinations of signal-to-noise ratios and root mean square delays, thereby optimizing the filter coefficients.
[0061] For example, according to the solution of the prior art, 6 groups of filter coefficients need to be stored. Assume that when the filter coefficients are pre-calculated, there are 10 SNR values and 6 root mean square delay values, which together have a total of 60*6 groups of filter coefficients. If the LTE system has 2 RS REs in each RB on the CRS symbol, that is, each RB has 12 subcarriers, but only 2 RSREs carry CSR, and there is no reference signal on the remaining 10 subcarriers; 3 RBs are used as a filtering unit (one RB contains 2 CRS REs), that is, there are 6 RSREs on 3 RBs, and the channel estimation values on all 36 subcarriers are estimated using 6 RSREs, then the dimension of the filter coefficients is 36*6, that is, 36 groups of coefficients are used to filter the 6 LS channel estimation values in 3 RBs to obtain the channel estimation values of all RE positions in 3 RBs. According to the above parameter settings, a total of 60*6*36*6=77760 filter coefficients need to be stored.
[0062] Based on the filter coefficient optimization scheme in this application, such as Figure 2 As shown, when a set of SNR and RMS delay parameters are used, it is not necessary to store all the coefficients in the 6 groups of filter coefficients corresponding to PCI mode 6. In fact, only the filter coefficients in the two cases of PCI mode 6 equal to 5 and PCI mode 6 equal to 0 need to be calculated, and then the first 5 rows when PCI mode 6 equals 5 are extracted, and then combined with the coefficients corresponding to PCI mode 6 equal to 0, that is, the corresponding coefficients in all cases of PCI mode 6 can be included. When 3 RBs are used as a filtering unit, the number of filter coefficients can be changed from 6*36*6 to 41*6; that is, based on the filter coefficient optimization scheme in the present application, according to 3 RBs as a filtering unit, 10 SNR values, and 6 RMS delay values, it is necessary to store 60*41*3 coefficients. Compared with the original 60*6*36*6, the hardware storage resources are reduced by more than 90%, effectively reducing the occupation of hardware storage resources.
[0063] The method of the above embodiment obtains N preset groups of combinations of signal-to-noise ratio and root mean square delay, and determines the filter coefficients of each combination of signal-to-noise ratio and root mean square delay when PCI mode 6 is equal to 0 and PCI mode 6 is equal to 5, and then obtains the filter coefficient matrices corresponding to each combination of signal-to-noise ratio and root mean square delay according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, and the filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value, thereby optimizing the filter coefficients; compared with the prior art, when the number of preset SNR and root mean square delay parameters remains unchanged, the filter coefficient optimization method in the present application can effectively reduce the number of filter coefficients to be stored without reducing the filtering performance, so that the required hardware storage resources are greatly reduced, and the hardware resource storage amount is effectively reduced.
[0064] In one embodiment, after adding the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 to the tail of the filter coefficients when PCI mode 6 is equal to 0 to obtain the filter coefficient matrices corresponding to each combination of signal-to-noise ratio and root mean square delay, the following further comprises:
[0065] When the filter coefficient matrix has N columns, the first 0.5N columns in the filter coefficient matrix are used as the filter coefficient matrix.
[0066] Specifically, in the embodiment of the present application, the filter coefficient matrices corresponding to each combination of signal-to-noise ratio and root mean square delay are obtained according to the 1st to 5th rows of the filter coefficients when PCI mod 6 is equal to 5 and the filter coefficients when PCI mod 6 is equal to 0; wherein the filter coefficient matrix includes the filter coefficients when PCI mod 6 is equal to any value, thereby realizing the first optimization of the filter coefficients. In the embodiment of the present application, the filter coefficients can be further optimized for the second time, that is, when the filter coefficient matrix is N columns, the first 0.5N columns in the filter coefficient matrix are used as the filter coefficient matrix; that is, based on the research, it is further found that the relationship between the elements in the 1st column and the Nth column in the filter coefficient matrix is an inversion plus conjugate relationship, that is, after inverting and conjugating the 1st column in the filter coefficient matrix, the Nth column in the filter coefficient matrix can be obtained; therefore, based on the elements in the 1st column of the filter coefficient matrix, the elements in the Nth column can also be calculated, so that when the filter coefficient matrix is N columns, the first 0.5N columns in the filter coefficient matrix can be stored as the filter coefficient matrix in the storage unit of the filter.
[0067] Exemplarily, for a set of SNR and RMS delay, the filter coefficient secondary optimization process is as follows: Figure 3 As shown, for the 41*6 filter coefficients, according to the characteristics of each column of data, only the first 3 columns need to be stored, that is, only 41*3 coefficients need to be stored. Optionally, when the filter coefficient matrix is N columns, the last 0.5N columns in the filter coefficient matrix can be used as the filter coefficient matrix; thereby realizing the second optimization of the filter coefficients, further reducing the number of filter coefficients to be stored, greatly reducing the required hardware storage resources, and effectively reducing the storage capacity of hardware resources. That is, based on the filter coefficient optimization method in this application, when calculating the filter coefficients in the LTE system, only the two cases of PCI mode 6 equal to 5 and PCI mode 6 equal to 0 need to be considered. According to the filter coefficients corresponding to these two cases, the filter coefficient matrix corresponding to all 6 cases of PCI mode 6 is combined to reduce storage resources; and then according to the characteristics of the filter coefficients in each column of the filter coefficient matrix, only half of the number of filter coefficients need to be stored, thereby further reducing the demand for storage resources.
[0068] The method of the above embodiment is based on the inverted and conjugate relationship between the elements in the 1st column and the Nth column in the filter coefficient matrix. Therefore, based on the elements in the 1st column of the filter coefficient matrix, the elements in the Nth column can also be calculated. Therefore, when the filter coefficient matrix has N columns, the first 0.5N columns in the filter coefficient matrix can be used as the filter coefficient matrix, thus realizing the second optimization of the filter coefficients, further reducing the number of filter coefficients to be stored, greatly reducing the required hardware storage resources, and effectively reducing the hardware resource storage capacity.
[0069] In one embodiment, after obtaining the target filter coefficient matrix, the method further includes:
[0070] The filter coefficient matrix is stored in the memory cell of the filter.
[0071] Specifically, in the prior art, due to the huge number of filter coefficients and the small capacity of the storage unit in the filter, the filter coefficients cannot be effectively stored in the storage unit of the filter, but the filter coefficients are stored in a storage unit such as DDR outside the filter, and then in the process of using the filter coefficients, it is necessary to frequently interact with the external storage unit, resulting in information interaction and calculation delays. The present application effectively reduces the number of filter coefficients to be stored by optimizing the filter coefficients, so that the required hardware storage resources are greatly reduced, so the optimized filter coefficients can be stored in the storage unit of the filter, thereby saving hardware storage resources, and at the same time can effectively solve the problem of information interaction calculation delay caused by storing the filter coefficients in the external storage unit when the number of filter coefficients is too large in the prior art.
[0072] After obtaining the filter coefficient matrices corresponding to the combinations of each signal-to-noise ratio and root mean square delay, the following is also included:
[0073] According to the signal-to-noise ratio and root mean square delay of the target multipath channel, a target filter coefficient matrix is determined from the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay; the target multipath channel is the multipath channel between the terminal and the target cell;
[0074] According to the value of PCI mode 6 of the target cell and the target filter coefficient matrix, the filter coefficient corresponding to the target cell is determined.
[0075] Specifically, after obtaining the filter coefficient matrices corresponding to the combinations of various signal-to-noise ratios and root mean square delays, a target filter coefficient matrix can be determined from the filter coefficient matrices corresponding to the combinations of various signal-to-noise ratios and root mean square delays according to the signal-to-noise ratios and root mean square delays of the multipath channel between the terminal and the target cell; optionally, a combination closest to the signal-to-noise ratio and root mean square delay of the multipath channel between the terminal and the target cell can be determined from a plurality of preset combinations of signal-to-noise ratios and root mean square delays; and then the filter coefficient matrix corresponding to the determined closest combination is used as the target filter coefficient matrix.
[0076] After determining the target filter coefficient matrix, since the target filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value, the filter coefficient corresponding to the target cell can be determined from the target filter coefficient matrix according to the value of PCI mode 6 of the target cell.
[0077] Optionally, the filter coefficient corresponding to the target cell is determined according to the Xth row to the Yth row in the target filter coefficient matrix; wherein X and Y are determined based on the following formula:
[0078] Take 5 as the minuend, the value of the PCI modulo 6 of the target cell as the subtrahend, and the difference between the minuend and the subtrahend is taken as the target result;
[0079] The sum of the target result and 1 is taken as X;
[0080] The sum of the number of all subcarriers in the filter unit of the filter and the target result is taken as Y.
[0081] Specifically, according to the value of PCI mode 6 of the target cell, a row coefficient of 5-PCI mode 6 value + [1: the number of all subcarriers in the filtering unit of the filter] can be selected from the target filter coefficient matrix as the coefficient used by the current cell; optionally, if 3 RBs are one filtering unit, the number of all subcarriers in the filtering unit of the filter is 36; if 2 RBs are one filtering unit, the number of all subcarriers in the filtering unit of the filter is 24; if 1 RB is one filtering unit, the number of all subcarriers in the filtering unit of the filter is 12.
[0082] That is, 5 is taken as the minuend, the value of PCI mod 6 of the target cell is taken as the subtrahend, and the difference between the minuend and the subtrahend is taken as the target result; the sum of the target result and 1 is taken as X; the sum of the number of all subcarriers in the filtering unit of the filter and the target result is taken as Y; and then the Xth to Yth rows in the target filter coefficient matrix are taken as the filter coefficients corresponding to the target cell, so as to accurately and efficiently determine the filter coefficients corresponding to the target cell from the filter coefficient matrix according to the value of PCI mod 6 of the target cell.
[0083] For example, Figure 4 As shown, after obtaining the signal-to-noise ratio and root mean square delay of the multipath channel between the terminal and the target cell, the table lookup index can be calculated, and then the corresponding target filter coefficient matrix can be determined according to the table lookup index; wherein the target filter coefficient matrix is stored in the table current Coff. Then, the value of PCI mode 6 of the target cell is calculated; finally, the row coefficient of 5-PCI mode 6 value + [1: the number of all subcarriers in the filter unit of the filter] is selected from the target filter coefficient matrix as the coefficient used by the current cell, thereby achieving the effect of accurately and efficiently determining the filter coefficient corresponding to the target cell from the filter coefficient matrix according to the value of PCI mode 6 of the target cell.
[0084] In one embodiment, after determining the filter coefficient corresponding to the target cell, the method further includes:
[0085] According to the filter coefficient corresponding to the target cell, the least squares LS channel estimation result is filtered to obtain the minimum mean square error MMSE channel estimation result.
[0086] Specifically, after determining the filter coefficient corresponding to the target cell according to the value of PCI mode 6 of the target cell and the target filter coefficient matrix, the least squares LS channel estimation result can be filtered according to the filter coefficient corresponding to the target cell, so as to quickly and accurately obtain the minimum mean square error MMSE channel estimation result, thereby improving the efficiency and accuracy of channel estimation.
[0087] The filter coefficient optimization device provided by the present invention is described below. The filter coefficient optimization device described below and the filter coefficient optimization method described above can be referred to each other.
[0088] Figure 5 : is a schematic diagram of the structure of the filter coefficient optimization device provided by the present invention. The filter coefficient optimization device provided by this embodiment includes:
[0089] An acquisition module 710 is used to acquire N preset groups of signal-to-noise ratio and root mean square delay combinations;
[0090] A processing module 720, for determining the filter coefficient when PCI mode 6 is equal to 0 and the filter coefficient when PCI mode 6 is equal to 5 respectively under various combinations of signal-to-noise ratio and root mean square delay;
[0091] The optimization module 730 is used to obtain the filter coefficient matrix corresponding to each combination of signal-to-noise ratio and root mean square delay according to the 1st to 5th rows of the filter coefficient when PCI mode 6 is equal to 5 and the filter coefficient when PCI mode 6 is equal to 0; the filter coefficient matrix includes the filter coefficient when PCI mode 6 is equal to any value.
[0092] Optionally, the optimization module 730 is specifically used to: add the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 to the tail of the filter coefficients when PCI mode 6 is equal to 0, and obtain the filter coefficient matrices corresponding to each combination of signal-to-noise ratio and root mean square delay.
[0093] Optionally, the optimization module 730 is specifically configured to: when the filter coefficient matrix has N columns, use the first 0.5N columns in the filter coefficient matrix as the filter coefficient matrix.
[0094] Optionally, the optimization module 730 is specifically used to: store the filter coefficient matrix in a storage unit of the filter.
[0095] Optionally, the optimization module 730 is specifically used to: determine a target filter coefficient matrix from filter coefficient matrices corresponding to respective combinations of signal-to-noise ratios and root mean square delays according to a signal-to-noise ratio and a root mean square delay of a target multipath channel; the target multipath channel is a multipath channel between a terminal and a target cell;
[0096] According to the value of PCI mode 6 of the target cell and the target filter coefficient matrix, the filter coefficient corresponding to the target cell is determined.
[0097] Optionally, the optimization module 730 is specifically configured to: determine the filter coefficient corresponding to the target cell according to the Xth row to the Yth row in the target filter coefficient matrix; wherein X and Y are determined based on the following formula:
[0098] Take 5 as the minuend, the value of the PCI modulo 6 of the target cell as the subtrahend, and the difference between the minuend and the subtrahend is taken as the target result;
[0099] The sum of the target result and 1 is taken as X;
[0100] The sum of the number of all subcarriers in the filter unit of the filter and the target result is taken as Y.
[0101] Optionally, the optimization module 730 is specifically used to: filter the least squares LS channel estimation result according to the filter coefficient corresponding to the target cell to obtain the minimum mean square error MMSE channel estimation result.
[0102] The device of the embodiment of the present invention is used to execute the method in any of the aforementioned method embodiments. Its implementation principle and technical effects are similar and will not be repeated here.
[0103] Figure 6An example of a physical structure diagram of an electronic device is provided, and the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute a filter coefficient optimization method, which includes: obtaining a preset N group of combinations of signal-to-noise ratios and root mean square delays; under each combination of signal-to-noise ratios and root mean square delays, determining the filter coefficients when PCI mode 6 is equal to 0 and the filter coefficients when PCI mode 6 is equal to 5 respectively; according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, obtaining the filter coefficient matrices corresponding to each combination of signal-to-noise ratios and root mean square delays respectively; the filter coefficient matrix includes the filter coefficients when PCI mode 6 is equal to any value.
[0104] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0105] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the filter coefficient optimization method provided by the above-mentioned methods, and the method includes: obtaining N preset groups of signal-to-noise ratio and root mean square delay combinations; under each combination of signal-to-noise ratio and root mean square delay, determining the filter coefficient when PCI mode 6 is equal to 0 and the filter coefficient when PCI mode 6 is equal to 5 respectively; according to the 1st to 5th rows of the filter coefficient when PCI mode 6 is equal to 5 and the filter coefficient when PCI mode 6 is equal to 0, obtaining the filter coefficient matrix corresponding to each combination of signal-to-noise ratio and root mean square delay; the filter coefficient matrix includes the filter coefficient when PCI mode 6 is equal to any value.
[0106] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the processor executes the above-mentioned filter coefficient optimization method, the method comprising: obtaining N preset groups of signal-to-noise ratio and root mean square delay combinations; under each combination of signal-to-noise ratio and root mean square delay, determining the filter coefficient when PCI mode 6 is equal to 0 and the filter coefficient when PCI mode 6 is equal to 5 respectively; according to the 1st to 5th rows of the filter coefficient when PCI mode 6 is equal to 5 and the filter coefficient when PCI mode 6 is equal to 0, obtaining the filter coefficient matrix corresponding to each combination of signal-to-noise ratio and root mean square delay respectively; the filter coefficient matrix includes the filter coefficient when PCI mode 6 is equal to any value.
[0107] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0108] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A filter coefficient optimization method, characterized in that: include: Obtaining N preset combinations of signal-to-noise ratio and root mean square delay; Under each combination of the signal-to-noise ratio and the root mean square delay, respectively determine the filter coefficient when the physical cell identifier PCI mode 6 is equal to 0 and the filter coefficient when the PCI mode 6 is equal to 5; According to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay are obtained; The filter coefficient matrix includes filter coefficients when PCI modulo 6 is equal to any value; The filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay are obtained according to the 1st to 5th rows of the filter coefficients when the PCI mode 6 is equal to 5 and the filter coefficients when the PCI mode 6 is equal to 0, including: Adding the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 to the tail of the filter coefficients when PCI mode 6 is equal to 0, to obtain the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay; After adding the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 to the tail of the filter coefficients when PCI mode 6 is equal to 0 to obtain the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay, the method further includes: When the filter coefficient matrix has N columns, the first 0.5N columns of the filter coefficient matrix are used as the filter coefficient matrix.
2. The filter coefficient optimization method according to claim 1, characterized in that: After obtaining the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay, the method further includes: According to the signal-to-noise ratio and root mean square delay of the target multipath channel, a target filter coefficient matrix is determined from the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay; the target multipath channel is a multipath channel between the terminal and the target cell; The filter coefficient corresponding to the target cell is determined according to the value of PCI mode 6 of the target cell and the target filter coefficient matrix.
3. The filter coefficient optimization method according to claim 2, characterized in that: The step of determining the filter coefficient corresponding to the target cell according to the value of PCI mode 6 of the target cell and the target filter coefficient matrix includes: According to the Xth to Yth rows in the target filter coefficient matrix, the filter coefficient corresponding to the target cell is determined; wherein X and Y are determined based on the following formula: Take 5 as the minuend, the value of the PCI modulo 6 of the target cell as the subtrahend, and the difference between the minuend and the subtrahend is taken as the target result; The sum of the target result and 1 is taken as X; The sum of the number of all subcarriers in the filtering unit of the filter and the target result is taken as Y.
4. The filter coefficient optimization method according to claim 3, characterized in that: After determining the filter coefficient corresponding to the target cell, it also includes: According to the filter coefficient corresponding to the target cell, the least squares LS channel estimation result is filtered to obtain the minimum mean square error MMSE channel estimation result.
5. A filter coefficient optimization device, characterized in that: include: An acquisition module, used to acquire N preset groups of signal-to-noise ratio and root mean square delay combinations; A processing module, used for determining the filter coefficient when PCI mode 6 is equal to 0 and the filter coefficient when PCI mode 6 is equal to 5 respectively under each combination of the signal-to-noise ratio and the root mean square delay; An optimization module, used to obtain filter coefficient matrices corresponding to each combination of the signal-to-noise ratio and the root mean square delay according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0; The filter coefficient matrix includes filter coefficients when PCI mode 6 is equal to any value; the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay are obtained according to the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 and the filter coefficients when PCI mode 6 is equal to 0, including: Adding the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 to the tail of the filter coefficients when PCI mode 6 is equal to 0, to obtain the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay; After adding the 1st to 5th rows of the filter coefficients when PCI mode 6 is equal to 5 to the tail of the filter coefficients when PCI mode 6 is equal to 0 to obtain the filter coefficient matrices corresponding to the combinations of the signal-to-noise ratio and the root mean square delay, the method further includes: When the filter coefficient matrix has N columns, the first 0.5N columns of the filter coefficient matrix are used as the filter coefficient matrix.
6. An electronic device 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 filter coefficient optimization method according to any one of claims 1 to 4 is implemented.
7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the filter coefficient optimization method according to any one of claims 1 to 4 is implemented.
