A two-dimensional amplitude characteristic correction method suitable for ultra-wideband channels

Abstract

A two-dimensional amplitude characteristic correction method suitable for ultra-wideband channels includes: S1, calculating the center frequency points required for establishing the calibration matrix acquisition data based on the channel's operating frequency band parameters and instantaneous operating bandwidth parameters, and generating a center frequency point sequence FRE1. 1*N S2. The control channel operates at the pre-set center frequency sequence FRE1, acquiring data, calculating spectral characteristics, and smoothing data for each center frequency until all center frequencies are processed, generating a compensation matrix C; S3. Calculate the values ​​of each compensation frequency point based on the required channel center frequency and bandwidth; then calculate interpolation coefficients based on each compensation frequency point value, complete the two-dimensional compensation coefficient calculation based on the compensation matrix, and generate the spectral correction value JZ_COE corresponding to each compensation frequency point. m S4. Based on the output of step S3, add the spectrum to be calibrated to the spectrum correction value to complete the final spectrum calibration.

Description

Technical Field

[0001] This invention relates to a two-dimensional amplitude characteristic correction method applicable to ultra-wideband channels, particularly a correction method when the channel amplitude characteristic changes two-dimensionally with the switching of frequency and local oscillator frequency within the instantaneous operating bandwidth, belonging to the field of signal processing technology. Background Technology

[0002] With the continuous iteration and development of design technology and application requirements, microwave channels and devices used in various fields are showing a trend of increasingly wider bandwidths. To ensure system performance, the designed microwave channel must have flat amplitude performance within its operating bandwidth. Due to the influence of cascaded components such as amplitude limiting protection circuits, low-noise amplifiers, and multi-stage power amplifiers, the actual amplitude-frequency response often fluctuates significantly with frequency within the operating bandwidth. This channel characteristic often leads to severe distortion of the received spectrum, affecting the dynamic range and sensitivity of the receiving system. Correction and compensation are needed in practical applications to ensure channel flatness.

[0003] Currently, existing correction technologies can be divided into two categories: analog correction and digital correction. Analog correction is generally performed by constructing an equalizer, which consists of lumped parameters such as inductors (L), capacitors (C), and resistors (R), and is essentially an inverse filter. Analog equalizers have the advantages of high processing speed, small size, low power consumption, and minimal latency. However, they require the pre-acquisition of amplitude-frequency characteristics over the entire operating bandwidth and can only compensate for fixed-variation types such as monotonic rises and falls or convex and concave variations in amplitude-frequency characteristics. Therefore, their reliability is low, and they cannot be applied to ultra-wideband channels with complex amplitude-frequency characteristic variations. Digital correction utilizes digital signal processing algorithms, offering better reliability. During digital correction, noise floor data at each frequency point of the channel is collected, and the collected data is analyzed to estimate the amplitude inconsistencies at each frequency point. For example, correction methods based on the subspace principle perform joint online estimation of amplitude-frequency errors at each frequency point, finding the optimal solution through iterative calculations. Some researchers have approximated the amplitude error characteristics of broadband channels using ARMA models, achieving amplitude-frequency error estimation for broadband incident signals. Furthermore, by establishing a broadband channel model and solving for the model parameters using collected data, the amplitude variation characteristics of the broadband channel are fitted, and compensation is completed based on this. 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 an architecture of fixed local oscillator + adjustable local oscillator + amplification and attenuation module + filtering module. The amplitude variation characteristics of the channel are determined by the local oscillator frequency and the amplification and filtering modules, exhibiting a two-dimensional distribution. However, existing methods only focus on compensating for amplitude and phase inconsistencies within the instantaneous operating bandwidth, ignoring the gain changes caused by local oscillator switching, resulting in poor compensation effects across the entire bandwidth. At the same time, they are limited by the spectral resolution of digital signals and cannot adapt to situations with drastic in-band changes. Summary of the Invention

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

[0005] The objective of this invention is achieved through the following technical solutions:

[0006] A two-dimensional amplitude characteristic correction method applicable to ultra-wideband channels is proposed. First, raw noise data under different local oscillator and bandwidth conditions within the ultra-wideband channel is acquired. The acquired results are transformed to the frequency domain and filtered. Based on this, a two-dimensional smoothing correction matrix is ​​established. When correction is required, the compensation value is calculated in real time based on the current operating center frequency and bandwidth by calculating interpolation coefficients and combining them with the two-dimensional smoothing correction matrix, thus completing channel compensation at any frequency point. Experimental results show that this method can effectively compensate for channel amplitude inconsistencies across the entire bandwidth.

[0007] Specifically, the steps include the following:

[0008] S1, based on the lower limit of the channel's operating frequency band (FRE) Lower Operating frequency band upper limit FRE Upper and instantaneous operating bandwidth BW All The center frequency points required for establishing the calibration matrix data acquisition are calculated, and the center frequency point sequence FRE1 is generated. 1*N Simultaneously, the effective bandwidth BW is set based on the instantaneous operating bandwidth of the channel. valid ;

[0009] S2. The control channel operates at the pre-set center frequency sequence FRE1, acquiring data from the i-th center frequency point to obtain the acquired data sequence. Perform spectral calculation and smoothing on the data sequence to obtain the spectral characteristics corresponding to the i-th center frequency point. After all center frequency points have been processed, compensation matrix C is generated;

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

[0011] S4. Using the spectrum data of the frequency point to be compensated, SAM_M_F mWith the spectrum correction value JZ_COE m This completes 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 limit of the operating frequency band (FRE) of the ultra-wideband channel. Lower Operating frequency band upper limit FRE Upper and instantaneous operating 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 for the compensation matrix C in step S2 are as follows:

[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 operating center frequency of the ultra-wideband channel to FRE1(i), with frequency f s Perform sampling at time t1 to obtain the sequence of the i-th sampling point. Where N s =f s *t1;

[0022] S2.3: For the sampling point sequence Perform a Fast Fourier Transform and logarithm operation to obtain the corresponding spectral data. Where FFT(·) represents Fast Fourier Transform:

[0023]

[0024] S2.4: Set the smoothing coefficient vector Where N fa If the number is even, the spectrum data will be... and Perform a multiplication-accumulation operation and record the result.

[0025]

[0026] Where p is the ordinal number.

[0027] S2.5: Settings The elements in the array correspond to their indices, where the index 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 indices greater than -BW valid / 2 and less than BW valid The element value of / 2 is recorded as SAM_F_f i Let the length of a be N. s1 And record the selected index sequence as SAM_F_Ind;

[0030] S2.7: Set SAM_F_fa1 i Record in matrix The i-th row;

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

[0032] Step S3: Spectrum correction value JZ_COE m The specific calculation steps are as follows:

[0033] S3.1: Records various parameters of the ultra-wideband channel during normal operation, mainly including: the operating center frequency Fre work Working instantaneous bandwidth BW work Total working hours T work and AD sampling rate fs during operation work Let the total amount of sampled data to be compensated be...

[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 will be SAMALL with N Thre Divide the segment into segments according to its length, and record the total number of segments as M; otherwise, divide it into N segments.Thre The value is updated to the total length of the sampled data to be compensated, and the total number of segments M is recorded as 1; each segment of data to be compensated is recorded as... Initialize ordinal m = 1;

[0035] S3.3: Calculate the segmented data to be compensated, SAM_M m The frequency indices of each element in the table are given, where the index of the n1th element is:

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

[0037] S3.4: Take the m-th segment of data to be compensated, SAM_M m Perform a Fast Fourier Transform on it to obtain its corresponding spectral data SAM_M_F m Where FFT(·) represents Fast Fourier Transform:

[0038] SAM_M_F m =FFT(SAM_M m )

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

[0040] S3.6: Take the center frequency Fre of the channel that needs compensation. work Simultaneously, take SAM_M m The index record corresponding to the n1th element is Fre. n1 ;

[0041] S3.7: FRE1 1*N Subtract Fre from all elements work This yields 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, and record their positions as a1, a2, a3, and a4 respectively. 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 This yields the new vector SAM_F_Ind_TMP. 1*NAnd 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, and record their positions as c1, c2, c3, and c4 respectively. 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: Combine the elements in S with C_TMP 4*4 Multiply each element correspondingly, sum the results of all multiplications, and fill the final result into JZ_COE. m Take the n1th element, where C_TMP 4*4 The definition is as follows:

[0047]

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

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

[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 for spectrum correction in step S4 are as follows:

[0056] S4.1: Select SAM_M_F m and JZ_COE m Using the following formula for SAM_M_Fm Perform 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, then go to step 1; otherwise, this module ends, SAM_M_F m This refers to the spectral data of each segment after correction; if time-domain sampling data is required, then SAM_M_F... m This can be achieved by performing an inverse Fourier transform.

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

[0060] (1) Based on the two-dimensional smooth correction matrix, interpolation and compensation coefficient generation method, the present invention realizes the digital domain amplitude-frequency compensation algorithm, which effectively solves the problem of poor amplitude-frequency consistency index of ultra-wideband analog channel.

[0061] (2) This invention achieves compensation for amplitude-frequency characteristics at any frequency point in the ultra-wideband channel by establishing a two-dimensional smooth correction matrix for the full-band sweep receiver, interpolation and real-time calculation of compensation coefficients, etc., and solves the problem that traditional methods can only be used for amplitude compensation within the instantaneous working bandwidth and are difficult to cover the ultra-wideband spectrum.

[0062] (3) Based on the two-dimensional smooth correction matrix, interpolation and compensation coefficient generation method, the present invention realizes the digital domain amplitude-frequency compensation algorithm, which greatly reduces the requirements of the system for the analog channel and reduces the design difficulty of the system.

[0063] (4) Based on the two-dimensional smooth correction matrix, interpolation and compensation coefficient generation method, the present invention realizes the digital domain amplitude-frequency compensation algorithm, 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. Attached Figure Description

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

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

[0066] Figure 3 The time domain and spectrum diagrams of sampled data at some center frequency points are shown.

[0067] Figure 4 Let C be the calculated compensation matrix.

[0068] Figure 5 Time domain and spectrum diagram of the sampled data at the center frequency point after compensation.

[0069] Figure 6 To compensate for fluctuations and changes in total energy at different center frequency points before and after. Detailed Implementation

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

[0071] Taking a real-world ultra-wideband channel system as an example, the channel operates in the frequency range of 28GHz-35GHz, with an instantaneous operating bandwidth of 500MHz, and the back-end acquisition system has a maximum sampling rate of 2560MHz. Figure 2 A system connection diagram is provided. Figure 3 The uncompensated channel sampling data and spectral characteristics are presented. The channel center frequency was controlled to vary from 28.25 GHz to 34.75 GHz in 500 MHz increments, with an instantaneous operating bandwidth of 500 MHz. The acquisition system sampling frequency was 1280 MHz, and the acquisition time for each center frequency was 100 μs. The uncompensated channel output signal was acquired, resulting in 14 data segments. The time domain and spectrum of some frequency points are shown below. Figure 3 As shown, it can be seen that the amplitude within each center frequency band exhibits varying degrees of random fluctuation, and there are also differences in channel amplitude between different center frequencies.

[0072] A two-dimensional amplitude characteristic correction method suitable for ultra-wideband channels, such as Figure 1 As shown, it includes:

[0073] S1, based on the lower limit of the channel's operating frequency band (FRE) Lower Operating frequency band upper limit FRE Upper and instantaneous operating bandwidth BW All The center frequency points required for establishing the calibration matrix data acquisition are calculated, and the center frequency point sequence FRE1 is generated. 1*N Simultaneously, the effective bandwidth BW is set based on the instantaneous operating bandwidth of the channel. valid :

[0074] S1.1 Record the lower limit of the operating frequency band (FRE) of the ultra-wideband channel. Lower Operating frequency band upper limit FRE Upper and instantaneous operating bandwidth BW All In this embodiment, FRE Lower For 28GHz, FRE Upper 35GHz, instantaneous operating bandwidth BW All 500MHz;

[0075] S1.2: Set the center frequency sequence FRE1 1*N A total of 131 center frequency points were set, 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 operates at the pre-set center frequency sequence FRE1, acquiring data from the i-th center frequency point to obtain the acquired data sequence. Perform spectral calculation and smoothing on the data sequence to obtain the spectral characteristics corresponding to the i-th center frequency point. After all center frequency points have been processed, 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 For 2560MHz, f s The following conditions must be met:

[0081] f s >2*BW Valid

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

[0083] S2.3: For the sampling point sequence Perform a Fast Fourier Transform and logarithm operation to obtain the corresponding spectral data. Where FFT(·) represents Fast Fourier Transform:

[0084]

[0085] S2.4: Set the smoothing coefficient vector Where N fa If the number is even, the spectrum data will be... and Perform a multiplication-accumulation operation and record the result.

[0086]

[0087] S2.5: Settings The elements in the array correspond to their indices, where the index 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 indices greater than -BW valid / 2 and less than BW valid The element value of / 2 is recorded as SAM_F_fa1 i Let its length be N. s1 And record the selected index sequence as SAM_F_Ind;

[0090] S2.7: Set SAM_F_fa1 i Record in matrix The i-th row;

[0091] S2.8: Let i = i + 1; if i is less than N, then go to step S2.2; if i equals N, then this module ends processing and outputs matrix C. The final matrix C has a size of 131 * 3840. Figure 4 The distribution of the compensation matrix C values ​​is given;

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

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

[0094] S3.2: Set the threshold value NThre The value is 12800, and the sampled data to be compensated is SAMALL with N. Thre Divide the segment into segments according to the segment length, and record the total number of segments as M = 10; initialize m = 1;

[0095] S3.3: Calculate the segmented data to be compensated, SAM_M m The frequency indices of each element in the table are given, where the index of the n1th element is:

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

[0097] S3.4: Take the m-th segment of data to be compensated, SAM_M m Perform a Fast Fourier Transform on it to obtain its corresponding spectral data SAM_M_F m Where FFT(·) represents Fast Fourier Transform:

[0098] SAM_M_F m =FFT(SAM_M m )

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

[0100] S3.6: Take the center frequency Fre of the channel that needs compensation. work Simultaneously, take SAM_M m The index record corresponding to the n1th element is Fre. n1 ;

[0101] S3.7: FRE1 1*N Subtract Fre from all elements work This yields 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, and record their positions as a1, a2, a3, and a4 respectively. 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 This yields the new vector SAM_F_Ind_TMP. 1*NAnd 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, and record their positions as c1, c2, c3, and c4 respectively. 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: Combine the elements in S with C_TMP 4*4 Multiply each element correspondingly, sum the results of all multiplications, and fill the final result into JZ_COE. m Take the n1th element, where C_TMP 4*4 The definition is as follows:

[0111]

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

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

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

[0115] S4.1: Select SAM_M_F m and JZ_COE m Using the following formula for SAM_M_F m Perform 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, then go to step 1; otherwise, this module ends, SAM_M_F m This refers to the spectral data of each segment after correction; if time-domain sampling data is required, then SAM_M_F... m This can be achieved by performing an inverse Fourier transform.

[0118] Compensation was performed on 14 frequency points between 28.25GHz and 34.75GHz following the above procedure. Figure 5 The time domain and spectrum of some frequency 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 frequencies is good. Figure 6 The changes in in-band ripple and total energy at different center frequencies before and after compensation are presented. The in-band ripple at each center frequency decreased from a maximum of 12dB to 0.9dB before and after compensation, and the maximum difference in total energy at each center frequency decreased from 7.86dB to 0.5dB. The channel amplitude consistency was significantly improved, verifying the effectiveness of the proposed method.

[0119] A two-dimensional amplitude characteristic correction device suitable for ultra-wideband channels includes 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, the center frequencies required for acquiring data to establish the calibration matrix are calculated based on the channel's operating frequency band parameters and instantaneous operating bandwidth parameters, generating a center frequency sequence. Considering that the effective bandwidth within the band is less than the channel's instantaneous bandwidth, the effective bandwidth also needs to be set. In the spectrum characteristic calculation and smoothing module, the channel is controlled to operate within the set center frequency sequence. Data is acquired, spectrum characteristics are calculated, and smoothing is performed for each center frequency until all center frequencies are processed, generating a compensation matrix. In the two-dimensional compensation coefficient calculation module, the values ​​of each compensation frequency point are calculated based on the required channel center frequencies and bandwidth. Interpolation coefficients are then calculated based on these compensation frequency point values, and the two-dimensional compensation coefficient calculation is completed based on the compensation matrix, generating the spectrum correction value corresponding to each compensation frequency point. Finally, in the spectrum correction module, the spectrum to be calibrated is added to the spectrum correction value to complete the final spectrum correction.

[0121] The contents not described in detail in this specification are common knowledge to those skilled in the art.

[0122] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

Claims

1. A two-dimensional amplitude characteristic correction method suitable for ultra-wideband channels, characterized in that, include: S1, based on the lower limit of the channel's operating frequency band (FRE) Lower Operating frequency band upper limit FRE Upper and instantaneous operating bandwidth BW All The center frequency points required for establishing the calibration matrix data acquisition are calculated, and the center frequency point sequence FRE1 is generated. 1*N Simultaneously, the effective bandwidth BW is set based on the instantaneous operating bandwidth of the channel. valid ; S2. The control channel operates at the center frequency sequence FRE1, acquiring data from the i-th center frequency point to obtain the acquired data sequence. Perform spectral calculation and smoothing on the data sequence to obtain the spectral characteristics corresponding to the i-th center frequency point. After all center frequency points have been processed, compensation matrix C is generated; S3. Collect channel data to be compensated. The channel center frequency Fre that needs compensation work and bandwidth BW work Calculate the index value SAM_M corresponding to the frequency point to be compensated. m ; and according to Fre work and bandwidth BW work Calculate the interpolation coefficients S, and combine them with the compensation matrix C to complete the two-dimensional compensation coefficient calculation, generating the spectral correction value JZ_COE corresponding to the data to be compensated. m ; S4. Using the spectrum data of the frequency point to be compensated, SAM_M_F m With the spectrum correction value JZ_COE m This completes the final spectrum correction.

2. The two-dimensional amplitude characteristic correction method 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 limit of the operating frequency band (FRE) of the ultra-wideband channel. Lower Operating frequency band upper limit FRE Upper and instantaneous operating 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 amplitude characteristic correction method according to claim 1, characterized in that, The specific calculation steps for the compensation matrix C in step S2 are as follows: S2.1: Initialize ordinal number i = 0, set sampling time t1 and AD chip sampling frequency f s , where f s The following conditions must be met: f s >2*BW Valid S2.2: Control the operating center frequency of the ultra-wideband channel to FRE1(i), with frequency f s Perform sampling at time t1 to obtain the sequence of the i-th sampling point. Where N s =f s *t1; S2.3: For the sampling point sequence Perform a Fast Fourier Transform and logarithm operation to obtain the corresponding spectral data. Where FFT(·) represents Fast Fourier Transform: S2.4: Set the smoothing coefficient vector Where N fa If the number is even, the spectrum data will be... and Perform a multiplication-accumulation operation and record the result. Where: p is the ordinal number; S2.5: Settings The elements in the array correspond to their indices, where the index 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 indices greater than -BW valid / 2 and less than BW valid The element value of / 2 is recorded as SAM_F_fa1 i Let its length be N. s1 And record the selected index sequence as SAM_F_Ind; S2.7: Set SAM_F_fa1 i Record the i-th row of matrix C, where the size of matrix C is N*N. s1 ; S2.8: Let i = i + 1; if i is less than N, then go to step S2.2; if i is equal to N, then the process ends and output matrix C.

4. The two-dimensional amplitude characteristic correction method according to claim 3, characterized in that, Step S3: Spectrum correction value JZ_COE m The specific calculation steps are as follows: S3.1: Record various parameters of the ultra-wideband channel during normal operation, including: operating center frequency Fre work Working instantaneous bandwidth BW work Total working hours T work and AD sampling rate fs during operation work Let the total amount of sampled data to be compensated be... 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 will be SAMALL with N Thre Divide the segment into segments according to its length, and record the total number of segments as M; otherwise, divide it into N segments. Thre The value is updated to the total length of the sampled data to be compensated, and the total number of segments M is recorded as 1; each segment of data to be compensated is recorded as... Initialize ordinal m = 1; S3.3: Calculate the segmented data to be compensated, SAM_M m The frequency indices of each element in the table are given, where the index of the n1th element is: -fs work / 2+fs work / (N Thre -1)*n1 n1=0,1,2,…,N Thre -1 S3.4: Take the m-th segment of data to be compensated, SAM_M m Perform a Fast Fourier Transform on it to obtain its corresponding spectral data SAM_M_F m Where FFT(·) represents Fast Fourier Transform: SAM_M_F m =FFT(SAM_M m ) S3.5: Initialization Let n1 be a vector of all zeros, and let n1 = 0; S3.6: Take the center frequency Fre of the channel that needs compensation. work Simultaneously, take SAM_M m The index record corresponding to the n1th element is Fre. n1 ; S3.7: FRE1 1*N Subtract Fre from all elements work This yields 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, and record their positions as a1, a2, a3, and a4 respectively. 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 This yields 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, and record their positions as c1, c2, c3, and c4 respectively. 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: Combine the elements in S with C_TMP 4*4 Multiply each element correspondingly, sum the results of all multiplications, and fill the final result into JZ_COE. m Take the n1th element, where C_TMP 4*4 The definition is as follows: S3.13: Let n1 = n1 + 1; if n1 is less than N Thre If yes, proceed to step S3.6; otherwise, proceed to step S3.

14. S3.14: Let m = m + 1; if m is less than M + 1, then go to step S3.3; otherwise, the process ends, and the spectrum data SAM_M_F corresponding to each segment of data to be compensated is output. m and JZ_COE m And initialize m to 1.

5. The two-dimensional amplitude characteristic correction method 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 as follows: x＝1,2,3,4 y＝1,2,3,4 Both x and y are ordinal numbers.

6. The two-dimensional amplitude characteristic correction method according to claim 4, characterized in that, The specific calculation steps for spectrum correction in step S4 are as follows: S4.1: Select SAM_M_F m and JZ_COE m Using the following formula for SAM_M_F m Perform 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, then go to step 1; otherwise, this module ends, SAM_M_F m This refers to the spectral data of each segment after correction; if time-domain sampling data is required, then SAM_M_F... m This can be achieved by performing an inverse Fourier transform.

7. A two-dimensional amplitude characteristic correction device suitable for ultra-wideband channels, characterized in that, It includes a preprocessing module, a spectral characteristic calculation and smoothing module, a two-dimensional compensation coefficient calculation module, and a spectral correction module. In the preprocessing module, the center frequency points required for establishing the calibration matrix are calculated based on the channel's operating frequency band parameters and instantaneous operating bandwidth parameters, generating a center frequency point sequence. Considering that the effective bandwidth within the band is less than the channel's instantaneous bandwidth, the effective bandwidth also needs to be set. In the spectrum characteristic calculation and smoothing module, the channel is controlled to operate within the set center frequency point sequence. Data at each center frequency point is collected, spectrum characteristics are calculated, and smoothed until all center frequency points are processed, generating a compensation matrix. In the two-dimensional compensation coefficient calculation module, the values ​​of each compensation frequency point are calculated based on the center frequency point and bandwidth of the channel to be compensated. Then, interpolation coefficients are calculated based on each compensation frequency point value, and two-dimensional compensation coefficients are calculated 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.

8. A computer-readable storage medium having stored thereon computer program instructions, which, when loaded and run by a processor, cause the processor to perform the method as described in 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 program code for performing the method as described in any one of claims 1 to 6.

Images