Amplitude characteristic two-dimensional correction method suitable for ultra-wideband channel

Abstract

The invention discloses an amplitude characteristic two-dimensional correction method suitable for an ultra-wideband channel, and the method comprises the steps: S1, carrying out the calculation of a center frequency point needed by the establishment of calibration matrix collection data according to a channel working frequency band parameter and an instantaneous working bandwidth parameter, generating a center frequency point sequence FRE11 * N, and setting an effective frequency band width; s2, controlling a channel to work at a set center frequency point sequence FRE1, performing acquisition, spectral characteristic calculation and smoothing on data of each center frequency point until all center frequency points are processed, and generating a compensation matrix C; s3, calculating each compensation frequency point value according to the channel center frequency point and the bandwidth which need to be compensated; calculating an interpolation coefficient according to each compensation frequency point value, completing two-dimensional compensation coefficient calculation based on the compensation matrix, and generating a frequency spectrum correction value JZCOEm corresponding to each compensation frequency point; and S4, according to an output result of the step S3, adding the frequency spectrum to be corrected and the frequency spectrum correction value to complete final frequency spectrum correction.

Description

Technical Field

[0001] The present invention relates to a two-dimensional correction method for amplitude characteristics of an ultra-wideband channel, in particular to a correction method when the channel amplitude characteristics change two-dimensionally with the frequency in the instantaneous working bandwidth and the local oscillator frequency switching, and belongs to the technical field of signal processing. Background Art

[0002] With the continuous iterative development of design technology and application requirements, microwave channels and microwave devices used in various fields show a trend of increasingly wider bandwidths. In order to ensure system performance, the designed microwave channels must have flat amplitude performance within the working bandwidth. Due to the influence of the cascade of various device modules such as amplitude limiting protection circuits, low-noise amplifiers, and multi-stage power amplifiers, the actual amplitude-frequency response often fluctuates greatly with frequency within the working bandwidth, and this channel characteristic often leads to serious distortion of the received spectrum, affecting indicators such as the dynamic range and sensitivity of the receiving system. Correction and compensation are required in practical applications to ensure channel flatness.

[0003] At present, the existing correction technologies can be divided into two categories: analog correction and digital correction. Among them, analog correction is generally carried out by constructing an equalizer, which is composed of lumped parameters such as inductance (L), capacitance (C), and resistance (R). It is essentially an inverse filter. The analog equalizer has the characteristics of high processing speed, small size, low power consumption and minimum delay, but it needs to obtain the amplitude-frequency characteristics within the entire working bandwidth in advance, and can only compensate for fixed change types such as monotonic rise and fall, concave and convex changes in amplitude-frequency characteristics. Therefore, the reliability is low and it cannot be applied to ultra-wideband channels with complex amplitude-frequency characteristics. Digital correction uses digital signal processing algorithms and has better reliability. During the digital correction process, the noise floor data of each frequency point of the channel is collected, and the collected data is analyzed to estimate the inconsistency of the amplitude of each frequency point: such as the correction method based on the subspace principle, which jointly estimates the amplitude-frequency error of each frequency point online, and finds the optimal solution through iterative calculations; some researchers use the ARMA model to approximate the amplitude error characteristics of the broadband channel, and realize the amplitude-frequency error estimation of the broadband incident signal; for example, by establishing a broadband channel model and using the collected data to solve the model parameters, the amplitude change characteristics of the broadband channel are fitted, and compensation is completed on this basis. The above methods have achieved some results and progress, but none of them are suitable for amplitude and phase compensation under ultra-wideband channel conditions. The reasons include: ultra-wideband channels generally adopt the architecture of fixed local oscillator + adjustable local oscillator + amplification and attenuation module + filtering module, and its channel amplitude variation characteristics are jointly determined by the local oscillator frequency and the amplification and filtering modules, showing the characteristics of two-dimensional distribution. The existing methods only focus on compensating for the amplitude and phase inconsistency within the instantaneous working bandwidth, ignoring the gain changes brought about by the local oscillator switching, resulting in poor compensation effect within the full bandwidth; at the same time, due to the limitation of the spectrum resolution of digital signals, they cannot adapt to the situation of more drastic changes within the band. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and solve the problem of effectively compensating for channel amplitude inconsistency within the full bandwidth.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A two-dimensional correction method for amplitude characteristics of ultra-wideband channels is proposed. First, the original noise data under different local oscillator and bandwidth conditions in the ultra-wideband channel is collected, and the collected results are transformed into the frequency domain and filtered. On this basis, a two-dimensional smooth correction matrix is ​​established; when correction is required, the compensation value is calculated in real time based on the interpolation coefficient according to the current working center frequency and bandwidth, and the two-dimensional smooth correction matrix is ​​combined to complete the channel compensation at any frequency point. Experimental results show that this method can effectively compensate for channel amplitude inconsistency within the full bandwidth.

[0007] The specific steps include:

[0008] S1, according to the lower limit of the channel working frequency band FRE Lower 、FRE upper limit of working frequency band Upper And instantaneous working bandwidth BW All , calculate the center frequency points required to establish the calibration matrix and collect data, and generate the center frequency point sequence FRE1 1*N At the same time, according to the instantaneous working bandwidth of the channel, set the effective bandwidth BW valid ;

[0009] S2, the control channel works at the set center frequency sequence FRE1, collects data of the i-th center frequency, and obtains the collected data sequence Calculate and smooth the spectrum of the data sequence to obtain the spectrum characteristics corresponding to the i-th center frequency point After all center frequency points are processed, the compensation matrix C is generated;

[0010] S3. Collect channel data to be compensated The channel center frequency Fre that needs to be compensated work and bandwidth BW work , calculate the subscript value SAM_M corresponding to the frequency point to be compensated m ; and according to Fre work and bandwidth BW work Calculate the interpolation coefficient S, combine it with the compensation matrix C to complete the two-dimensional compensation coefficient calculation, and generate the spectrum correction value JZ_COE corresponding to the data to be compensated m ;

[0011] S4. Using the frequency spectrum data SAM_M_F of the frequency point to be compensated mand spectrum correction value JZ_COE m , completing the final spectrum correction.

[0012] The center frequency sequence FRE1 in step S1 1*N and effective bandwidth BW valid The specific calculation steps are as follows:

[0013] S1.1: Record the lower frequency limit FRE of the ultra-wideband channel Lower 、FRE upper limit of working frequency band Upper And instantaneous working bandwidth BW All ;

[0014] S1.2: Set the center frequency sequence FRE1 1*N , the value of the nth element in the sequence is:

[0015]

[0016] n=0,1,2,…,N-1

[0017] S1.3: Set the effective bandwidth BW valid =BW All *coff, where coff is the ratio of the acquisition data bandwidth to the analog channel bandwidth, and the value is greater than 1.

[0018] The specific calculation steps of the compensation matrix C in step S2 are:

[0019] S2.1: Initialize parameter i=0, set sampling time t1 and AD chip sampling frequency f s , where f s The following conditions must be met:

[0020] f s >2*BW Valid

[0021] S2.2: Control the ultra-wideband channel operating center frequency to FRE1(i), with frequency f s Sampling at time t1 to obtain the i-th sampling point sequence Where N s =f s *t1;

[0022] S2.3: Sampling point sequence Perform fast Fourier transform and logarithm operation to obtain the corresponding spectrum data Where FFT(·) stands for Fast Fourier Transform:

[0023]

[0024] S2.4: Setting the smoothing coefficient vector Where N fa is an even number, the spectrum data and Perform multiplication and addition operations and record the results after multiplication and addition as

[0025]

[0026] Where: p is an ordinal number.

[0027] S2.5: Settings Each element in the array corresponds to a subscript, where the subscript value of the nth element is:

[0028] -f s / 2+f s / (N s -1)*nn=0,1,2,…,N s -1

[0029] S2.6: Select all subscripts greater than -BW valid / 2 and less than BW valid / 2, recorded as SAM_F_f i , let its length be N s1 , and record the selected subscript sequence as SAM_F_Ind;

[0030] S2.7: SAM_F_fa1 i Enter into matrix The i-th row of

[0031] S2.8: Let i=i+1; if i is less than N, go to step S2.2; if i is equal to N, the module processing ends and the matrix C is output.

[0032] Step S3: spectrum correction value JZ_COE m The specific calculation steps are:

[0033] S3.1: Record various parameters of the ultra-wideband channel during normal operation, including: working center frequency Fre work , working instantaneous bandwidth BW work , total working time T work And AD sampling rate fs during operation work Assume that the total sampled data to be compensated is

[0034] S3.2: If the total length of the sampled data to be compensated exceeds the threshold value N Thre , then the sampled data to be compensated is SAMALL with N Thre Segment for segment length, record the total number of segments as M; otherwise, NThre The value of is updated to the total length of the sampled data to be compensated, and the total number of segments M is recorded to be 1; each segment of data to be compensated is recorded as Initialize the ordinal number m = 1;

[0035] S3.3: Calculate the segmented data to be compensated SAM_M m Each element in corresponds to a frequency subscript, where the subscript corresponding to the n1th element is:

[0036] -fs work / 2+fs work / (N Thre -1)*n1 n1=0,1,2,…,N Thre -1

[0037] S3.4: Get the mth segment of data to be compensated SAM_M m , perform fast Fourier transform on it and obtain its corresponding spectrum data SAM_M_F m , where FFT(·) stands for Fast Fourier Transform:

[0038] SAM_M_F m =FFT(SAM_M m )

[0039] S3.5: Initialization is a vector of all zeros, let n1 = 0;

[0040] S3.6: Get the center frequency Fre of the channel to be compensated work , and take SAM_M m The subscript corresponding to the n1th element in is recorded as Fre n1 ;

[0041] S3.7: Put FRE1 1*N Subtract Fre from all elements in work , get the new vector FRE1_TMP 1*N , and round down all elements in FRE1_TMP;

[0042] S3.8: Find the positions of the four values ​​-2, -1, 0, and 1 in FRE1_TMP, record their positions as a1, a2, a3, and a4 respectively, and subtract -2, -1, 0, and 1 from the a1, a2, a3, and a4 elements in FRE1 respectively, and record the subtracted values ​​as b1, b2, b3, and b4;

[0043] S3.9: Subtract Fre from all elements in SAM_F_Ind n1 , get the new vector SAM_F_Ind_TMP 1*N, and round down all elements in SAM_F_Ind_TMP;

[0044] S3.10: Find the positions of the four values ​​-2, -1, 0, and 1 in SAM_F_Ind_TMP, record their positions as c1, c2, c3, and c4 respectively, and subtract -2, -1, 0, and 1 from the c1, c2, c3, and c4 elements in SAM_F_Ind respectively, and record the subtracted values ​​as d1, d2, d3, and d4;

[0045] S3.11: Generate a smoothing matrix S based on b1, b2, b3, b4 and d1, d2, d3, d4 4*4 ;

[0046] S3.12: Compare the elements in S with C_TMP 4*4 Multiply the corresponding elements in the table, add the results of all the elements after multiplication, and fill the final result into JZ_COE m The n1th element of C_TMP 4*4 is defined as follows:

[0047]

[0048] S3.13: Let n1 = n1 + 1; if n1 is less than N Thre , then go to step S3.6; otherwise go to step S3.14;

[0049] S3.14: Let m=m+1; if m is less than M+1, go to step S3.3; otherwise, the module ends and outputs the spectrum data SAM_M_F corresponding to each segment of the data to be compensated. m and JZ_COE m , and initialize m to 1.

[0050] In step S3.11, a smoothing matrix S is generated based on b1, b2, b3, b4 and d1, d2, d3, d4. 4*4 The method is:

[0051]

[0052] x=1,2,3,4

[0053] y=1,2,3,4

[0054] Both x and y are ordinal numbers.

[0055] The specific calculation steps of the spectrum correction in step S4 are:

[0056] S4.1: Select SAM_M_F m and JZ_COE m , use the following formula to calculate SAM_M_Fm To make a correction:

[0057] SAM_M_F m (n1) = SAM_M_F m (n1) / 10^(JZ_COE m (n1) / 20)

[0058] S4.2: Let m = m + 1; if m is less than M + 1, go to step 1; otherwise, this module ends, SAM_M_F m That is, the spectrum data of each segment after correction; if time domain sampling data is required, SAM_M_F m Just perform an inverse Fourier transform.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] (1) The present invention implements a digital domain amplitude-frequency compensation algorithm based on a two-dimensional smoothing correction matrix, interpolation, and compensation coefficient generation method, and effectively solves the problem of poor amplitude-frequency consistency index of an ultra-wideband analog channel.

[0061] (2) The present invention achieves compensation for the amplitude-frequency characteristics of any frequency point in the ultra-wideband channel by establishing a two-dimensional smoothing correction matrix for the full-band of the sweeping receiver, interpolation, and real-time calculation of compensation coefficients, thereby solving the problem that the traditional method can only be used for amplitude compensation within the instantaneous working bandwidth and is difficult to cover the ultra-wideband spectrum.

[0062] (3) The present invention implements a digital domain amplitude-frequency compensation algorithm based on a two-dimensional smoothing correction matrix, interpolation, and compensation coefficient generation method, which greatly reduces the system's requirements for analog channels and reduces the difficulty of system design.

[0063] (4) The present invention implements a digital domain amplitude-frequency compensation algorithm based on a two-dimensional smoothing correction matrix, interpolation, and compensation coefficient generation method, reduces the hardware debugging time of the analog channel, and ensures the rapid implementation and verification of the subsequent signal processing algorithm on the hardware platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 The figure is a flowchart of the steps of the method of the present invention.

[0065] Figure 2 This is the connection block diagram of an ultra-wideband channel system.

[0066] Figure 3 It is the time domain and spectrum diagram of sampling data of some central frequency points.

[0067] Figure 4 The compensation matrix C is calculated.

[0068] Figure 5 Time domain and spectrum diagram of some center frequency sampling data after compensation.

[0069] Figure 6 To compensate for the intra-band fluctuations and total energy changes at different center frequencies before and after. DETAILED DESCRIPTION

[0070] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0071] Taking an actual ultra-wideband channel system as an example, the channel operating frequency range is 28GHz-35GHz, the instantaneous working bandwidth is 500MHz, and the maximum sampling rate of the back-end acquisition system is 2560MHz. Figure 2 The system connection block diagram is given. Figure 3 The channel sampling data and spectrum characteristics without compensation are given. The control channel center frequency changes from 28.25GHz to 34.75GHz, the change step is 500MHz, the instantaneous working bandwidth is 500MHz, the acquisition system sampling frequency is 1280MHz, and the acquisition time of each center frequency is 100us. The uncompensated channel output signal is collected, and a total of 14 segments of acquisition data are obtained. The time domain and spectrum diagrams of some frequency point data are shown in Figure 3 As shown, it can be seen that the amplitude within the band of each center frequency point has different degrees of random fluctuations, and there are also differences in the channel amplitude between different center frequencies.

[0072] A two-dimensional correction method for amplitude characteristics of ultra-wideband channels, such as Figure 1 As shown, including:

[0073] S1, according to the lower limit of the channel working frequency band FRE Lower 、FRE upper limit of working frequency band Upper And instantaneous working bandwidth BW All , calculate the center frequency points required to establish the calibration matrix and collect data, and generate the center frequency point sequence FRE1 1*N At the same time, according to the instantaneous working bandwidth of the channel, set the effective bandwidth BW valid :

[0074] S1.1. Record the lower limit FRE of the operating frequency band of the ultra-wideband channel Lower 、FRE upper limit of working frequency band Upper And instantaneous working bandwidth BW All In this embodiment, FRE Lower 28GHz, FRE Upper The instantaneous working bandwidth BW is 35GHz. All 500MHz;

[0075] S1.2: Set the center frequency sequence FRE1 1*N , there are 131 center frequency points in total, and the value of the nth element in the sequence is:

[0076]

[0077] n=0,1,2,…,N-1

[0078] S1.3: Set the effective bandwidth BW valid It is 640MHz.

[0079] S2, the control channel works at the set center frequency sequence FRE1, collects data of the i-th center frequency, and obtains the collected data sequence Calculate and smooth the spectrum of the data sequence to obtain the spectrum characteristics corresponding to the i-th center frequency point After all center frequency points are processed, the compensation matrix C is generated:

[0080] S2.1: Initialize parameter i=0, set sampling time t1 to 3us and AD chip sampling frequency f s is 2560MHz, f s The following conditions must be met:

[0081] f s >2*BW Valid

[0082] S2.2: Control the ultra-wideband channel operating center frequency to FRE1(i), with frequency f s Sampling at time t1 to obtain the i-th sampling point sequence Where N s =f s *t1;

[0083] S2.3: Sampling point sequence Perform fast Fourier transform and logarithm operation to obtain the corresponding spectrum data Where FFT(·) stands for Fast Fourier Transform:

[0084]

[0085] S2.4: Setting the smoothing coefficient vector Where N fa is an even number, the spectrum data and Perform multiplication and addition operations and record the results after multiplication and addition as

[0086]

[0087] S2.5: Settings Each element in the array corresponds to a subscript, where the subscript value of the nth element is:

[0088] -f s / 2+f s / (N s -1)*nn=0,1,2,…,N s -1

[0089] S2.6: Select all subscripts greater than -BW valid / 2 and less than BW valid / 2, recorded as SAM_F_fa1 i , let its length be N s1 , and record the selected subscript sequence as SAM_F_Ind;

[0090] S2.7: SAM_F_fa1 i Enter into matrix The i-th row of

[0091] S2.8: Let i=i+1; if i is less than N, go to step S2.2; if i is equal to N, the module processing ends and the matrix C is output. The size of the final matrix C is 131*3840. Figure 4 The distribution of the compensation matrix C value is given;

[0092] S3. Collect channel data to be compensated The channel center frequency Fre that needs to be compensated work and bandwidth BW work , calculate the subscript value SAM_M corresponding to the frequency point to be compensated m ; and according to Fre work and bandwidth BW work Calculate the interpolation coefficient S, combine it with the compensation matrix C to complete the two-dimensional compensation coefficient calculation, and generate the spectrum correction value JZ_COE corresponding to the data to be compensated m ; Taking the 28.25 GHz center frequency point to be compensated as an example, other center frequencies to be compensated are similar.

[0093] S3.1: Record various parameters of the ultra-wideband channel during normal operation, including: working center frequency Fre work The working instantaneous bandwidth BW is 28.25GHz work 500MHz, total working time T work 100us and AD sampling rate fs during operation work is 1280MHz. Assume that the total sampled data to be compensated is

[0094] S3.2: Setting the threshold value NThre is 12800, and the sample data to be compensated is SAMALL with N Thre Segment for segment length, record the total number of segments as M is 10; initialize m = 1;

[0095] S3.3: Calculate the segmented data to be compensated SAM_M m Each element in corresponds to a frequency subscript, where the subscript corresponding to the n1th element is:

[0096] -fs work / 2+fs work / (N Thre -1)*n1 n1=0,1,2,…,N Thre -1

[0097] S3.4: Get the mth segment of data to be compensated SAM_M m , perform fast Fourier transform on it and obtain its corresponding spectrum data SAM_M_F m , where FFT(·) stands for Fast Fourier Transform:

[0098] SAM_M_F m =FFT(SAM_M m )

[0099] S3.5: Initialization is a vector of all zeros, let n1 = 0;

[0100] S3.6: Get the center frequency Fre of the channel to be compensated work , and take SAM_M m The subscript corresponding to the n1th element in is recorded as Fre n1 ;

[0101] S3.7: Put FRE1 1*N Subtract Fre from all elements in work , get the new vector FRE1_TMP 1*N , and round down all elements in FRE1_TMP;

[0102] S3.8: Find the positions of the four values ​​-2, -1, 0, and 1 in FRE1_TMP, record their positions as a1, a2, a3, and a4 respectively, and subtract -2, -1, 0, and 1 from the a1, a2, a3, and a4 elements in FRE1 respectively, and record the subtracted values ​​as b1, b2, b3, and b4;

[0103] S3.9: Subtract Fre from all elements in SAM_F_Ind n1 , get the new vector SAM_F_Ind_TMP 1*N, and round down all elements in SAM_F_Ind_TMP;

[0104] S3.10: Find the positions of the four values ​​-2, -1, 0, and 1 in SAM_F_Ind_TMP, record their positions as c1, c2, c3, and c4 respectively, and subtract -2, -1, 0, and 1 from the c1, c2, c3, and c4 elements in SAM_F_Ind respectively, and record the subtracted values ​​as d1, d2, d3, and d4;

[0105] S3.11: Generate a smoothing matrix S based on b1, b2, b3, b4 and d1, d2, d3, d4 4*4 :

[0106]

[0107] x=1,2,3,4

[0108] y=1,2,3,4

[0109] Both x and y are ordinal numbers.

[0110] S3.12: Compare the elements in S with C_TMP 4*4 Multiply the corresponding elements in the table, add the results of all the elements after multiplication, and fill the final result into JZ_COE m The n1th element of C_TMP 4*4 is defined as follows:

[0111]

[0112] S3.13: Let n1 = n1 + 1; if n1 is less than N Thre , then go to step S3.6; otherwise go to step S3.14;

[0113] S3.14: Let m=m+1; if m is less than M+1, go to step S3.3; otherwise, the module ends and outputs the spectrum data SAM_M_F corresponding to each segment of the data to be compensated. m and JZ_COE m , and initialize m to 1.

[0114] S4. Using the frequency spectrum data SAM_M_F of the frequency point to be compensated m and spectrum correction value JZ_COE m , complete the final spectrum correction

[0115] S4.1: Select SAM_M_F m and JZ_COE m , use the following formula to calculate SAM_M_F m To make a correction:

[0116] SAM_M_F m (n1) = SAM_M_F m (n1) / 10^(JZ_COE m (n1) / 20)

[0117] S4.2: Let m = m + 1; if m is less than M + 1, go to step 1; otherwise, this module ends, SAM_M_F m That is, the spectrum data of each segment after correction; if time domain sampling data is required, SAM_M_F m Just perform an inverse Fourier transform.

[0118] According to the above process, 14 frequency points within 28.25GHz to 34.75GHz are compensated. Figure 5 The time domain and spectrum diagrams of some frequency point data after compensation are given. It can be seen that the in-band fluctuation amplitude is significantly reduced after compensation, and the channel amplitude consistency between different center frequency points is good. Figure 6 The in-band fluctuation and total energy changes of different center frequencies before and after compensation are given. The in-band fluctuation of each center frequency is reduced from a maximum of 12dB to 0.9dB before and after compensation, and the maximum difference in total energy of each center frequency is reduced from 7.86dB to 0.5dB. The channel amplitude consistency is significantly improved, which verifies the effectiveness of the proposed method.

[0119] A two-dimensional correction device for amplitude characteristics of an ultra-wideband channel, comprising a preprocessing module, a spectrum characteristic calculation and smoothing module, a two-dimensional compensation coefficient calculation module, and a spectrum correction module:

[0120] In the preprocessing module, according to the channel working frequency band parameters and instantaneous working bandwidth parameters, the center frequency points required for establishing the calibration matrix to collect data are calculated to generate a center frequency point sequence. At the same time, considering that the effective bandwidth in the band is smaller than the instantaneous bandwidth of the channel, the effective bandwidth width needs to be set; in the spectrum characteristic calculation and smoothing module, the channel is controlled to work in the set center frequency point sequence, and the data of each center frequency point is collected, the spectrum characteristics are calculated and smoothed until all center frequency points are processed and the compensation matrix is ​​generated; in the two-dimensional compensation coefficient calculation module, the values ​​of each compensation frequency point are calculated according to the center frequency point and bandwidth of the channel to be compensated. Subsequently, the interpolation coefficient is calculated according to the value of each compensation frequency point, and the two-dimensional compensation coefficient calculation is completed based on the compensation matrix to generate the spectrum correction value corresponding to each compensation frequency point; finally, in the spectrum correction module, the spectrum to be corrected is added to the spectrum correction value to complete the final spectrum correction.

[0121] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

[0122] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims

1. A two-dimensional correction method for amplitude characteristics of an ultra-wideband channel, characterized in that: include: S1, according to the lower limit of the channel working frequency band FRE Lower 、FRE upper limit of working frequency band Upper And instantaneous working bandwidth BW All , calculate the center frequency points required to establish the calibration matrix and collect data, and generate the center frequency point sequence FRE1 1*N At the same time, according to the instantaneous working bandwidth of the channel, set the effective bandwidth BW valid ; S2, the control channel works at the center frequency sequence FRE1, collects data of the i-th center frequency, and obtains the collected data sequence Calculate and smooth the spectrum of the data sequence to obtain the spectrum characteristics corresponding to the i-th center frequency point After all center frequency points are processed, the compensation matrix C is generated; S3. Collect channel data to be compensated The channel center frequency Fre that needs to be compensated work and bandwidth BW work , calculate the subscript value SAM_M corresponding to the frequency point to be compensated m ; and according to Fre work and bandwidth BW work Calculate the interpolation coefficient S, combine it with the compensation matrix C to complete the two-dimensional compensation coefficient calculation, and generate the spectrum correction value JZ_COE corresponding to the data to be compensated m ; S4. Using the frequency spectrum data SAM_M_F of the frequency point to be compensated m and spectrum correction value JZ_COE m , completing the final spectrum correction.

2. The two-dimensional correction method for amplitude characteristics according to claim 1, characterized in that: The center frequency sequence FRE1 in step S1 1*N and effective bandwidth BW valid The specific calculation steps are as follows: S1.1: Record the lower frequency limit FRE of the ultra-wideband channel Lower 、FRE upper limit of working frequency band Upper And instantaneous working bandwidth BW All ; S1.2: Set the center frequency sequence FRE1 1*N , the value of the nth element in the sequence is: n=0,1,2,…,N-1 S1.3: Set the effective bandwidth BW valid =BW All *coff, where coff is the ratio of the acquisition data bandwidth to the analog channel bandwidth, and the value is greater than 1.

3. The two-dimensional correction method for amplitude characteristics according to claim 1, characterized in that: The specific calculation steps of the compensation matrix C in step S2 are: S2.1: Initialize the sequence number i=0, set the sampling time t1 and the AD chip sampling frequency f s , where f s The following conditions must be met: f s >2*BW Valid S2.2: Control the ultra-wideband channel operating center frequency to FRE1(i), with frequency f s Sampling at time t1 to obtain the i-th sampling point sequence Where N s =f s *t1; S2.3: Sampling point sequence Perform fast Fourier transform and logarithm operation to obtain the corresponding spectrum data Where FFT(·) stands for Fast Fourier Transform: S2.4: Setting the smoothing coefficient vector Where N fa is an even number, the spectrum data and Perform multiplication and addition operations and record the results after multiplication and addition as Where: p is an ordinal number; S2.5: Settings Each element in the array corresponds to a subscript, where the subscript value of the nth element is: -f s / 2+f s / (N s -1)*n n=0,1,2,…,N s -1 S2.6: Select all subscripts greater than -BW valid / 2 and less than BW valid / 2, recorded as SAM_F_fa1 i , let its length be N s1 , and record the selected subscript sequence as SAM_F_Ind; S2.7: SAM_F_fa1 i Record the i-th row of matrix C. The matrix size of matrix C is N*N s1 ; S2.8: Let i=i+1; if i is less than N, go to step S2.2; if i is equal to N, the process ends and the matrix C is output.

4. The two-dimensional correction method for amplitude characteristics according to claim 3, characterized in that: Step S3: spectrum correction value JZ_COE m The specific calculation steps are: S3.1: Record various parameters of the ultra-wideband channel when it is working normally, including: the working center frequency Fre work , working instantaneous bandwidth BW work , total working time T work And AD sampling rate fs during operation work ; Assume the total sampled data to be compensated is S3.2: If the total length of the sampled data to be compensated exceeds the threshold value N Thre , then the sampled data to be compensated is SAMALL with N Thre Segment for segment length, record the total number of segments as M; otherwise, N Thre The value of is updated to the total length of the sampled data to be compensated, and the total number of segments M is recorded to be 1; each segment of data to be compensated is recorded as Initialize the ordinal number m = 1; S3.3: Calculate the segmented data to be compensated SAM_M m Each element in corresponds to a frequency subscript, where the subscript corresponding to the n1th element is: -fs work / 2+fs work / (N Thre -1)*n1 n1=0,1,2,…,N Thre -1 S3.4: Get the mth segment of data to be compensated SAM_M m , perform fast Fourier transform on it and obtain its corresponding spectrum data SAM_M_F m , where FFT(·) stands for Fast Fourier Transform: SAM_M_F m =FFT(SAM_M m ) S3.5: Initialization is a vector of all zeros, let n1 = 0; S3.6: Get the center frequency Fre of the channel to be compensated work , and take SAM_M m The subscript corresponding to the n1th element in is recorded as Fre n1 ; S3.7: Put FRE1 1*N Subtract Fre from all elements in work , get the new vector FRE1_TMP 1*N , and round down all elements in FRE1_TMP; S3.8: Find the positions of the four values ​​-2, -1, 0, and 1 in FRE1_TMP, record their positions as a1, a2, a3, and a4 respectively, and subtract -2, -1, 0, and 1 from the a1, a2, a3, and a4 elements in FRE1 respectively, and record the subtracted values ​​as b1, b2, b3, and b4; S3.9: Subtract Fre from all elements in SAM_F_Ind n1 , get the new vector SAM_F_Ind_TMP 1*N , and round down all elements in SAM_F_Ind_TMP; S3.10: Find the positions of the four values ​​-2, -1, 0, and 1 in SAM_F_Ind_TMP, record their positions as c1, c2, c3, and c4 respectively, and subtract -2, -1, 0, and 1 from the c1, c2, c3, and c4 elements in SAM_F_Ind respectively, and record the subtracted values ​​as d1, d2, d3, and d4; S3.11: Generate a smoothing matrix S based on b1, b2, b3, b4 and d1, d2, d3, d4 4*4 ; S3.12: Compare the elements in S with C_TMP 4*4 Multiply the corresponding elements in the table, add the results of all the elements after multiplication, and fill the final result into JZ_COE m The n1th element of C_TMP 4*4 is defined as follows: S3.13: Let n1 = n1 + 1; if n1 is less than N Thre , then go to step S3.6; otherwise go to step S3.14; S3.14: Let m=m+1; if m is less than M+1, go to step S3.3; otherwise, the process ends and the spectrum data SAM_M_F corresponding to each segment of the data to be compensated is output. m and JZ_COE m , and initialize m to 1.

5. The two-dimensional correction method for amplitude characteristics according to claim 4, characterized in that: In step S3.11, a smoothing matrix S is generated based on b1, b2, b3, b4 and d1, d2, d3, d4. 4*4 The method is: x＝1,2,3,4 y＝1,2,3,4 Both x and y are ordinal numbers.

6. The two-dimensional correction method for amplitude characteristics according to claim 4, characterized in that: The specific calculation steps of the spectrum correction in step S4 are: S4.1: Select SAM_M_F m and JZ_COE m , use the following formula to calculate SAM_M_F m To make a correction: SAM_M_F m (n1)=SAM_M_F m (n1) / 10^(JZ_COE m (n1) / 20) S4.2: Let m = m + 1; if m is less than M + 1, go to step 1; otherwise, this module ends, SAM_M_F m That is, the spectrum data of each segment after correction; if time domain sampling data is required, SAM_M_F m Just perform an inverse Fourier transform.

7. A two-dimensional correction device for amplitude characteristics of an ultra-wideband channel, characterized in that: Including pre-processing module, spectrum characteristic calculation and smoothing module, two-dimensional compensation coefficient calculation module, spectrum correction module: In the pre-processing module, the central frequency points required for establishing the calibration matrix and collecting data are calculated according to the channel working frequency band parameters and the instantaneous working bandwidth parameters, and a central frequency point sequence is generated. At the same time, considering that the effective bandwidth in the band is smaller than the instantaneous bandwidth of the channel, the effective frequency bandwidth needs to be set; in the spectrum characteristic calculation and smoothing module, the channel is controlled to work in the set central frequency point sequence, and the data of each central frequency point is collected, the spectrum characteristics are calculated and smoothed until all the central frequency points are processed, and a compensation matrix is ​​generated; In the two-dimensional compensation coefficient calculation module, each compensation frequency value is calculated according to the channel center frequency and bandwidth that need to be compensated; Then, the interpolation coefficients are calculated according to the values ​​of each compensation frequency point, and the two-dimensional compensation coefficients are calculated based on the compensation matrix to generate the spectrum correction values ​​corresponding to each compensation frequency point; Finally, in the spectrum correction module, the spectrum to be corrected is added to the spectrum correction value to complete the final spectrum correction.

8. A computer-readable storage medium having computer program instructions stored thereon, wherein when the computer program instructions are loaded and executed by a processor, the processor is caused to execute the method according to any one of claims 1 to 6.

9. A computer program product stored on a non-transitory computer readable medium, the computer program product comprising a program code for carrying out the method according to any one of claims 1 to 6.

Images