FMCW biological radar indoor mutual interference suppression method

Abstract

The invention belongs to the technical field of radar mutual interference suppression, and particularly relates to an FMCW biological radar indoor mutual interference suppression method, which comprises the following steps: S1, constructing a deep reinforcement learning (DRL)-based adaptive signal subspace projection mutual interference suppression framework, comprising a frequency modulation interference feature recognition module, a Hansel-SVD module, a signal subspace projection module, a reward calculation module, a regulation factor diagonal matrix construction module and a depth deterministic policy gradient algorithm (DDPG) agent. And S2, training a frequency modulation interference feature recognition regression neural network mapping model by using the simulation data set to serve as a frequency modulation interference feature recognition module. According to the method, a self-adaptive signal subspace projection interference suppression framework is constructed based on the DRL technology, and when the problem of mutual interference suppression of a non-convex optimization problem is faced, dynamic changes of the problem are adapted through rewards fed back by the environment, so that the technical problem that the optimization process of a previous signal separation method is faced with NP difficulties is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of radar mutual interference suppression, and in particular to a method for suppressing indoor mutual interference of an FMCW bio-radar. Background Art

[0002] Compared to sensors like wearable devices and cameras, millimeter-wave bioradar offers advantages such as contactless operation, reduced privacy infringement, all-weather and all-light sensing capabilities, and the provision of fine-grained sensor information. It has been widely used in related fields such as smart homes and smart healthcare. With the further development of millimeter-wave bioradar technology, the number of millimeter-wave radar devices in daily life will gradually increase, providing all-weather support. When multiple FMCW bioradar devices operate in the same indoor environment, the limited available spectrum resources and the instantaneous superposition of different radar signals will inevitably cause interference, thereby impairing the target perception radar's ability to extract and identify target signature signals. When the frequency modulation slopes of the jamming radar and the target perception radar are the same, the high overlap in their frequency modulation patterns makes it very easy for the perception radar to detect false target reflections unrelated to the actual target, resulting in false target detection results. When the frequency modulation slopes of the two radars differ, the jamming radar's transmitted signal will overlay the actual target echo in the spectrum, causing significant deviations in the target signal's time-frequency characteristics. Mutual interference not only reduces the accuracy and real-time performance of activity monitoring, but also prevents timely and accurate identification and early warning of real falls, posing a potential threat to the safety and health of the monitored individuals. Therefore, we propose a method for suppressing indoor mutual interference using FMCW bioradar to address this issue. Summary of the Invention

[0003] (1) Technical problems solved

[0004] In view of the shortcomings of the existing technology, the present invention provides a method for suppressing indoor mutual interference of FMCW bioradar, which solves the problems raised in the above background technology.

[0005] (2) Technical solution

[0006] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0007] A method for suppressing indoor mutual interference of FMCW bioradar comprises the following steps:

[0008] S1: Construct a framework for suppressing mutual interference based on adaptive signal subspace projection based on deep reinforcement learning (DRL).

[0009] It includes FM interference feature recognition module, Hankel-SVD module, signal subspace projection module, reward calculation module, control factor diagonal matrix construction module, and deep deterministic policy gradient algorithm (DDPG) agent;

[0010] S2: Use the simulation data set to train a FM interference feature recognition regression neural network mapping model as the FM interference feature recognition module;

[0011] S3: The control factor diagonal matrix construction module constructs the DRL continuous action space data into a diagonal matrix and inputs it into the signal subspace projection module;

[0012] S4: Based on the characteristics of useful signals being continuous in time domain and sparse in frequency domain, a DRL reward calculation module is constructed with the goal of maximizing the useful signal component in the original signal;

[0013] S5: Use the simulation data set to pre-train the DRL framework, then use the measured data to fine-tune it to obtain the mutual interference suppression DRL model, and use the obtained DRL model to suppress the mutual interference signal in the original signal.

[0014] Furthermore, the simulation data set in S2 is: in the same perception space, based on the given rectangular room parameters and the installation coordinates of the perception radar and the interference radar, and the perception radar and the interference radar perceive the same target at the same time, through the FMCW bioradar mutual interference signal model in the indoor environment, the original data of the perception radar being interfered with is obtained.

[0015] Furthermore, the FM interference feature recognition module in S2 is: constructing a data set based on the Monte Carlo simulation method to train a regression neural network, and the data set is based on the signal interference noise ratio , the proportion of interference components , the frequency modulation slope of the jamming signal emitted by the jamming radar The label is the label, and the analog original IF signal data of N sampling points is the feature.

[0016] Furthermore, the control factor diagonal matrix construction module in S3 is used to construct the DRL continuous action space data into a diagonal matrix. The DRL continuous action space data is , the negative regulation factor represents the weakening of the corresponding subspace, and the positive regulation factor represents the retention or enhancement of the corresponding subspace. The constructed diagonal matrix of regulation factors is .

[0017] Furthermore, the signal subspace projection module in S3 is used to separate the low-rank approximation matrix of the useful signal by using the diagonal matrix of the control factor and the eigenvalues ​​and eigenvectors input into the module: , the low-rank approximation matrix of other signals is ,in Represents the identity matrix.

[0018] Furthermore, the Hankel-SVD module in S1 is used to convert the discrete raw data of the input system into Lifting to Hankel matrix After that, the eigenvalue matrix of the discrete original data is obtained by SVD decomposition With the eigenvector and , and input the eigenvalues ​​and eigenvectors into the signal subspace projection module.

[0019] Furthermore, the specific content of using the obtained DRL model in S5 to suppress the mutual interference signal in the original signal is: the original signal, that is, the IF signal of the interfered FMCW bioradar, is input into the mutual interference suppression DRL model after passing through a low-pass filter with a cutoff frequency matching the effective perception distance of the radar, and a high-fidelity target useful signal is separated from the interfered signal.

[0020] (3) Beneficial effects

[0021] Compared with the prior art, the present invention provides a method for suppressing indoor mutual interference of FMCW bioradar, which has the following beneficial effects:

[0022] The present invention constructs an adaptive signal subspace projection interference suppression framework based on DRL technology. When faced with the mutual interference suppression problem of non-convex optimization problems, the dynamic changes of the reward adaptation problem of environmental feedback are used to solve the NP-hard technical problems faced by the previous signal separation method optimization process. The pre-trained FM interference feature recognition regression neural network mapping model serves as the FM interference feature recognition module in the above-mentioned interference suppression framework. While improving the DRL training efficiency, it reduces the DRL observation space dimension and avoids the risk of state space explosion. A mutual interference suppression action space suitable for DRL is constructed, and the DRL action space data is constructed as an adaptive signal subspace projection matrix. The target useful signal is obtained by adaptively projecting the non-orthogonal signal subspace, and the signal projection adjustment factor is adaptively adjusted in a variety of complex indoor environments, thereby separating the time-frequency characteristics of the useful signal with high precision. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 Schematic diagram of the method flow of the present invention;

[0024] Figure 2 This is a schematic diagram of the framework structure of the mutual interference suppression based on deep reinforcement learning adaptive signal subspace projection of the present invention. DETAILED DESCRIPTION

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 making creative efforts are within the scope of protection of the present invention.

[0026] Example

[0027] like Figure 1-2 As shown, an embodiment of the present invention provides a method for suppressing indoor mutual interference of FMCW bio-radar, comprising the following steps:

[0028] S1: Construct a framework for suppressing mutual interference based on adaptive signal subspace projection based on deep reinforcement learning (DRL).

[0029] It includes FM interference feature recognition module, Hankel-SVD module, signal subspace projection module, reward calculation module, control factor diagonal matrix construction module, deep deterministic policy gradient algorithm (DDPG) agent, such as Figure 2 As shown;

[0030] The Hankel-SVD module is used to convert the discrete raw data of the input system into Lifting to Hankel matrix The particularity of the Hankel matrix construction can highlight the continuity of the useful signal in the time domain in the matrix, and the eigenvalues ​​obtained by SVD decomposition With the eigenvector 、 It can effectively reflect this continuity.

[0031] S2: Use the simulation data set to train a FM interference feature recognition regression neural network mapping model as the FM interference feature recognition module;

[0032] The simulation data set is as follows: in the same perception space, based on the given rectangular room parameters and the installation coordinates of the perception radar and the jamming radar, and the perception radar and the jamming radar perceive the same target at the same time, the raw data of the perception radar being interfered with is obtained through the FMCW bio-radar mutual interference signal model in the indoor environment.

[0033] The mutual interference signal between two FMCW bioradars in an indoor environment consists of the direct path interference signal generated in the simulated indoor environment, the target scattered interference signal, and the target useful echo signal. The mathematical formula of the above mutual interference signal model is as follows:

[0034] The transmitted signal of the perception radar can be expressed as:

[0035]

[0036] in is the initial phase of the signal, Indicates all time ,in It's fast time, Indicates slow time, Indicates the carrier frequency, Indicates the frequency modulation slope.

[0037] The instantaneous distance between the target and the radar is:

[0038]

[0039] in , represents the initial distance, Indicates the target movement speed.

[0040] The useful signal in the target echo of the perception radar can be approximately expressed as:

[0041]

[0042] Direct path interference, that is, the signal from the transmitting antenna of the interfering radar and reaching the receiving antenna of the sensing radar, causes the signal delay received by the sensing radar to be:

[0043]

[0044] The signal delay of the sensing radar caused by target scattering interference, that is, the signal from the transmitting antenna of the jamming radar to the receiving antenna of the sensing radar after being scattered by the target, is:

[0045]

[0046] The delay caused by direct path interference and scattered interference can be expressed as:

[0047]

[0048] The interference occurrence time is defined as , the duration of the interference is Therefore, the interference signal received by the perception radar is expressed as:

[0049]

[0050] is the initial phase of the intrusion radar's transmitted signal; the echo signal and the interference signal have the same carrier frequency. The FMCW radar system receives both simultaneously. The IF signal obtained by mixing the interference signal with the conjugate signal of the transmitted signal can be approximately expressed as:

[0051]

[0052] in Represents the inherent noise floor of the radar. The simulated original IF signal output by the mutual interference signal model is Data of each fast time dimension.

[0053] The FM interference feature recognition module is as follows: a data set is constructed based on the Monte Carlo simulation method to train a regression neural network. The data set is based on the signal interference noise ratio , the proportion of interference components , interference signal FM slope The label is the label, and the analog original IF signal data of N sampling points is the feature.

[0054] S3: The control factor diagonal matrix construction module constructs the DRL continuous action space data into a diagonal matrix and inputs it into the signal subspace projection module;

[0055] The control factor diagonal matrix construction module is used to construct the DRL continuous action space data into a diagonal matrix. The DRL continuous action space data is , the negative regulation factor represents the weakening of the corresponding subspace, and the positive regulation factor represents the retention or enhancement of the corresponding subspace. The constructed diagonal matrix of regulation factors is .

[0056] The signal subspace projection module is used to separate the low-rank approximation matrix of the useful signal using the diagonal matrix of the control factor and the eigenvalues ​​and eigenvectors input into the module: , the low-rank approximation matrix of other signals is ,in Represents the identity matrix.

[0057] S4: Based on the characteristics of useful signals being continuous in time domain and sparse in frequency domain, a DRL reward calculation module is constructed with the goal of maximizing the useful signal component in the original signal;

[0058] Through the objective function and constraints Calculate the reward obtained by the agent in the current round, and then input the reward value into the DDPG agent.

[0059] S5: Pre-train the DRL framework using a simulation dataset, then fine-tune it using measured data to obtain a mutual interference suppression DRL model, and use the obtained DRL model to suppress the mutual interference signal in the original signal;

[0060] The specific content of using the obtained DRL model to suppress the mutual interference signal in the original signal is: the original signal, that is, the IF signal of the interfered FMCW bioradar, is passed through a low-pass filter with a cutoff frequency matching the effective perception distance of the radar, and then input into the mutual interference suppression DRL model to separate the high-fidelity target useful signal from the interfered signal.

[0061] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for suppressing indoor mutual interference of FMCW bio-radar, characterized by: The steps include: S1: Construct a framework for mutual interference suppression based on deep reinforcement learning (DRL) adaptive signal subspace projection. It includes FM interference feature recognition module, Hankel-SVD module, signal subspace projection module, reward calculation module, control factor diagonal matrix construction module, and deep deterministic policy gradient algorithm (DDPG) agent; S2: Use the simulation data set to train a FM interference feature recognition regression neural network mapping model as a FM interference feature recognition module; S3: The control factor diagonal matrix construction module constructs the DRL continuous action space data into a diagonal matrix and inputs it into the signal subspace projection module; S4: Based on the characteristics of useful signals being continuous in time domain and sparse in frequency domain, a DRL reward calculation module is constructed with the goal of maximizing the useful signal component in the original signal; S5: Use the simulation data set to pre-train the DRL framework, and then use the measured data to fine-tune it to obtain the mutual interference suppression DRL model, and use the obtained DRL model to suppress the mutual interference signal in the original signal.

2. The method for suppressing indoor mutual interference of FMCW bio-radar according to claim 1, characterized in that: The simulation data set in S2 is: in the same perception space, based on the given rectangular room parameters and the installation coordinates of the perception radar and the interference radar, and the perception radar and the interference radar perceive the same target at the same time, and obtain the original data of the perception radar being interfered through the FMCW bio-radar mutual interference signal model in the indoor environment.

3. The method for suppressing indoor mutual interference of FMCW bio-radar according to claim 1, characterized in that: The FM interference feature recognition module in S2 is: based on the Monte Carlo simulation method, a data set is constructed to train a regression neural network. The data set is based on the signal interference noise ratio , the proportion of interference components , the frequency modulation slope of the jamming signal emitted by the jamming radar The label is the label, and the analog original IF signal data of N sampling points is the feature.

4. The method for suppressing indoor mutual interference of FMCW bio-radar according to claim 1, characterized in that: The control factor diagonal matrix construction module in S3 is used to construct the DRL continuous action space data into a diagonal matrix. The DRL continuous action space data is , the negative regulation factor represents the weakening of the corresponding subspace, and the positive regulation factor represents the retention or enhancement of the corresponding subspace. The constructed diagonal matrix of regulation factors is .

5. The method for suppressing indoor mutual interference of FMCW bio-radar according to claim 4, characterized in that: The signal subspace projection module in S3 is used to separate the low-rank approximate matrix of the useful signal by using the diagonal matrix of the control factor and the eigenvalues ​​and eigenvectors input into the module: , the low-rank approximation matrix of other signals is ,in Represents the identity matrix.

6. The method for suppressing indoor mutual interference of FMCW bio-radar according to claim 1, characterized in that: The Hankel-SVD module in S1 is used to upgrade the discrete original data of the input system to a Hankel matrix, obtain the eigenvalues ​​and eigenvectors of the discrete original data through SVD decomposition, and input the eigenvalues ​​and eigenvectors into the signal subspace projection module.

7. The method for suppressing indoor mutual interference of FMCW bio-radar according to claim 1, characterized in that: The specific content of using the obtained DRL model to suppress the mutual interference signal in the original signal in S5 is: the original signal, that is, the IF signal of the interfered FMCW bioradar, is input into the mutual interference suppression DRL model after passing through a low-pass filter with a cutoff frequency matching the effective perception distance of the radar, and a high-fidelity target useful signal is separated from the interfered signal.

Images