Software defined radio-based noise suppression and signal reconstruction method

Abstract

The invention provides a noise suppression and signal reconstruction method based on software defined radio, and relates to the field of software defined radio and signal processing. The method comprises the steps that a preset two-channel SDR platform is used, a receiving signal containing target information is obtained, and the receiving signal comprises a step frequency continuous wave signal; performing smooth filtering processing on the received signal to obtain a filtered signal; reconstructing the filtering signal by using a least square method with a penalty term to obtain a target signal; and performing phase correction on the target signal by using a phase division and elimination method to obtain a correction result, and extracting target information from the correction result. According to the invention, noise interference can be handled, and the problem of signal phase incoherence is solved.

Description

Technical Field

[0001] The present invention relates to the field of software defined radio (SDR) and signal processing, and in particular to a noise suppression and signal reconstruction method based on SDR. Background Art

[0002] As an emerging radio broadcast communication technology, software defined radio is highly flexible and open. The core of this technology lies in the dynamic programmability of software, that is, software can be used to control a general hardware platform to achieve various wireless communication functions, breaking the deadlock that system functions depend on hardware design. Compared with traditional radio communication technology, SDR technology can achieve a larger bandwidth and has broad application prospects. For example, SDR technology can play a role in the field of radar detection. The advantages of low cost and flexible operation of the SDR platform conform to the current development trend of miniaturization, lightweight and digitalization of radar systems. It can be well integrated into various complex radar systems to achieve target detection with a smaller footprint.

[0003] Since the local oscillators of the transmit channel and the receive channel in the low-cost SDR platform are independent of each other, the reset of the local oscillator center frequency caused by the device restart will cause phase offset. In addition, the mixing time of the baseband signal and the transmit channel local oscillator is unpredictable, which will also lead to the incoherent characteristics of the system. This characteristic will affect its target detection performance in the field of radar technology, so incoherent compensation of the signal phase is required. At present, studies have used single-pole dual-control switches or local oscillator buffers to achieve phase compensation. Considering the efficiency of software development and the practicality of the equipment, the additional components increase the complexity and difficulty of operation of the system. In contrast, it is more efficient to solve the phase incoherence problem based on the signal processing level.

[0004] In addition to the phase offset problem, SDR technology has to deal with noise interference in complex environments. Strong noise may cause distortion of the echo signal. At present, there are various studies on signal noise suppression methods. Commonly used methods such as window function method and moving average filter method have the advantage of simple processing process, but these methods have problems such as signal distortion caused by smoothing signals and weak processing ability for sudden signal. The flexibility of SDR technology provides multiple possibilities for noise suppression methods. Other methods can be used in addition to the commonly used noise suppression methods to combine the advantages of each method. Summary of the invention

[0005] In view of the advantages of Stepped Frequency Continuous Waveform (SFCW) signal, such as easy implementation, high distance resolution and strong anti-interference ability, the present invention provides a noise suppression and signal reconstruction method based on software defined radio, which realizes the transmission and reception of SFCW signals and performs signal processing with the help of a dual-channel SDR platform.

[0006] The present invention provides a noise suppression and signal reconstruction method based on software defined radio, comprising: using a preset dual-channel SDR platform to obtain a received signal containing target information, wherein the received signal includes a stepped frequency continuous wave signal; performing smoothing filtering on the received signal to obtain a filtered signal; using a least square method with a penalty term to reconstruct the filtered signal to obtain a target signal; using a phase division and cancellation method to perform phase correction on the target signal to obtain a correction result, and extracting the target information from the correction result.

[0007] Furthermore, obtaining a received signal containing target information includes: generating a baseband signal by a dual-channel SDR platform; up-converting the baseband signal according to a stepped local oscillator frequency to obtain a transmitted signal; obtaining an echo signal after the transmitted signal is reflected by the target; and down-converting the echo signal according to the local oscillator frequency to obtain a received signal containing target information.

[0008] Furthermore, performing smoothing filtering on the received signal includes: filtering the received signal using a window function filtering method, wherein the window function includes a Hanning window.

[0009] Furthermore, the filtered signal is reconstructed using the least squares method with a penalty term to obtain the target signal, including: determining an expression form of the target signal, wherein the expression form includes target parameters to be solved; fitting the filtered signal using the least squares method to obtain a fitted signal; defining an objective function based on the difference between the filtered signal and the fitted signal and a penalty condition; wherein the penalty condition is determined based on a penalty term and a preset penalty coefficient, and the penalty term is defined based on the target parameter defined by the Tikhonov regularization method; and solving the target parameter by optimizing the objective function to obtain the reconstructed target signal.

[0010] Furthermore, determining the expression form of the target signal includes sorting the filtered signal according to the expression form of the baseband signal and removing the noise part in the filtered signal to obtain the target signal shown in the following formula:

[0011]

[0012] In the formula, is the target signal of channel i, i=1,2; f0 is the frequency of the baseband signal; t is the time; a1, b1, c1 is the target parameter to be solved in the target signal of channel 1, a2, b 2, c2 is the target parameter to be solved in the target signal of channel 2.

[0013] Furthermore, at each step frequency point of the stepped frequency continuous wave signal, the objective function is defined according to the following formula:

[0014]

[0015]

[0016] In the formula, , is the penalty coefficient and penalty term of channel i; For punishment conditions; represents the Euclidean norm; The nth sampling point of the filtered signal of channel i corresponding to each step frequency point; is the nth sampling point of the fitting signal of channel i corresponding to each step frequency point; N is the total number of sampling points of the filtered signal of channel i corresponding to each step frequency point.

[0017] Furthermore, the penalty term is defined according to the following formula:

[0018]

[0019]

[0020]

[0021] In the formula, is the training set feature matrix of channel i; is the training set data vector of channel i; M is the total number of data in the training set; is the mth data of the training set of the filtered signal of channel i, represents the transpose of the matrix, I is the identity matrix, Represents the inverse transformation of a matrix.

[0022] Furthermore, by optimizing the objective function to solve the target parameters, obtaining the reconstructed target signal also includes: splitting the filtered signal into a training set and a verification set according to a predetermined ratio; for the training set, minimizing the function value of the objective function by adjusting the penalty term to obtain the optimal solution of the target parameter; using the verification set to verify the optimal solution, and if the verification passes, determining the optimal solution as the solved target parameter; substituting the solved target parameter into the expression form of the target signal to obtain the reconstructed target signal.

[0023] Furthermore, using the phase division and cancellation method to perform phase correction on the target signal includes: for the target signal, selecting one of the dual channels as a measurement channel and the other as a reference channel; dividing the target signal of the measurement channel by the target signal of the reference channel to obtain a correction result.

[0024] Furthermore, extracting the target information from the correction result includes: performing pulse compression on the correction result to obtain a compressed signal; and extracting the target information from a peak value of the compressed signal.

[0025] Compared with the prior art, the noise suppression and signal reconstruction method based on software defined radio provided by the present invention has at least the following beneficial effects:

[0026] (1) Compared with the general mean method and least squares estimation method, this method has better robustness, can fit the data well, and effectively improves the prediction accuracy of the signal model.

[0027] (2) This method can correct the phase incoherence of the echo signal at the receiving end without adding any external accessories. This method reduces the cost of system construction, and the implementation of the compensation method is based on the signal processing level, which greatly reduces the workload and improves work efficiency, which is conducive to the use of SDR technology in fields such as radar detection.

[0028] (3) This method provides a basis for radar systems based on the SDR platform, ensures the accuracy of radar detection, broadens the application vision of SDR technology, and can be applied to penetrating radar systems such as wall-penetrating radar and ground-penetrating radar. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0030] Figure 1 A flowchart of a noise suppression and signal reconstruction method based on software defined radio according to an embodiment of the present invention is schematically shown;

[0031] Figure 2 Schematically shows a flow chart of obtaining a received signal according to an embodiment of the present invention;

[0032] Figure 3 Schematically shows a flow chart of reconstructing a filtered signal according to an embodiment of the present invention;

[0033] Figure 4 Schematically shows a flow chart for solving target parameters according to an embodiment of the present invention;

[0034] Figure 5A schematic diagram of a noise suppression and signal reconstruction method based on software defined radio according to an embodiment of the present invention is shown;

[0035] Figure 6 A flowchart of performing phase correction on a target signal according to an embodiment of the present invention is schematically shown;

[0036] Figure 7 The processing result diagram of the general mean value method is schematically shown;

[0037] Figure 8 The processing result diagram of the general least squares estimation method is schematically shown;

[0038] Fig. 9 A diagram schematically shows a processing result of a noise suppression and signal reconstruction method based on software defined radio according to an embodiment of the present invention;

[0039] Fig.10 The figure schematically shows a comparison of the processing results of the method according to the embodiment of the present invention and the general least squares estimation method. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings. 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.

[0041] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0042] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0043] In view of the advantages of stepped frequency continuous wave signal (SFCW), which is easy to realize, has high distance resolution and strong anti-interference ability, the present invention provides a noise suppression and signal reconstruction method based on software defined radio. A dual-channel SDR device is selected as the hardware platform and MATLAB is used as the software platform. The dual-channel SDR platform is used to realize the transmission and reception of SFCW signals, and MATLAB is used for signal processing.

[0044] Figure 1 The flowchart of the noise suppression and signal reconstruction method based on software defined radio according to an embodiment of the present invention is schematically shown.

[0045] like Figure 1 As shown, the noise suppression and signal reconstruction method based on software defined radio according to this embodiment may include steps S1 to S4.

[0046] Step S1, using a preset dual-channel SDR platform to obtain a received signal containing target information, wherein the received signal includes a stepped frequency continuous wave signal.

[0047] Step S2, performing smoothing filtering on the received signal to obtain a filtered signal.

[0048] Step S3, reconstructing the filtered signal using the least square method with a penalty term to obtain the target signal.

[0049] Step S4, using the phase division and cancellation method to perform phase correction on the target signal to obtain a correction result, and extract the target information from the correction result.

[0050] Through the embodiment of the present invention, the received signal is first smoothed and filtered to improve the signal quality. Then, the target signal is reconstructed using the least square method with a penalty term to cope with the interference of noise. The penalty term is used to avoid overfitting in the process of reconstructing the target signal. Finally, the phase division and cancellation method is used to perform phase correction to solve the problem of signal phase incoherence.

[0051] Figure 2 The flowchart of obtaining a received signal according to an embodiment of the present invention is schematically shown.

[0052] like Figure 2 As shown, in this embodiment, the above step S1 of obtaining a received signal containing target information may include steps S11 to S14.

[0053] Step S11, a baseband signal is generated by a dual-channel SDR platform.

[0054] The baseband signals generated by the two channels of the dual-channel SDR platform are of the same form, namely and , t is the time.

[0055] Step S12, up-converting the baseband signal according to the stepped local oscillator frequency to obtain a transmission signal.

[0056] The signal step frequency modulation is realized by transmitting the step change of the local oscillator frequency through the SDR platform. The transmission signal corresponding to the up-conversion process is and .

[0057] Step S13, the transmitted signal is reflected by the target to obtain an echo signal.

[0058] After being reflected by the target, the echo signal obtained by the SDR receiving end is and .

[0059] Step S14, down-converting the echo signal according to the local oscillator frequency to obtain a received signal containing target information.

[0060] The SDR receiving local oscillator frequency changes synchronously with the transmitting local oscillator frequency. After the down-conversion process, the received signal containing the target information is obtained as follows: and .

[0061] Specifically, the two transmit channels of the dual-channel SDR platform share a transmit local oscillator, and the generated local oscillator signal is ; Two receiving channels share a receiving local oscillator, and the generated local oscillator signal is The above signals are expressed as:

[0062]

[0063]

[0064]

[0065] In the formula, is the frequency of the baseband signal, is the local oscillator frequency, Indicates the relative phase deviation between the transmitted local oscillator and the baseband signal. It represents the relative phase deviation between the received local oscillator and the baseband signal. When i is 1, it represents channel 1. When i is 2, it represents channel 2. exp(·) represents the natural exponential function. j is the imaginary unit.

[0066] The baseband signal generated by the dual-channel SDR platform completes the up-conversion process, and the final transmission signal is expressed as:

[0067]

[0068] In the formula, is the transmission signal of channel i, is the amplitude introduced by the up-conversion process of channel i.

[0069] The transmitted signal is reflected by the target and a certain delay is introduced to obtain the echo signal of channel i. It is expressed as:

[0070]

[0071] In the formula, is the target amplitude of channel i, The delay introduced by the target detected by channel i.

[0072] After receiving the echo signal, the SDR receiver performs down-conversion processing to obtain the received signal of channel i (that is, the signal after the echo signal and the receiving local oscillator are mixed), which are expressed as and :

[0073]

[0074] In the formula, is the amplitude introduced by the down-conversion process of channel i, It is the phase offset term between the transmitting channel and the receiving channel. Each local oscillator step change will cause the offset term to change, resulting in phase incoherence.

[0075] In this embodiment, the step S2 of performing smoothing filtering on the received signal includes: filtering the received signal using a window function filtering method, wherein the window function includes a Hanning window.

[0076] The received signal is filtered using the window function and Filter and obtain the filtered signal and , in order to solve the signal top distortion problem caused by SDR equipment, thereby improving signal quality and facilitating subsequent reconstruction.

[0077] Specifically, the top distortion problem of the SFCW echo signal caused by the instability of the SDR device can be initially processed by the window function filtering method, and the Hanning window is usually used to complete the filtering process. The main lobe width of the Hanning window is smaller, which reduces the sidelobe leakage compared to the rectangular window. It can not only smooth the signal but also ensure the accuracy of the frequency domain analysis. The window function h(n) of the Hanning window and the processed filtered signal are expressed as:

[0078]

[0079]

[0080]

[0081] In the formula, is the received signal of channel i The discrete Fourier transform of for The filtered signal of channel i after smoothing filtering The discrete Fourier transform of is the spectrum function of the Hanning window, N is the total number of sampling points of the channel data corresponding to each step frequency point, is the frequency domain variable, Will Discretization is performed to realize discrete Fourier transform, and cos(·) represents the cosine function.

[0082] Considering that the SFCW signal changes in steps and has multiple step frequency points, each step frequency point has multiple sampling points, and N is the total number of sampling points.

[0083] Next, the least square method with a penalty term is used to reconstruct the target signal. This embodiment introduces a penalty term on the basis of the general least square method fitting process to reconstruct the target signal. and Reconstruct the amplitude and phase parameters respectively, that is, fit the target signal containing useful information under the influence of noise and .

[0084] The process of extracting the target signal from noise interference can be understood as a curve fitting problem, that is, finding the most suitable parameters under the premise of noise influence to reconstruct the useful signal containing the target information.

[0085] Figure 3 The flowchart for reconstructing a filtered signal according to an embodiment of the present invention is schematically shown.

[0086] like Figure 3 As shown, in this embodiment, the above step S3 may further include steps S31 to S34.

[0087] Step S31, determining the expression form of the target signal, wherein the expression form includes the target parameters to be solved.

[0088] For example, for SFCW signals, the transmitted baseband signal frequency As a known condition, it can be used for signal reconstruction. and The expression form is then used to extract the real part of the signal, the filtered signals of channel 1 and channel 2 and Can be expressed as:

[0089]

[0090] Where a1, b 1, c1 is the target parameter to be solved in the target signal of channel 1, a2, b 2, c2 is the target parameter to be solved in the target signal of channel 2, is the noise part that we want to remove in the filtered signal of channel i, and sin(·) represents the sine function.

[0091] The form of the reconstructed target signal is determined by the form of the dual-channel filtered signal. After removing the noise part, the target signals of channel 1 and channel 2 are and It can be expressed as:

[0092]

[0093] Step S32, using the least square method to fit the filtered signal to obtain a fitting signal.

[0094] Step S33, defining an objective function according to the difference between the filtered signal and the fitted signal and a penalty condition; wherein the penalty condition is determined based on a penalty term and a preset penalty coefficient, and the penalty term is defined based on the Tikhonov space regularization method for the target parameter.

[0095] In this embodiment, at each step frequency point of the step frequency continuous wave signal, the objective function is defined according to the following formula:

[0096]

[0097]

[0098] In the formula, , is the penalty coefficient and penalty term of channel i, i=1,2; For punishment conditions; represents the Euclidean norm, which is calculated by taking the square root of the sum of the squares of all elements; The nth sampling point of the filtered signal of channel i corresponding to each step frequency point; is the nth sampling point of the fitting signal of channel i corresponding to each step frequency point; N is the total number of sampling points of the filtered signal of channel i corresponding to each step frequency point.

[0099] According to the filtered signal and the determined target signal form, this embodiment introduces a penalty term based on the general least squares method. , and choose the penalty coefficient , in order to control the penalty condition and prevent overfitting, that is, to minimize the function value of the objective function.

[0100] In addition, the target signal parameters to be estimated are only amplitude and phase, and the model complexity requirement is relatively low. The Tikhonov regularization method is selected to define the penalty term. It can be seen that The elements in contain the target parameters to be solved in the target signal of channel i.

[0101] It should be understood that Tikhonov regularization, also known as ridge regression or L2 regularization, is a regularization method for linear regression. It controls the complexity and generalization ability of the model by adding an L2 norm penalty term to the objective function of the model. This penalty term penalizes the sum of squares of the weights of the model to avoid excessive weights, thereby alleviating the overfitting problem. This method introduces a regularization term in the loss function and balances the fitting ability and generalization ability of the model by adjusting the regularization coefficient. Tikhonov regularization has a wide range of applications in practical applications and can effectively improve the performance and stability of the model.

[0102] Step S34, solving the target parameters by optimizing the target function to obtain the reconstructed target signal.

[0103] Figure 4 A flowchart for solving target parameters according to an embodiment of the present invention is schematically shown. Figure 5 The principle diagram of the noise suppression and signal reconstruction method based on software defined radio according to an embodiment of the present invention is schematically shown.

[0104] Please combine Figure 4 and Figure 5 In this embodiment, the above step S34 may further include step S341 to step S344.

[0105] Step S341, splitting the filtered signal into a training set and a validation set according to a predetermined ratio.

[0106] According to a predetermined ratio, a part of the original data of the filtered signal is randomly extracted as a training set, and the other part is used as a validation set to calculate the objective function.

[0107] Step S342, for the training set, by adjusting the penalty term, to minimize the function value of the objective function and obtain the optimal solution of the objective parameter.

[0108] For the training set, a linear regression solution based on normal equations is used to determine the target parameters to be solved so as to minimize the function value of the objective function.

[0109] Specifically, the penalty term containing the objective parameter to be solved is defined as follows:

[0110]

[0111]

[0112]

[0113] In the formula, is the training set feature matrix of channel i; is the training set data vector of channel i; M is the total number of data in the training set; is the mth data of the training set of the filtered signal of channel i, represents the transpose of the matrix, I is the identity matrix, Represents the inverse transformation of a matrix.

[0114] Step S343, using the verification set to verify the optimal solution, if the verification passes, the optimal solution is determined as the solved target parameter.

[0115] After the optimal solution of the target parameter is obtained using the training set, the optimal solution is verified using the validation set. The verification process is the same as the process of optimizing the target parameter of the training set. For example, the verification can be performed according to the following formula:

[0116]

[0117] In the formula, is the function value of the objective function of channel i; is the total number of data in the validation set of channel i; The lth sampling point of the filtered signal of channel i corresponding to each step frequency point; Each step frequency point corresponds to the lth sampling point of the best fitting signal of channel i obtained by applying the above optimal solution.

[0118] Then, the function value of the objective function of channel i is determined When it is the minimum value, it can be considered that the verification has passed, and the optimal solution can be determined as the solved target parameter.

[0119] Step S344: Substitute the solved target parameters into the expression form of the target signal to obtain the reconstructed target signal.

[0120] With the help of Euler's formula, the reconstructed target signal It can be expressed as:

[0121]

[0122] In the formula, is the amplitude of the target signal of channel i; is the phase of the target signal of channel i, containing the target information of channel i.

[0123] Figure 6 The flowchart of performing phase correction on a target signal according to an embodiment of the present invention is schematically shown.

[0124] like Figure 6 As shown, in this embodiment, the above step S4 uses the phase division and cancellation method to perform phase correction on the target signal, and can further include steps S41 and S42.

[0125] Step S41, for the target signal, one of the dual channels is selected as the measurement channel and the other is selected as the reference channel.

[0126] Step S42, dividing the target signal of the measurement channel by the target signal of the reference channel to obtain a correction result.

[0127] The dual-channel SDR platform adopts dual-transmit and dual-receive diversity technology. The transmit channel shares a transmit local oscillator and the receive channel shares a receive local oscillator. Combined with the characteristics of the SFCW signal, channel 1 can be selected as the measurement channel and channel 2 as the reference channel. The phase division and cancellation method is used to divide the target signal of the measurement channel by the target signal of the reference channel to perform phase correction on the target signal to eliminate the phase offset term. , the correction result is It can be expressed as:

[0128]

[0129] In this embodiment, the above step S4 of extracting the target information from the correction result includes: performing pulse compression on the correction result to obtain a compressed signal; and extracting the target information from the peak value of the compressed signal.

[0130] For example, the calibration results The average value is taken by channel and then Fourier transform is performed to complete pulse compression, and finally the target information is extracted based on the peak value.

[0131] Through the embodiments of the present invention, the noise of the signal received by the SDR device can be effectively suppressed, providing a guarantee for accurate detection of the SDR-based radar system.

[0132] The following is an experimental verification of the noise suppression and signal reconstruction method based on software defined radio proposed in the present invention.

[0133] In this experiment, a dual-channel SDR platform was used to perform closed-loop testing of an external attenuator. Channel 1 was selected as the measurement channel, and the length of the external cable was 1.6m. The dual channels simultaneously collected 49 sets of data, each of which contained 128 time domain data corresponding to the received signal of the local oscillator step change. Each of the 128 channels of data was used as the time domain data of the channel corresponding to a step frequency point.

[0134] Figure 7 The following diagram schematically shows the processing result of the general mean value method. Figure 7 It can be seen that after the data is phase corrected, the general mean method is used to take the mean value by channel, and then the result of inverse Fourier transform is affected by noise. It should be understood that the result of taking the mean value by channel by the general mean method is frequency domain data with frequency as the variable, while the result of inverse Fourier transform is time domain data.

[0135] Figure 8 The figure schematically shows the processing result of the general least squares estimation method. Figure 8 It can be seen that the signal reconstruction based on the general least squares method is first performed on the time domain data, and then the phase correction is performed. It can be preliminarily determined that the signal reconstruction method based on the least squares method is effective. Compared with the mean method, the side lobe is reduced by about 6.9dB, and the cable length can be well extracted.

[0136] Fig. 9 The figure schematically shows the processing result of the noise suppression and signal reconstruction method based on software defined radio according to an embodiment of the present invention. Fig.10 The figure schematically shows a comparison of the processing results of the method according to the embodiment of the present invention and the general least squares estimation method.

[0137] Fig. 9 Take the average of 49 groups of data, and Fig.10 It is calculated that, using the method proposed in the present invention, the side lobe is reduced by about 4.1 dB compared with the general least square estimation method.

[0138] In addition, under the premise of using the method proposed in the present invention, taking the mean square error of channel 1 as an example, the penalty term is defined based on the Tikhonov regularization method, and the mean square error obtained by taking the average according to the number of steps is about 23.5 lower than the result of the Lasso (Least Absolute Shrinkage and Selection Operator) regularization method. It can be proved that selecting the penalty term based on the Tikhonov regularization method is more suitable for this embodiment.

[0139] From the above description, it can be seen that the above embodiments of the present invention achieve at least the following technical effects:

[0140] (1) Under the influence of noise, the target signal can be extracted through the penalized least squares estimation signal reconstruction method based on the window function method. The sidelobe suppression effect and anti-interference performance are better than the general mean method and least squares estimation method, and it can adapt to complex application scenarios such as wall-penetrating radar or ground-penetrating radar;

[0141] (2) Without adding any external accessories, the echo phase incoherence problem of the dual-channel SDR platform is solved at the signal processing level to improve the target distance positioning accuracy.

[0142] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in the order marked in different drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0143] In the description of the present invention, "plurality" means at least two, for example two, three, etc., unless otherwise clearly and specifically limited. In addition, the word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements.

[0144] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A noise suppression and signal reconstruction method based on software defined radio, characterized in that: include: Using a preset dual-channel SDR platform, obtaining a received signal containing target information, wherein the received signal includes a stepped frequency continuous wave signal; Performing smoothing filtering on the received signal to obtain a filtered signal; Reconstructing the filtered signal using a least square method with a penalty term to obtain a target signal; The target signal is phase corrected by using a phase division and cancellation method to obtain a correction result, and the target information is extracted from the correction result.

2. The noise suppression and signal reconstruction method based on software defined radio according to claim 1, characterized in that: The obtaining of a received signal containing target information comprises: Generating a baseband signal by the dual-channel SDR platform; Performing up-conversion processing on the baseband signal according to the stepped local oscillator frequency to obtain a transmission signal; After the transmitted signal is reflected by the target, an echo signal is obtained; The echo signal is down-converted according to the local oscillator frequency to obtain the received signal containing the target information.

3. The noise suppression and signal reconstruction method based on software defined radio according to claim 1, characterized in that: The step of performing smoothing filtering on the received signal comprises: The received signal is filtered using a window function filtering method, wherein the window function includes a Hanning window.

4. The noise suppression and signal reconstruction method based on software defined radio according to claim 1, characterized in that: The step of reconstructing the filtered signal using the least square method with a penalty term to obtain a target signal comprises: Determining an expression form of the target signal, wherein the expression form includes a target parameter to be solved; Fitting the filtered signal using the least squares method to obtain a fitted signal; Defining an objective function according to the difference between the filtered signal and the fitted signal and a penalty condition; wherein the penalty condition is determined based on a penalty term and a preset penalty coefficient, and the penalty term is defined based on the Tikhonov regularization method for defining the target parameter; The target parameters are solved by optimizing the target function to obtain a reconstructed target signal.

5. The noise suppression and signal reconstruction method based on software defined radio according to claim 4, characterized in that: Determining the expression form of the target signal includes: The filtered signal is sorted according to the expression form of the baseband signal, and the noise part in the filtered signal is removed to obtain the target signal shown in the following formula: In the formula, is the target signal of channel i, i=1,2; f0 is the frequency of the baseband signal; t is the time; a1, b 1, c1 is the target parameter to be solved in the target signal of channel 1, a2, b 2, c2 is the target parameter to be solved in the target signal of channel 2.

6. The noise suppression and signal reconstruction method based on software defined radio according to claim 5, characterized in that: At each step frequency point of the stepped frequency continuous wave signal, the objective function is defined according to the following formula: In the formula, , is the penalty coefficient and penalty term of channel i; For punishment conditions; represents the Euclidean norm; The nth sampling point of the filtered signal of channel i corresponding to each step frequency point; is the nth sampling point of the fitting signal of channel i corresponding to each step frequency point; N is the total number of sampling points of the filtered signal of channel i corresponding to each step frequency point.

7. The noise suppression and signal reconstruction method based on software defined radio according to claim 6, characterized in that: The penalty term is defined according to the following formula: In the formula, is the training set feature matrix of channel i; is the training set data vector of channel i; M is the total number of data in the training set; is the mth data of the training set of the filtered signal of channel i, represents the transpose of the matrix, I is the identity matrix, Represents the inverse transformation of a matrix.

8. The noise suppression and signal reconstruction method based on software defined radio according to claim 4, characterized in that: The step of solving the target parameter by optimizing the target function to obtain the reconstructed target signal further comprises: Splitting the filtered signal into a training set and a validation set according to a predetermined ratio; For the training set, the function value of the objective function is minimized by adjusting the penalty term to obtain the optimal solution of the objective parameter; Using the verification set to verify the optimal solution, if the verification passes, determining the optimal solution as the solved target parameter; Substitute the solved target parameters into the expression form of the target signal to obtain the reconstructed target signal.

9. The noise suppression and signal reconstruction method based on software defined radio according to claim 1, characterized in that: The phase correction of the target signal using the phase division and cancellation method comprises: For the target signal, one of the dual channels is selected as a measurement channel and the other is selected as a reference channel; The target signal of the measurement channel is divided by the target signal of the reference channel to obtain the correction result.

10. The noise suppression and signal reconstruction method based on software defined radio according to claim 1, characterized in that: Extracting the target information from the correction result comprises: Performing pulse compression on the correction result to obtain a compressed signal; The target information is extracted from a peak value of the compressed signal.

Images