CN116405129A - Self-interference cancellation method, system, device and medium based on improved LMS algorithm - Google Patents

Abstract

The present invention relates to the technical field of digital signal processing, and discloses a self-interference cancellation method, system, equipment and medium based on an improved LMS algorithm, wherein the method includes: according to the transmission signal and filter transmitted by the transmission end of the full-duplex ISAC system Calculate the output signal of the filter; carry out error calculation according to the baseband receiving signal and the output signal in the full-duplex ISAC system, and obtain the error signal; according to the error signal and the improved skip line function , construct a step adjustment function, and iteratively update the weight coefficient according to the step adjustment function; iteratively update the error signal according to the updated weight coefficient, when the power of the updated error signal reaches a minimum value, The updated error signal and its corresponding output signal are output. The improved algorithm of the invention enables the self-interference cancellation process in the digital domain of the full-duplex ISAC system to maintain relatively fast convergence while maintaining low computational complexity.

Description

Self-interference cancellation method, system, device and medium based on improved LMS algorithm

technical field

The present invention relates to the technical field of digital signal processing, in particular to a self-interference cancellation method, system, device and medium based on an improved LMS algorithm.

Background technique

At present, the least mean square algorithm——LMS algorithm is widely used in digital domain self-interference cancellation of full-duplex ISAC systems due to its simple implementation. The algorithm can obtain a faster convergence speed, but the steady-state error is larger; when the step size factor is smaller, the algorithm can obtain a smaller steady-state error, but the convergence speed is slower. The inherent defect of the fixed step value cannot reasonably coordinate the convergence speed and steady-state error, and various variable-step-size LMS algorithms have emerged one after another: such as the variable-step-size LMS algorithm proposed by Qin Jingfan et al. based on the Sigmoid function——SVSLMS algorithm. Establish a new nonlinear function relationship between the LMS algorithm and the step size, which overcomes the contradiction between the convergence speed of the LMS algorithm and the steady-state error, but when the error is close to zero in the SVSLMS algorithm, the variable step size function changes too much and does not have a slow change The characteristics of the SVSLMS algorithm still have a large step size change in the adaptive steady-state stage; another example is the TCLMS algorithm based on the variable step length LMS algorithm proposed by using the characteristics of the skip line function, which calculates The complexity is low, and under the premise that the steady-state error is the same as the aforementioned SVSLMS algorithm, it has faster convergence speed and tracking speed, and solves the contradiction between convergence speed and steady-state error to a greater extent. However, due to the current application requirements, The convergence speed of the TCLMS algorithm needs to be further improved, and it is easily affected by the interference of uncorrelated noise at the signal input end, which affects the stability of the algorithm. Therefore, there is an urgent need for an algorithm that can maintain fast convergence and low computational complexity in the process of self-interference cancellation in the digital domain of a full-duplex ISAC system.

Contents of the invention

The present invention provides a self-interference cancellation method, system, equipment and medium based on an improved LMS algorithm. By improving the algorithm, the self-interference cancellation process in the digital domain of a full-duplex ISAC system can be performed even in a low signal-to-noise ratio state. It still maintains good performance and maintains low computational complexity while maintaining fast convergence.

In order to achieve the above object, the first aspect of the present invention provides a self-interference cancellation method based on the improved LMS algorithm, comprising the following steps:

calculating the output signal of the filter according to the transmission signal sent by the transmitting end of the full-duplex ISAC system and the weight coefficient of the filter;

Performing error calculation according to the baseband receiving signal and the output signal in the full-duplex ISAC system to obtain an error signal;

Construct a step size adjustment function according to the error signal and the improved skip line function, and iteratively update the weight coefficient according to the step size adjustment function;

The error signal is iteratively updated according to the updated weight coefficient, and when the power of the updated error signal reaches a minimum value, the updated error signal and its corresponding output signal are output.

Further, before calculating the output signal of the filter according to the transmission signal sent by the transmitting end of the full-duplex ISAC system and the weight coefficient of the filter, the method includes:

The filter is initialized so that the weight coefficient is 0.

Further, the step size adjustment function is constructed according to the error signal and the improved skip line function, including:

Substituting the error function as an independent variable into the improved skip line function, according to the principle of step size adjustment, adding an absolute value to the error signal and introducing adjustment parameters to obtain the step size adjustment function.

Further, the expression of the step size adjustment function is:

In the formula, μ(n) is the step adjustment function, α is the first adjustment parameter, β is the second adjustment parameter, m is the third adjustment parameter, and α, β, m are all greater than 0, and e(n) is the error function.

Further, the baseband received signal is a signal after antenna domain self-interference cancellation, analog domain self-interference cancellation and ADC quantization in the full-duplex ISAC system, and the expression of the baseband received signal is:

r(n)=s _I (n)+d(n)+ε(n)

In the formula, r(n) is the baseband received signal, s _I (n) is the residual self-interference signal after antenna domain self-interference cancellation and analog domain self-interference cancellation, d(n) is the desired signal at the far end, ε(n) is additive noise.

Further, when the interference cancellation ratio reaches the maximum value, the iterative update process is ended, and the updated error signal and its corresponding output signal are output; wherein, the calculation formula of the interference cancellation ratio is as follows:

In the formula, ICR is the interference cancellation ratio, P _r is the power of baseband received signal, P _d is the power of remote desired signal, σ _ε ² is the power of additive noise signal, E{|e(∞)| ² } is The mean square error as n tends to infinity.

The second aspect of the present invention provides a self-interference cancellation system based on the improved LMS algorithm, including:

The output signal acquisition module is used to calculate the output signal of the filter according to the transmission signal sent by the full-duplex ISAC system transmission end and the weight coefficient of the filter;

An error signal acquisition module, configured to perform error calculation according to the baseband receiving signal and the output signal in the full-duplex ISAC system, to obtain an error signal;

A weight coefficient updating module, configured to construct a step adjustment function according to the error signal and the improved skip line function, and iteratively update the weight coefficient according to the step adjustment function;

The loop cutoff output module is configured to iteratively update the error signal according to the updated weight coefficient, and output the updated error signal and its corresponding output signal when the power of the updated error signal reaches a minimum value.

Further, the weight coefficient update module package:

The step size function construction module is used for substituting the error function into the improved skip line function as an independent variable, according to the step size adjustment principle, adding an absolute value to the error signal and introducing a step size adjustment parameter to obtain the step size Long adjustment function.

A third aspect of the present invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the above-mentioned The self-interference cancellation method based on the improved LMS algorithm described in any one of the first aspect.

The fourth aspect of the present invention provides a computer-readable storage medium, the computer-readable storage medium includes a stored computer program, wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned first The self-interference cancellation method based on the improved LMS algorithm described in any one of the aspects.

Compared with the prior art, the beneficial effects of the embodiments of the present invention are:

The present invention provides a self-interference cancellation method, system, device and medium based on the improved LMS algorithm, and constructs a step size adjustment function by establishing a new nonlinear relationship between the improved skip line function and the error signal, and then updates the weight coefficient The step size factor can dynamically change the size of the step size factor during the convergence process of the algorithm, so that the improved LMS algorithm can maintain a fast convergence while maintaining a low computational complexity, and it can be used under the condition of low signal-to-noise ratio , can still maintain good performance, and in the full-duplex synaesthesia integrated digital domain self-interference cancellation, the ICR in the digital domain is higher than other compared algorithms and has a faster convergence speed.

Description of drawings

In order to illustrate the technical solution of the present invention more clearly, the accompanying drawings used in the implementation will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some implementations of the present invention. As far as the skilled person is concerned, other drawings can also be obtained based on these drawings on the premise of not paying creative work.

Fig. 1 is a full-duplex ISAC system model diagram provided by a certain embodiment of the present invention;

Fig. 2 is a self-interference cancellation model diagram based on LMS adaptive filtering provided by an embodiment of the present invention;

Fig. 3 is a flowchart of a self-interference cancellation method based on an improved LMS algorithm provided by an embodiment of the present invention;

Fig. 4 is the performance simulation result diagram of the step size update change of the IVSSLMS, SVSLMS and TCLMS algorithms provided by a certain embodiment of the present invention;

Fig. 5 is the mean square error performance simulation result diagram under different variable step size LMS algorithms provided by a certain embodiment of the present invention;

Fig. 6 is when the signal-to-noise ratio provided by a certain embodiment of the present invention is 5, the mean square error performance simulation result diagram under different variable step size LMS algorithms;

Fig. 7 is the mean square error performance simulation result diagram under different variable step size LMS algorithms when the signal-to-noise ratio provided by a certain embodiment of the present invention is 1;

Fig. 8 is a comparison diagram of ICR under different variable step size LMS algorithms provided by a certain embodiment of the present invention;

Fig. 9 is a function model curve diagram of the step factor and the error when the parameters β and m provided by a certain embodiment of the present invention are constant, and α is 0.5, 1, 2, and 5 respectively;

Fig. 10 is a function model curve diagram of the step factor and the error when the parameters α and β provided by a certain embodiment of the present invention are constant, and m is 0.5, 1, 2, and 5 respectively;

Fig. 11 is a function model curve diagram of the step factor and the error when the parameters α and m provided by a certain embodiment of the present invention are constant, and β is 0.05, 0.1, 0.2, and 0.5 respectively;

Fig. 12 is a device diagram of a self-interference cancellation system based on an improved LMS algorithm;

Fig. 13 is a structural diagram of an electronic device provided by an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

It should be understood that the step numbers used herein are only for convenience of description, and are not intended to limit the execution order of the steps.

It should be understood that the terminology used in the description of the present invention is for the purpose of describing particular embodiments only and is not intended to limit the present invention. As used in this specification and the appended claims, the singular forms "a", "an" and "the" are intended to include plural referents unless the context clearly dictates otherwise.

The terms "comprising" and "comprising" indicate the presence of described features, integers, steps, operations, elements and/or components, but do not exclude the presence of one or more other features, integers, steps, operations, elements, components and/or The presence or addition of its collection.

The term "and/or" refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

Explanation of some terms involved in the present invention:

Synaesthesia integration: refers to the fusion of communication and perception functions, so that the communication system has both communication and perception functions. While transmitting information through wireless channels, it actively recognizes and analyzes the characteristics of the channel to perceive the surroundings. The physical characteristics of the environment, so that communication and perception functions mutually enhance each other.

Simultaneous full-duplex at the same frequency: Simultaneous full-duplex at the same frequency is considered to be a technology that can effectively improve spectrum efficiency. This technology realizes the transmission of signals in two directions on the same physical channel, that is, through The receiver eliminates the interference of its own transmitter signal, and simultaneously receives the same frequency signal from another node at the transmitter signal. Compared with the traditional time division duplex (TDD) and frequency division duplex (FDD), simultaneous full duplex on the same frequency can double the spectrum efficiency.

LMS algorithm: The least mean square algorithm, referred to as the LMS algorithm, is an improved algorithm of the steepest descent algorithm. It is an optimized extension after using the rapid descent method in the Wiener filter theory. It was first proposed by Widrow and Hoff. The algorithm does not need to know the statistical characteristics of the input signal and the expected signal, and the weight coefficient of the "current moment" is obtained by adding a proportional term of the negative mean square error gradient to the weight coefficient of the "previous moment". It has the characteristics of low computational complexity, good convergence in the environment where the signal is a stationary signal, its expected value unbiasedly converges to the Wiener solution, and the stability of the algorithm when the algorithm is implemented with finite precision, making the LMS algorithm an adaptive algorithm. The most stable and widely used algorithm.

Interference cancellation ratio: The interference cancellation ratio (Interference Cancellation Ratio, ICR) after self-interference cancellation is an important indicator for judging the interference cancellation ability. The larger the value of ICR, the better the self-interference suppression ability.

The full-duplex ISAC system model is shown in Figure 1. Due to the characteristics of the full-duplex ISAC system, the receiver will be affected by both the self-interference signal and the perception signal while receiving the signal. The remote receiver The same is true. In the present invention, the full-duplex ISAC system only considers the influence of the remote full-duplex communication sensing node, and does not consider the influence of the sensing target. The main challenge of the full-duplex ISAC system is that the power of the self-interfering signal in the full-duplex ISAC system is much higher than the power of the signal of interest, so it will saturate or even damage the receiver, which will make the receiver unable to receive the desired signal. It is self-interference cancellation.

The self-interference cancellation of full-duplex ISAC system is divided into three steps: antenna domain, analog domain and digital domain. When the self-interference signal is received by the receiving antenna at the receiving end, a part of the self-interference signal can be canceled by the antenna self-interference cancellation technology; after the self-interference cancellation in the antenna domain is completed, the remaining self-interference signal is self-interferenced in the analog domain before being quantized by the ADC. Interference cancellation; the signal after ADC quantization includes the remote desired signal, some self-interference signals and noise signals, and these signals are finally subjected to digital domain self-interference cancellation before baseband demodulation; so far, the self-interference of the entire full-duplex ISAC system has been completed Offset process. The self-interference cancellation method based on the improved LMS algorithm proposed by the present invention is the digital domain self-interference cancellation of the full-duplex ISAC system, which eliminates the residual self-interference signal in the baseband received signal.

For the self-interference cancellation in the digital domain of the full-duplex ISAC system model, the LMS adaptive filtering method is usually used, and the self-interference cancellation model based on the LMS adaptive filtering is shown in Figure 2. The mathematical description of the model is as follows:

y(n)=s ^H (n)w(n)

e(n)=r(n)-y(n)

w(n+1)=w(n)+2μe(n)s(n)

In the formula, μ is the step size factor, which is a constant; the superscript H represents the conjugate transposition of the matrix; w(n) is the weight coefficient of the M-order filter, n is the n moment; s(n) is the input signal; y(n) is the output signal of the filter; r(n) is the baseband receiving signal; e(n) is the error signal, which is used to feed back to the filter, so as to continuously adjust the weight coefficient of the filter. offset model. However, in the traditional fixed step size algorithm, when the step size factor is selected to a larger value, the algorithm can obtain a faster convergence speed, but the steady-state error is larger; when the step size factor is selected to a smaller value, the algorithm can obtain a smaller steady-state error. state error, but the convergence speed is slow. The inherent defect of the fixed step size value cannot reasonably coordinate the convergence speed and the steady-state error. Aiming at this problem, various variable step size LMS algorithms have been proposed one after another, the core of which is to dynamically change the size of the step size factor in the algorithm convergence process. The present invention also A solution to this problem is provided, and the step size factor of the weight coefficient is updated by establishing a new nonlinear relationship between the improved skip line function and the error signal.

In one embodiment, as shown in FIG. 3 , the first aspect of the present invention provides a self-interference cancellation method based on an improved LMS algorithm, including the following steps:

S1. Calculate the output signal of the filter according to the transmission signal sent by the transmitting end of the full-duplex ISAC system and the weight coefficient of the filter;

Specifically, the acquisition method of the filter output signal is the same as the acquisition method of the output signal in the conventional LMS adaptive filtering method, which is to perform conjugate transposition on the transmission signal transmitted by the transmission end of the full-duplex ISAC system, and then combine it with the current It is obtained by multiplying the weight coefficients of the M-order filter at the moment. However, in a specific embodiment, before step S1, the filter needs to be initialized so that the weight coefficient is 0, and an initial value is assigned to the input signal and the error signal, so that the filter can converge quickly and work better.

S2. Perform error calculation according to the baseband receiving signal and output signal in the full-duplex ISAC system to obtain an error signal;

Specifically, the error signal is the difference between the baseband received signal and the output signal of the filter in the full-duplex ISAC system. The signal after self-interference cancellation and ADC quantization, the expression of the baseband received signal is:

r(n)=s _I (n)+d(n)+ε(n)

where s _I (n) is the residual self-interference signal after antenna domain self-interference cancellation and analog domain self-interference cancellation, d(n) is the desired signal at the far end, and ε(n) is additive noise.

According to Figure 1 and Figure 2, it can be seen that the baseband received signal is r(n) including the remote desired signal, part of the self-interference signal and noise signal after the ADC quantization is completed, and the signal is finally self-assessed in the digital domain before baseband demodulation Interference cancellation. The error signal is used as the feedback signal for the LMS algorithm to update the weight coefficient vector. In the self-interference cancellation model, after the adaptive algorithm converges, the signal can be expressed as the remaining signal in the digital domain after self-interference cancellation, that is, the remote desired signal; And according to the improved LMS algorithm (also known as IVSSLMS algorithm) in the present invention, the weight coefficient in the filter is updated, so that the error signal of the final output is as close as possible to the far-end desired signal, thereby achieving the effect of adaptive interference cancellation Purpose.

S3. Construct a step size adjustment function according to the error signal and the improved skip line function, and iteratively update the weight coefficient according to the step size adjustment function;

In a specific embodiment, step S3 specifically includes: substituting the error function as an argument into the improved skip line function, adding an absolute value to the error signal and introducing adjustment parameters according to the step size adjustment principle to obtain a step size adjustment function. The IVSSLMS algorithm proposed by the present invention establishes a new nonlinear relationship between the improved skip line function and the error signal to construct a step size adjustment function, and then update the step size factor and weight coefficient to establish a balance between complexity and convergence .

Specifically, the step size adjustment function of the present application starts from the most basic skip line function and improves it, and the expression of the improved skip line function is as follows:

Substituting the error function as an independent variable into the improved skip line function, introducing the first adjustment parameter to the error function substituted into the improved skip line function, and then obtaining the first variable step size function, the formula expression is as follows:

In the formula, μ ₁ (n) is the first variable step function, and α is the first adjustment parameter.

The first variable step size function formed by introducing the error function and the first adjustment parameter into the improved skip line function improves the convergence speed and tracking speed on the basis of the LMS algorithm.

Introduce the curve parameters in the first variable step function to realize the controllable function curve, and obtain the controllable variable step function, the formula expression is as follows:

In the formula, μ _k (n) is a controllable variable step function, and k is a curve parameter.

Through the analysis of the controllable variable step size function, it can be known that the smaller the value of k, the greater the maximum convergence value of the curve, but the faster the curve descends at this time; the larger the k value, the slower the curve descends, but the maximum convergence of the curve The smaller the value. Therefore, for the comprehensive analysis of the convergence speed of the comprehensive curve and the complexity of the steady-state error, the value of k is preferably 2.

Introduce the second adjustment parameter of the value range of the control function to the controllable variable step size function that determines the value of k, and perform a square operation on the denominator including the error function to obtain the second variable step size function, and its formula expression is as follows:

In the formula, μ ₂ (n) is the second variable step function, and β is the second adjustment parameter, which is also a coefficient controlling the value range of the function. The second variable step function introducing the second adjustment parameter establishes a balance between the curve convergence speed and the steady-state error, and further reduces the complexity of the algorithm.

In order to meet the step size adjustment principle of the variable step size LMS adaptive filtering algorithm, the absolute value is added to the error signal in the second variable step size function and the third adjustment parameter is introduced to obtain the step size adjustment function; among them, the step size adjustment principle is : In the initial convergence stage or when the unknown system parameters change, the step size should be relatively large in order to have a faster convergence speed or tracking speed of the time-varying system; after the algorithm converges, the adjustment step size should be kept small In order to achieve a small steady-state imbalance. That is to say, the variable step size LMS algorithm has a larger step size factor at the initial stage of convergence to obtain a faster convergence speed; when it is about to enter the steady state stage, the step size factor is slowly reduced to a smaller value to obtain a better steady-state performance. Therefore, adding a third adjustment parameter can dynamically adjust the step size function to meet the step size adjustment principle, and the step size factor is a positive number, and adding an absolute value to the error function makes the step size adjustment function an even function to meet the requirements.

In a specific embodiment, the formula expression of the step size adjustment function is as follows:

In the formula, μ(n) is the step size adjustment function, and m is the third adjustment parameter.

The main invention point in the present invention is to improve the LMS algorithm, so that the traditional fixed step size becomes a step size adjustment function formed by the new nonlinear relationship established by the improved skip line function and the error signal. The IVSSLMS algorithm proposed by the present invention has established a balance between complexity and convergence. Compared with the classic variable step size LMS algorithm - SVSLMS and TCLMS algorithms, the performance simulation results of the step size update changes of the three algorithms are shown in the figure 4, as can be seen from the curve changes of different functions in Fig. 4, the SVSLMS algorithm and the TCLMS algorithm produce a sudden mutation when the error factor is close to 0, and the IVSSLMS algorithm proposed by the present invention overcomes this shortcoming, and when the error signal is close to When the step size is 0, the change is relatively gentle, which not only provides a larger step size when the error is large, but also improves the convergence speed of the algorithm.

After constructing the step size adjustment function, the weight coefficient is updated by the following formula:

w(n+1)=w(n)+2μ(n)e(n)s(n)

In the formula, w(n) is the weight coefficient of the filter at the current moment, and w(n+1) is the weight coefficient of the filter at the next moment.

S4. Iteratively updating the error signal according to the updated weight coefficient, and outputting the updated error signal and its corresponding output signal when the power of the updated error signal reaches a minimum value;

Specifically, after the weight coefficient is updated, enter the loop steps S1-S3 of the IVSSLMS algorithm until the power of the updated error signal reaches the minimum value, and then end the loop, and output the updated error signal and its corresponding output signal. In general, the evaluation standard of filter performance is usually measured by mean square error (Mean Square Error, MSE), that is, the smaller the mean square error, the better the performance, and the mean square error formula is expressed as follows:

MSE=E[e ² (n)]

In the formula, E is the mean value. The convergence criterion of the LMS algorithm is to minimize the power of the error signal, which is also used in the present invention. That is to say, when the weight coefficient vector of the adaptive filter is close to the Wiener solution, the filter weight coefficient vector can be equivalent to the best self-interference channel estimation, and the self-interference cancellation performance at this time reaches the best.

The simulation effect diagram of the mean square error performance under different variable step size LMS algorithms is shown in Figure 5. Under the same input signal and the same simulation conditions, the convergence speed of the SVSLMS algorithm is the slowest, and it starts to converge after 300 iterations; the TCLMS algorithm Convergence begins after 200 iterations; and the IVSSLMS algorithm proposed by the present invention begins to converge after 100 iterations, showing a faster convergence speed, and when the 500th iteration occurs, the phase of the signal changes, it can be seen that the present invention Compared with the other two algorithms, the proposed improved algorithm maintains better tracking ability. Fig. 6, 7 respectively represent the mean square error performance simulation effect figure of different variable step length LMS algorithms under low signal-to-noise ratio, can see from the figure that the IVSSLMS algorithm proposed by the present invention can still keep good under low signal-to-noise ratio conditions performance.

In a specific embodiment, when the interference cancellation ratio reaches the maximum value, the iterative update process is ended, and the updated error signal and its corresponding output signal are output; wherein, the calculation formula of the interference cancellation ratio is as follows:

In the formula, ICR is the interference cancellation ratio, P _r is the power of baseband received signal, P _d is the power of remote desired signal, σ _ε ² is the power of additive noise signal, E{|e(∞)| ² } is The mean square error as n tends to infinity.

The interference cancellation ratio ICR after self-interference cancellation in the digital domain is also an important indicator for judging the interference cancellation ability. The larger the ICR value, the better the self-interference cancellation ability. The value of the interference cancellation ratio can be used as the end of the cycle process. condition, when the ICR reaches the maximum value, the loop process ends.

The comparison chart of ICR under different variable step size LMS algorithms is shown in Figure 8. From Figure 8, it can be seen that the IVSSLMS algorithm is applied to full-duplex ISAC digital domain self-interference cancellation, and the algorithm can achieve convergence after 50 iterations, while The TCLMS algorithm and the SVSLMS algorithm need 200 and 250 iterations respectively to achieve the convergence process. Compared with the other two algorithms, this algorithm can achieve an interference cancellation ratio of 43dB, which is more than 2dB higher than the other two algorithms. The offset effect can not only meet the normal communication of the full-duplex ISAC system, but also achieve a faster convergence speed and a higher interference cancellation ratio.

Therefore, the process of the improved LMS algorithm - IVSSLMS algorithm is summarized in the following table:

当e(n)→∞，/>

可以初步判断步长μ(n)的范围是(0，0.5β)，β是调节μ(n)取值范围的系数，影响步长μ(n)变化的上限，α、m是调节μ(n)变化趋势的系数，影响步长的变化速率。α, β, m are all greater than 0, and when e(n)→0,

When e(n)→∞, />

It can be preliminarily judged that the range of the step size μ(n) is (0, 0.5β), β is the coefficient to adjust the value range of μ(n), and affects the upper limit of the change of the step size μ(n), α and m are the adjustments of μ( n) The coefficient of the change trend, which affects the rate of change of the step size.

The present invention studies the influence of parameters α, β and m on the step factor by controlling the variable method, that is, by fixing two parameters and changing another parameter to determine its influence on the fluctuation of the step factor. When β=1, m=2, and α are 0.5, 1, 2, and 5 respectively, the function model curves of step factor and error are shown in Fig. 9, and all four curves satisfy the step size adjustment principle. When the error is 4 (the error is relatively large at this time), most of the curves in the figure have reached or approached the maximum value of the step size factor, indicating that the convergence speed at this time is very fast, which is conducive to quickly changing towards the direction of error reduction ; As the error gradually decreases to zero, the step factor also gradually decreases to zero. However, since the steady-state error is also related to the step size, when the system is close to convergence, the rapidly changing step size will have a greater impact on the steady-state misalignment, which may lead to larger oscillations. Therefore, when the system is close to convergence, in order to ensure the performance of the algorithm, the step size factor is required to be slowly reduced to zero. Therefore, considering the above factors, the α value should be selected within the range of 1 to 4, and the preferred α=2 of the present invention, when the algorithm has a large error, the value of the step factor is larger, which ensures that the algorithm convergence speed is faster, When the error is close to zero, the step length curve is gradually close to the zero scale line, and the change range is relatively gentle, and the algorithm obtains higher stability.

When α=2, β=1, and m take 0.5, 1, 2, and 5 respectively, the function model curve of the step factor and the error is shown in Figure 10. When the value of m is small, the error e(n) is close to When the step size is 0, the step size cannot decrease smoothly, so m should be greater than 1. In order to ensure that the step size can quickly track the synchronous change of the error, the value of m should be made larger, but when m is too large, the complexity of the function will increase, the amount of calculation will increase, and the step size will be 0 within the small error range. Therefore, considering all factors, the value of m is preferably 3.

When α=2, m=3, and β are respectively 0.05, 0.1, 0.2, and 0.5, the function model curves of the step size factor and the error are shown in Figure 11. It can be seen from the relationship between the iteration step size and the parameter β that for For the same initial error, the larger the value of β, the faster the convergence speed of the algorithm in the initial convergence stage. If the value of β is too large, although the convergence speed is improved, the value of the step size μ(n) corresponding to e(n) after the algorithm converges is also large, resulting in a large steady-state error and algorithm divergence. If the demand for convergence speed is high in practical applications, a larger β value can be selected; if the demand for steady-state error is high, a smaller β value can be selected. However, it should be noted that when the value of β is too small, the iterative change interval of the step size μ(n) is small, and the variable step size LMS algorithm will degenerate approximately into a fixed step size LMS algorithm at this time, and the convergence speed will not be greatly improved. improve. Therefore, the present invention preferably sets α to 2, m to 3, and β to variable parameters, but μ(n) is proportional to β, and the optimal value of parameter β can be determined according to actual needs.

In the embodiment of the present invention, a self-interference cancellation method based on the improved LMS algorithm is designed based on the inherent defects of the fixed step size value, which cannot reasonably coordinate the convergence speed and steady-state error, and the existing improved LMS algorithm. The output signal of the filter is calculated according to the transmission signal sent by the transmitter of the full-duplex ISAC system and the weight coefficient of the filter; the error is calculated according to the baseband receiving signal and output signal in the full-duplex ISAC system, and the error signal is obtained; according to The error signal and the improved skip line function are used to construct a step adjustment function, and the weight coefficient is iteratively updated according to the step adjustment function; the error signal is iteratively updated according to the updated weight coefficient, when the power of the updated error signal When the minimum value is reached, the technical scheme of outputting the updated error signal and its corresponding output signal; so that the improved algorithm in the self-interference cancellation in the digital domain of the full-duplex ISAC system can still maintain the low signal-to-noise ratio. Good performance and low computational complexity while maintaining fast convergence.

It should be noted that although the various steps in the above flow chart are displayed sequentially according to the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise specified in the present invention, the execution of these steps is not strictly limited to the order, and these steps can be executed in other orders.

In another embodiment, as shown in FIG. 12 , the second aspect of the present invention provides a self-interference cancellation system based on an improved LMS algorithm, including:

The output signal acquisition module 11 is used to calculate the output signal of the filter according to the transmission signal sent by the full-duplex ISAC system transmission end and the weight coefficient of the filter;

The error signal acquisition module 12 is used for performing error calculation according to the baseband receiving signal and the output signal in the full-duplex ISAC system to obtain the error signal;

Weight coefficient update module 13, is used for according to error signal and improved skip line function, constructs step size adjustment function, and according to step size adjustment function weight coefficient is iteratively updated;

The loop cutoff output module 14 is configured to iteratively update the error signal according to the updated weight coefficient, and output the updated error signal and its corresponding output signal when the power of the updated error signal reaches a minimum value.

In a specific embodiment, the weight coefficient updating module 13 includes:

The step size function construction module is used to substitute the error function as an independent variable into the improved skip line function, and according to the step size adjustment principle, add an absolute value to the error signal and introduce a step size adjustment parameter to obtain a step size adjustment function.

In a specific embodiment, the system further includes a filter initialization module 10, configured to initialize the filter so that the weight coefficient is 0.

It should be noted that each module in the above self-interference cancellation system based on the improved LMS algorithm can be fully or partially realized by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can invoke and execute the corresponding operations of the above-mentioned modules. For the specific definition of a self-interference cancellation system based on the improved LMS algorithm, please refer to the above-mentioned definition of a self-interference cancellation method based on the improved LMS algorithm. The two have the same function and effect, and will not be repeated here.

A third aspect of the present invention provides an electronic device, the electronic device comprising:

Processor, memory and bus;

The bus is used to connect the processor and the memory;

The memory is used to store operation instructions;

The processor is configured to invoke the operation instruction, and the executable instruction causes the processor to execute the operation corresponding to the self-interference cancellation method based on the improved LMS algorithm as shown in the first aspect of the present invention.

An optional embodiment provides an electronic device, as shown in FIG. 13 , the electronic device 5000 shown in FIG. 13 includes: a processor 5001 and a memory 5003 . Wherein, the processor 5001 is connected to the memory 5003 , such as through a bus 5002 . Optionally, the electronic device 5000 may further include a transceiver 5004 . It should be noted that in practical applications, the transceiver 5004 is not limited to one, and the structure of the electronic device 5000 does not limit the embodiment of the present invention.

The processor 5001 may be a CPU, a general processor, DSP, ASIC, FPGA or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute the various illustrative logical blocks, modules and circuits described in connection with the present disclosure. The processor 5001 may also be a combination that implements computing functions, for example, a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

Bus 5002 may include a path for communicating information between the components described above. The bus 5002 can be a PCI bus or an EISA bus, etc. The bus 5002 can be divided into address bus, data bus, control bus and so on. For ease of representation, only one thick line is used in FIG. 13 , but it does not mean that there is only one bus or one type of bus.

Memory 5003 can be ROM or other types of static storage devices that can store static information and instructions, RAM or other types of dynamic storage devices that can store information and instructions, and can also be EEPROM, CD-ROM or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or can be used to carry or store desired program code in the form of instructions or data structures and can be programmed by a computer Any other medium accessed, but not limited to.

The memory 5003 is used to store application program codes for executing the solutions of the present invention, and the execution is controlled by the processor 5001 . The processor 5001 is configured to execute the application program code stored in the memory 5003, so as to realize the content shown in any one of the foregoing method embodiments.

Among them, electronic devices include but are not limited to: mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc. Mobile terminals such as digital TVs, desktop computers, etc. and fixed terminals.

The fourth aspect of the present invention provides a computer-readable storage medium, and a computer program is stored on the computer-readable storage medium. When the program is executed by a processor, an improved LMS algorithm based on the improved LMS algorithm shown in the first aspect of the present invention is implemented. Self-interference cancellation method.

Another embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is run on a computer, the computer can execute the corresponding content in the foregoing method embodiments.

In addition, an embodiment of the present invention also proposes a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of the above method are implemented.

In summary, in order to solve the inherent defects of the fixed step value that cannot reasonably coordinate the convergence speed and steady-state error and the problems existing in the existing improved LMS algorithm, the present invention provides a self-interference cancellation method, system, For equipment and media, the step size adjustment function is constructed by establishing a new nonlinear relationship between the improved skip line function and the error signal, and then the step size factor of the weight coefficient is updated, and the size of the step size factor can be dynamically changed during the algorithm convergence process , so that the improved LMS algorithm can maintain a low computational complexity while maintaining fast convergence, and can still maintain good performance under the condition of low signal-to-noise ratio. In domain self-interference cancellation, the ICR in the digital domain is higher than other compared algorithms and has a faster convergence speed.

Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for relevant parts, refer to part of the description of the method embodiment. It should be noted that the various technical features of the above-mentioned embodiments can be combined arbitrarily. For the sake of concise description, all possible combinations of the various technical features in the above-mentioned embodiments are not described. Where there is a contradiction, all should be deemed to be within the scope of this specification.

The above-mentioned examples only express several preferred implementation modes of the present invention, and the description thereof is relatively specific and detailed, but should not be construed as limiting the patent scope of the invention. It should be noted that those skilled in the art can make some improvements and substitutions without departing from the technical principle of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present invention. Therefore, the protection scope of the patent for the present invention should be based on the protection scope of the claims.

Claims (10)

1. A self-interference cancellation method based on an improved LMS algorithm, comprising the steps of:

calculating an output signal of a filter according to a transmission signal transmitted by a full duplex ISAC system transmitting end and a weight coefficient of the filter;

performing error calculation according to a baseband receiving signal and the output signal in the full duplex ISAC system to obtain an error signal;

constructing a step length adjusting function according to the error signal and the improved skip tongue function, and iteratively updating the weight coefficient according to the step length adjusting function;

and iteratively updating the error signal according to the updated weight coefficient, and outputting the updated error signal and the corresponding output signal when the power of the updated error signal reaches the minimum value.

2. The method for self-interference cancellation based on improved LMS algorithm as claimed in claim 1, wherein before said calculating the output signal of the filter according to the transmission signal transmitted from the transmitting end of the full duplex ISAC system and the weight coefficient of the filter, comprising:

initializing the filter to make the weight coefficient be 0.

3. A method of self-interference cancellation based on an improved LMS algorithm as claimed in claim 1, wherein said constructing a step size adjustment function based on said error signal and an improved skip tongue function comprises:

substituting the error function as an independent variable into an improved skip tongue function, adding an absolute value to the error signal according to a step length adjustment principle, and introducing an adjustment parameter to obtain the step length adjustment function.

4. A method of self-interference cancellation based on an improved LMS algorithm as claimed in claim 3, wherein said step size adjustment function is expressed as:

wherein μ (n) is a step size adjustment function, α is a first adjustment parameter, β is a second adjustment parameter, m is a third adjustment parameter, α, β, m are all greater than 0, and e (n) is an error function.

5. The method for self-interference cancellation based on improved LMS algorithm as claimed in claim 1, wherein said baseband received signal is a signal after antenna domain self-interference cancellation, analog domain self-interference cancellation and ADC quantization in said full duplex ISAC system, and the expression of said baseband received signal is:

r(n)＝s _I (n)+d(n)+ε(n)

where r (n) is the baseband received signal, s _I (n) is the residual self-interference signal after the self-interference cancellation of the antenna domain and the self-interference cancellation of the analog domain, d (n) is the far-end expected signal, and epsilon (n) is the additive noise.

6. A method of self-interference cancellation based on an improved LMS algorithm as claimed in claim 1, further comprising: when the interference cancellation ratio reaches the maximum value, ending the iterative updating process, and outputting an updated error signal and a corresponding output signal thereof; wherein, the calculation formula of the interference cancellation ratio is as follows:

wherein ICR is interference cancellation ratio, P _r For the power of the baseband received signal, P _d Work for remote desired signalRate, sigma _ε ² E { E (++E) is the power of the additive noise signal ² And n is the mean square error when n approaches infinity.

7. A self-interference cancellation system based on an improved LMS algorithm, comprising:

the output signal acquisition module is used for calculating the output signal of the filter according to the transmission signal transmitted by the transmitting end of the full-duplex ISAC system and the weight coefficient of the filter;

the error signal acquisition module is used for carrying out error calculation according to a baseband receiving signal and the output signal in the full-duplex ISAC system to obtain an error signal;

the weight coefficient updating module is used for constructing a step length adjusting function according to the error signal and the improved skip tongue function and iteratively updating the weight coefficient according to the step length adjusting function;

and the loop cut-off output module is used for carrying out iterative updating on the error signal according to the updated weight coefficient, and outputting the updated error signal and the corresponding output signal when the power of the updated error signal reaches the minimum value.

8. A self-interference cancellation system based on an improved LMS algorithm as claimed in claim 7, wherein said weight coefficient update module comprises;

the step length function construction module is used for substituting the error function as an independent variable into the improved skip tongue function, adding an absolute value to the error signal according to a step length adjustment principle, and introducing an adjustment parameter to obtain the step length adjustment function.

9. An electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the improved LMS algorithm-based self-interference cancellation method of any one of claims 1 to 6 when the computer program is executed.

10. A computer readable storage medium, characterized in that the computer readable storage medium comprises a stored computer program, wherein the computer program when run controls a device in which the computer readable storage medium is located to perform the self-interference cancellation method based on the modified LMS algorithm as claimed in any one of claims 1 to 6.

Images