Electronic fence based on ultra wide band radar technology

Abstract

The invention relates to the technical field of signal processing, and further relates to an electronic fence based on an ultra-wideband radar technology, which comprises a signal transmitting unit used for generating an ultra-wideband signal; the signal model construction unit is used for constructing a propagation model of the modulation signal in the space; the target object model construction unit is used for constructing an anisotropic scattering model of the target so as to calculate a radar cross section of the target; the receiving signal model building unit is used for building a receiving signal model according to the radar cross section of the target; the analysis unit is used for enhancing the detection capability of the target in the received signal and setting a detection threshold so as to determine whether the target exists in the received signal or not; and the target motion estimation unit is used for performing target motion process estimation by using a Kalman filter. According to the invention, the accuracy of the electronic fence is improved through the ultra-wideband radar technology, signal processing and multi-channel pulse-Doppler processing.

Description

Technical Field

[0001] The present invention belongs to the technical field of signal processing, and particularly relates to an electronic fence based on ultra-wideband radar technology. Background Art

[0002] Radar technology is a radio spectrum application widely used for target detection, tracking, and positioning. With the continuous progress of technology, radar systems play an increasingly important role in various fields such as aviation, meteorology, traffic management, and industrial fields. Radar systems detect target objects by transmitting radio waves and receiving the reflected signals, and their application scope is extensive, including aircraft navigation, vehicle autonomous driving, weather prediction, etc.

[0003] However, traditional radar systems have some challenges and limitations in certain aspects, such as the ability to distinguish target details, anti-interference performance, and the accuracy of target motion estimation. Therefore, new technologies are constantly proposed to solve these problems, and one of the key technologies is the electronic fence technology based on ultra-wideband radar, which uses ultra-wideband signals and advanced signal processing algorithms to improve the performance of radar systems.

[0004] Ultra-wideband radar technology is an emerging radar technology that uses ultra-wideband signals and has the following characteristics: relatively wide bandwidth: Ultra-wideband signals have a very wide bandwidth, so they have higher time resolution and can effectively distinguish approaching targets. Anti-interference: Ultra-wideband radar systems have a certain resistance to interference because they can use spectrum spreading and signal processing technologies to cancel interference. High-precision target motion estimation: Ultra-wideband radar systems can achieve more accurate target speed and direction estimation because they provide richer signal information. Summary of the Invention

[0005] The main purpose of the present invention is to provide an electronic fence based on ultra-wideband radar technology, which has the advantage of high target detection accuracy.

[0006] To solve the above problems, the technical solution of the present invention is realized as follows:

[0007] An electronic fence based on ultra-wideband radar technology, comprising: a signal transmitting unit, a signal propagation model construction unit, a target object model construction unit, a received signal model construction unit, an analysis unit, and a target motion estimation unit; the signal transmitting unit is used to generate an ultra-wideband signal, modulate the ultra-wideband signal through a pulse modulation function to obtain a modulated signal, and then transmit the modulated signal; the signal model construction unit is used to construct a propagation model of the modulated signal in space to describe the process of the modulated signal propagating from the radar emission point to the target and returning; the target object model construction unit is used to construct an anisotropic scattering model of the target according to the geometric shape, electromagnetic characteristics, and material parameters of the target, so as to calculate the radar cross section of the target; the received signal model construction unit is used to construct a received signal model according to the radar cross section of the target to describe the process of the received signal returning from the target to the radar receiver; the analysis unit is used to process the received signal using a matched filtering technique to enhance the characteristics of the received signal, apply a beamforming technique to improve the target resolution of the received signal by weighting and superimposing the received signals of multiple receiving channels, adopt an adaptive signal processing algorithm to suppress multipath interference and noise, and at the same time enhance the detection ability of the target in the received signal, and set a detection threshold to determine whether the target exists in the received signal; the target motion estimation unit is used to calculate a time-domain differential signal from the received signals of multiple receiving channels, combine the time-domain differential signals into a vector, estimate the speed and direction of the target based on multi-channel adaptive beamforming, and use a Kalman filter to estimate the target motion process.

[0008] Further, the modulated signal transmitted by the signal transmitting unit is represented by the following formula:

[0009] s(t) = A·cos(2πf c t)·g(t)·p(t);

[0010] where s(t) is the modulated signal; A·cos(2πf c t) represents the ultra-wideband signal; A is the amplitude of the ultra-wideband signal; f c is the center frequency of the ultra-wideband signal; g(t) is the pulse modulation function used to determine the time-domain characteristics of the signal; p(t) is the modulation pulse, expressed as:

[0011]

[0012] where N is the number of subcarriers in the modulation pulse; α i is the amplitude of each subcarrier; f i is the frequency of each subcarrier; φ i is the phase of each subcarrier.

[0013] Furthermore, the propagation model constructed by the signal model construction unit is represented by the following formula:

[0014]

[0015] where N p is the order of polynomial expansion; φ n is the phase of multipath reflection, considering the interference of different paths; p(t, r) represents the complex amplitude of the modulated signal at time t and position r; 4πR is the propagation path length of the modulated signal, where R represents the distance from the radar emission point to the target or reflection point;

[0016] represents the coefficient of polynomial expansion, where k is the wave number, R is the distance, and n is the ordinal number of the multipath; is the phase term of the signal, c is the speed of light, and φ n represents the phase of multipath reflection.

[0017] Furthermore, the anisotropic scattering model constructed by the target object model construction unit is represented by the following formula:

[0018]

[0019] where: RCS(θ, φ) is the radar cross section of the target object, representing the reflection intensity of the target to the modulated signal. It is a function related to the incident angle and azimuth angle, used to describe the reflection characteristics of the target in different directions; |F(θ, φ)| 2 represents the anisotropic scattering model of the target, which is obtained by modeling according to the geometric shape, electromagnetic characteristics, and material parameters of the target; σ(θ, φ) is the local radar cross section of the target, which represents the scattering cross section value of the target in different directions; θ is the direction angle of the target relative to the radar antenna. When θ is 0°, it means the target is directly in front of the radar antenna. A positive value represents the angle in the clockwise direction, and a negative value represents the angle in the counterclockwise direction; φ is the phase of the modulated signal.

[0020] Furthermore, the received signal model constructed by the received signal model construction unit is represented by the following formula:

[0021]

[0022] where r(t, r) represents the received signal; ∫ 目标区域 is an integral symbol, representing the integration over the entire region where the target is located. It is used to integrate the signals of various reflections within the region where the target is located to calculate the received signal; dΩ is the solid angle element, used to integrate the signals reflected in different directions; j is the imaginary symbol.

[0023] Furthermore, the process of the analysis unit processing the received signal through matched filtering technology, beamforming technology, and adaptive signal processing algorithms is expressed using the following formula:

[0024]

[0025] where s MF (t) represents the received signal processed by beamforming technology and adaptive signal processing algorithms; is an integral symbol, representing the integration over the entire time domain; s(t') represents the received signal at time t'; Z(t - t') is the matched filter; w(t - t') is the adaptive weight function, expressed using the following formula:

[0026] w(t - t') = w0·e jφ(t-t′) ;

[0027] where w0 is the adaptive amplitude, a set value; φ(t - t') represents the phase at time t - t'.

[0028] Furthermore, the detection threshold set by the analysis unit is expressed using the following formula:

[0029]

[0030] where μ is the mean of the received signal; σ is the standard deviation of the received signal; ɑ is the adaptive threshold coefficient, a set value, with a value range of 1 to 3.

[0031] Furthermore, the target motion estimation unit calculates the time-domain differential signal from the received signals of multiple receiving channels using the following formula:

[0032] d i (t) = r i (t, r) - r i (t - Δt, r)

[0033] where r i (t, r) is the received signal of the i-th channel for which we calculate the time-domain differential signal; d i (t) is the time-domain differential signal; Δt is the time interval; the time-domain differential signals of multiple channels are combined into a vector d(t) = [d1(t), d2(t), …, d m (t)], where m is the number of channels; through multi-channel pulse-Doppler processing, the velocity vector v(t) and the direction vector u(t) are estimated.

[0034] Further, for the target motion estimation unit, the method of using a Kalman filter to estimate the target motion process includes: decomposing the direction vector into direction components in three directions, namely x(t), y(t), and z(t); decomposing the velocity vector into velocity components in three directions, namely and Thus, the state vector of the target is represented as Perform state prediction using the Kalman filter through the following formula:

[0035]

[0036] Perform state update using the Kalman filter through the following formula:

[0037]

[0038] Among them, P pred (t) is the state prediction covariance matrix; K(t) is the Kalman gain; z(t) is the observation residual; x pred (t) is the state prediction vector; F(t) is the state transition matrix, which is represented by the following formula:

[0039]

[0040] P(t) is the state covariance matrix, which is represented by the following formula

[0041]

[0042] H(t) is the observation matrix, which is represented by the following formula:

[0043]

[0044] An electronic fence based on ultra-wideband radar technology of the present invention has the following beneficial effects: The present invention adopts ultra-wideband radar technology. By transmitting ultra-wideband signals, it has a higher bandwidth and time resolution, and can effectively distinguish approaching targets, improving the accuracy of target detection and positioning. This is crucial for various application scenarios, such as obstacle recognition in autonomous driving and ground target detection in aircraft navigation. The present invention introduces matched filtering technology, beamforming technology, and adaptive signal processing algorithms to suppress multipath interference and noise. The matched filtering technology can effectively extract target signals. The beamforming technology improves the signal-to-noise ratio by weighting and superimposing signals from multiple receiving channels. The adaptive signal processing algorithm further suppresses the influence of multipath interference and noise. Therefore, the present invention can provide more reliable target detection and positioning results in complex environments. The present invention uses multi-channel pulse-Doppler processing and Kalman filters to estimate the target motion, thereby improving the accuracy of speed and direction estimation. The multi-channel pulse-Doppler processing can more accurately estimate the target speed, and the Kalman filter further improves the motion estimation result through state prediction and update. This will help improve the path planning of autonomous driving vehicles and the navigation accuracy of aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 FIG. is a schematic structural diagram of an electronic fence based on ultra-wideband radar technology provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0047] The following will be described in detail respectively.

[0048] Example 1: Refer to Figure 1, An electronic fence based on ultra-wideband radar technology, comprising: a signal transmitting unit, a signal propagation model construction unit, a target object model construction unit, a received signal model construction unit, an analysis unit, and a target motion estimation unit; the signal transmitting unit is used to generate an ultra-wideband signal, modulate the ultra-wideband signal through a pulse modulation function to obtain a modulated signal, and then transmit the modulated signal; the signal model construction unit is used to construct a propagation model of the modulated signal in space to describe the process of the modulated signal propagating from the radar emission point to the target and returning; the target object model construction unit is used to construct an anisotropic scattering model of the target according to the geometric shape, electromagnetic characteristics, and material parameters of the target, so as to calculate the radar cross section of the target; the received signal model construction unit is used to construct a received signal model according to the radar cross section of the target to describe the process of the received signal returning from the target to the radar receiver; the analysis unit is used to process the received signal using a matched filtering technique to enhance the characteristics of the received signal, apply a beamforming technique, sum and superimpose the received signals of multiple receiving channels through weighting to improve the target resolution of the received signal, adopt an adaptive signal processing algorithm to suppress multipath interference and noise, and at the same time enhance the detection ability of the target in the received signal, and set a detection threshold to determine whether the target exists in the received signal; the target motion estimation unit is used to calculate the time-domain differential signal from the received signals of multiple receiving channels, combine the time-domain differential signals into a vector, estimate the speed and direction of the target based on multi-channel adaptive beamforming, and use a Kalman filter to estimate the target motion process.

[0049] Specifically, the signal transmitting unit is responsible for generating an ultra-wideband signal and modulating it through a pulse modulation function. The ultra-wideband signal refers to a signal with an extremely wide frequency band and usually has a very short pulse width. This ultra-wideband signal has many advantages in radar applications, including high resolution and the ability to penetrate obstacles. Through pulse modulation, the signal can be better distinguished from the target reflection signal.

[0050] The signal propagation model construction unit is responsible for constructing a propagation model of the modulated signal in space to describe the process of the signal propagating from the radar emission point to the target and returning. This is a crucial step because different targets and environments will have different effects on signal propagation. This model can accurately consider effects such as multipath propagation and signal attenuation, thereby improving the positioning accuracy of the system for the target.

[0051] The target object model construction unit constructs an anisotropic scattering model of the target according to the geometric shape, electromagnetic characteristics, and material parameters of the target. This model is used to calculate the radar cross section of the target, which refers to the intensity of the target reflecting the radar signal. By considering the detailed characteristics of the target, the system can more accurately identify the target and distinguish it from background noise.

[0052] The received signal model construction unit constructs a received signal model based on the radar cross section of the target to describe the process of the received signal returning from the target to the radar receiver. This helps the system understand the characteristics of the target reflected signal, including signal delay and phase change.

[0053] The analysis unit processes the received signal using matched filtering techniques to enhance the characteristics of the received signal. This includes beamforming techniques that improve target resolution by weighting and summing signals from multiple receiving channels. In addition, adaptive signal processing algorithms are used to suppress multipath interference and noise while enhancing the detection ability for the target. Setting a detection threshold helps determine whether a target exists in the received signal. These techniques improve the system's detection and positioning capabilities for the target.

[0054] The target motion estimation unit calculates the time-domain differential signal from the received signals of multiple receiving channels and combines them into a vector. Based on multi-channel adaptive beamforming, it can estimate the speed and direction of the target. Using a Kalman filter for target motion process estimation helps track the target's motion trajectory, which is very important in electronic fence applications.

[0055] Embodiment 2: The modulated signal transmitted by the signal transmitting unit is represented by the following formula:

[0056] s(t) = A·cos(2πf c t)·g(t)·p(t);

[0057] where, s(t) is the modulated signal; A·cos(2πf c t) represents the ultra-wideband signal; A is the amplitude of the ultra-wideband signal; f c is the center frequency of the ultra-wideband signal; g(t) is the pulse modulation function used to determine the time-domain characteristics of the signal; p(t) is the modulation pulse, expressed as:

[0058]

[0059] where, N is the number of subcarriers in the modulation pulse; α i is the amplitude of each subcarrier; f i is the frequency of each subcarrier; φ i is the phase of each subcarrier.

[0060] Specifically, the ultra-wideband signal A·cos(2πf c t) is a signal based on the cosine function and has an extremely wide frequency band. It consists of the amplitude A and the center frequency f cControl. The periodicity of the cosine function results in multiple frequency components of the signal in the frequency domain, enabling it to transmit rich information. The ultra-wideband signal is the frequency-domain component of the entire modulated signal, providing the frequency-domain characteristics of the signal, which can be used for data transmission or radar target detection. It typically exists as an envelope for containing multiple subcarriers of the modulated pulse.

[0061] The pulse modulation function g(t) is used to determine the time-domain characteristics of the signal. It is usually a short pulse that controls the pulse shape and time-domain distribution of the signal. The pulse modulation function determines the time-domain characteristics of the signal and can be used to adjust the pulse width and shape of the signal to meet specific application requirements. For example, the range resolution of a radar system can be controlled by adjusting the pulse width.

[0062] Modulated pulse p(t): It consists of multiple subcarriers, each with different frequencies, amplitudes, and phases. The main function of the modulated pulse is to combine the ultra-wideband signal with multiple frequency components. This combination allows multiple frequency components to be included in the signal, thereby increasing the frequency-domain information of the signal, which can be used for different applications such as multiple access transmission, spectrum spreading, and target recognition.

[0063] The ultra-wideband signal provides an extremely wide frequency band, allowing for more information transmission or higher resolution in radar. The modulated pulse with multiple subcarriers further expands the frequency-domain characteristics of the signal.

[0064] Embodiment 3: The propagation model constructed by the signal model construction unit is represented by the following formula:

[0065]

[0066] Where N p is the order of the polynomial expansion; φ n is the phase of the multipath reflection, considering the interference of different paths; p(t, r) represents the complex amplitude of the modulated signal at time t and position r; 4πR is the propagation path length of the modulated signal, where R represents the distance from the radar emission point to the target or reflection point;

[0067] represents the coefficient of the polynomial expansion, where k is the wave number, R is the distance, and n is the ordinal number of the multipath; is the phase term of the signal, c is the speed of light, and φ n represents the phase of the multipath reflection.

[0068] Specifically, represents the attenuation of the signal amplitude with distance r. A is the amplitude of the signal, and 4πR is the propagation path length. This part ensures that the signal has appropriate intensity at different distances, in line with the physical principle of propagation loss. Represents the coefficients of polynomial expansion, used to account for multipath effects. Multipath effects refer to the phenomenon where a signal experiences reflections or scatterings through multiple different paths during propagation, resulting in multiple versions of the signal arriving at the receiver simultaneously. The coefficients of polynomial expansion are used to represent the contributions of different multipath paths. N p is the order of polynomial expansion, n is the ordinal number of the multipath path, k is the wave number, and R is the distance.

[0069] Represents the phase term of the signal, taking into account the phase differences of multipath reflections. Phase is the offset or relative time delay of the signal waveform, caused by the signal reflections and propagations along different paths. φ n Represents the phase of multipath reflection. f c is the center frequency of the signal, and c is the speed of light, used to convert the distance R into time delay.

[0070] The main function of this formula is to describe the complex amplitude variations of the signal along the propagation path, including amplitude attenuation, multipath effects, and phase changes. The first term describes the decrease in the amplitude of the signal as the propagation distance increases. This is a physical phenomenon based on the length of the propagation path, ensuring that the signal has appropriate energy at different distances. The coefficient part of the polynomial expansion accounts for multipath effects, that is, the signal reaches the receiver after being reflected or scattered through different paths. These coefficients represent the signal distribution and intensity on different paths. The phase term describes the phase differences of the signals on different paths. This reflects the influence of multipath reflections, and the signal will have different phase offsets after passing through different paths.

[0071] Example 4: The anisotropic scattering model constructed by the target object model construction unit is represented by the following formula:

[0072]

[0073] Where: RCS(θ,φ) is the radar cross section of the target object, representing the reflection intensity of the target to the modulated signal. It is a function related to the incident angle and azimuth angle, used to describe the reflection characteristics of the target in different directions; |F(θ,φ)| 2 represents the anisotropic scattering model of the target, which is modeled based on the geometric shape, electromagnetic characteristics, and material parameters of the target; σ(θ,φ) is the local radar cross section of the target, which represents the cross section values of the target in different directions; θ is the direction angle of the target relative to the radar antenna. When θ is 0°, it means the target is directly in front of the radar antenna. Positive values represent angles in the clockwise direction, and negative values represent angles in the counterclockwise direction; φ is the phase of the modulated signal.

[0074] Specifically, the radar cross section RCS(θ,φ) represents the radar cross section of a target object in different directions. It is a function related to the incident angles (azimuth angle θ and phase angle φ). The radar cross section is used to describe the reflection intensity of a target for an incident signal and is a very important parameter in a radar system. The radar cross section is used to determine the reflection characteristics of a target in different directions. By measuring or modeling the radar cross sections in different directions, scattering information of the target at different angles can be obtained, which is used for target detection, tracking, and recognition in a radar system.

[0075] Anisotropic scattering model |F(θ,φ)| 2 Represents the anisotropic scattering model of a target, which is modeled based on the geometric shape, electromagnetic characteristics, and material parameters of the target. The anisotropic scattering model describes the differences in the reflection characteristics of a target in different directions. The anisotropic scattering model is used to describe the scattering characteristics of a target at different angles and directions. It reflects the influence of the physical structure and electromagnetic properties of the target on signal reflection.

[0076] The local radar cross section σ(θ,φ) represents the local radar cross section values of a target in different directions. It is a specific numerical representation of the radar scattering characteristics of the target. The local radar cross section is used to quantitatively describe the reflection intensity of a target in different directions. It is a key parameter in a radar system and is used to calculate the total radar cross section of a target.

[0077] Anisotropic scattering model |F(θ,φ)| 2 Describes the radar scattering characteristics of a target object in different directions. Its establishment is usually based on simulating or measuring the geometric shape, electromagnetic characteristics, and material parameters of the target. The following is the general process for establishing an anisotropic scattering model:

[0078] First, it is necessary to model the geometry of the target. This includes the size, shape, surface features of the target, as well as possible edges, corners, and concave and convex parts. Usually, the target shape can be represented by points, lines, and surfaces in a three-dimensional coordinate system. Next, it is necessary to establish an electromagnetic property model of the target. This includes electromagnetic parameters such as the conductivity, permittivity, and permeability of the target. These parameters determine the response of the target to incident electromagnetic waves. The electromagnetic property model can be determined based on the material composition and structure of the target. Using electromagnetic field theory and scattering theory, the scattering characteristics of the target in different directions can be calculated. This can be achieved by solving Maxwell's equations or applying the radar equation. Specific methods include: using numerical methods such as finite element analysis (FEM) or time-domain integral equation method (TDIE) to numerically simulate the physical process of the interaction between electromagnetic waves and the target. For large-sized targets, methods such as physical optics approximation (PO) or physical optics geometry (POGO) can be used for analysis. Using radar remote sensing data or experimental measurements to obtain the scattering characteristics of the target in different directions. Based on the calculation or measurement results, an anisotropic scattering model |F(θ,φ)| of the target can be established. 2 This model describes the reflection characteristics of the target in different directions, usually represented in polar coordinates, where θ represents the azimuth angle and φ represents the elevation angle. When establishing the anisotropic scattering model, it is usually necessary to parameterize the model so that it can be adjusted according to the specific target attributes. These parameters can include scattering intensity, phase, polar distribution, etc. After establishing the model, it is necessary to verify and adjust it to ensure that the model can accurately describe the scattering characteristics of the target. This can be done by comparing it with experimental data or other independent simulation methods.

[0079] The local radar cross section σ(θ,φ) describes the scattering cross section values of the target in different directions, which is usually a specific numerical representation of the reflection intensity of the target in a radar system. The process of calculating the local radar cross section involves analyzing and calculating the electromagnetic response of the target. The following is the general process of calculating the local radar cross section:

[0080] First, geometric modeling of the target is required. This includes the size, shape, surface features, and electromagnetic parameters of the target. Geometric modeling can represent the shape of the target using points, lines, and surfaces in a three-dimensional coordinate system. Determine the characteristics of the electromagnetic wave incident on the target, including frequency, wavelength, polarization direction, and incident angle (usually represented by azimuth angle θ and elevation angle φ). Using electromagnetic field theory, calculate the response of the target surface to the incident electromagnetic wave. This usually involves solving Maxwell's equations or applying the radar equation. The result of the calculation is the distribution of the reflected electromagnetic field at different points on the target surface. Integrate or simulate the reflected electromagnetic field on the target surface to calculate the scattered electromagnetic wave in different directions. This can be achieved through integral equations or numerical simulation methods. Based on the calculated scattered electromagnetic wave, calculate the local radar cross-section σ(θ,φ) at the azimuth angle θ and elevation angle φ. This usually involves comparing the intensity or power density of the reflected electromagnetic field with that of the incident electromagnetic wave. When calculating the local radar cross-section, the results usually need to be parameterized for adjustment according to specific target attributes. This can include normalization, unit conversion, or size correction, etc. After calculating the local radar cross-section, it needs to be verified and adjusted to ensure that the model can accurately describe the reflection characteristics of the target in different directions. This can be done by comparing with experimental data or other independent simulation methods.

[0081] Embodiment 5: The received signal model constructed by the received signal model construction unit is represented by the following formula:

[0082]

[0083] where r(t,r) represents the received signal; ∫ 目标区域 is an integral symbol, representing the integration over the entire region where the target is located. It is used to integrate various reflected signals within the region where the target is located to calculate the received signal; dΩ is the solid angle element, used to integrate the reflected signals in different directions; j is the imaginary symbol.

[0084] Specifically, the received signal r(t,r) is the signal received at time t and position r. It represents the variation of the signal received by the radar system in the time domain and space. The integral symbol ∫ 目标区域 represents an integration operation, integrating a certain parameter within the target region. In this formula, it is used to integrate various reflected signals within the target region to calculate the total received signal. The propagation model p(t,r) represents the complex amplitude of the modulated signal at time t and position r. It describes the propagation mode of the signal, considering the variations in time and space. is a complex exponential term, where Let \(f\) represent the operating frequency of the radar, \(R\) represent the distance from the radar emission point to the target or reflection point, and \(c\) represent the speed of light. This term takes into account the phase change of the signal with distance and frequency, and is usually used to represent the phase change of the signal during propagation.

[0085] There may be multiple reflecting objects in the target area, and each object will reflect the signal. Through integration, we consider the contributions of all these reflected signals to calculate the final received signal. The solid angle element \(d\Omega\) is used to integrate the reflected signals in different directions. It is actually a small solid angle element in the spherical coordinate system, used to determine the weight of the reflected signals in each direction.

[0086] Embodiment 6: The process of the analysis unit processing the received signal through the matched filtering technique, beamforming technique, and adaptive signal processing algorithm is represented by the following formula:

[0087]

[0088] where \(s\) MF (t) represents the received signal processed by the beamforming technique and the adaptive signal processing algorithm; is an integral symbol, representing the integration over the entire time domain; \(s(t')\) represents the received signal at time \(t'\); \(Z(t - t')\) is the matched filter; \(w(t - t')\) is the adaptive weight function, represented by the following formula:

[0089] w(t - t') = w0·e jφ(t-t′) ;

[0090] where \(w0\) is the adaptive amplitude, which is a set value; \(\varphi(t - t')\) represents the phase at time \(t - t'\).

[0091] Specifically, \(s\) MF (t) represents the received signal \(s\) MF (t) processed by the beamforming technique and the adaptive signal processing algorithm. It is the result of the input signal \(s(t')\) processed by the matched filter and the adaptive weight function \(w(t - t')\). \(s\) MF (t) represents the result of signal processing, which includes the effects of signal enhancement and noise suppression. Denotes the integration over the entire time domain, accumulating the signal at all time points. Through integration, signal processing can be applied to the entire time domain, ensuring that each moment is taken into account. s(t') represents the received signal at time t'. It is the originally received signal. The input signal s(t') contains target information and noise and is the starting point of signal processing. Z(t - t') is the matched filter, used to filter the input signal according to the desired signal characteristics. The matched filter is used to enhance the components that match the desired signal characteristics, helping to detect the target signal and reduce interference. w(t - t') is the adaptive weight function, which weights the signals at different time points according to the signal characteristics and interference conditions. The adaptive weight function is used to optimize signal processing to adapt to the dynamic changes of the signal and interference, improving the signal quality. w0 is the adaptive amplitude, usually a preset value, used to control the amplitude gain of the signal. φ(t - t') represents the phase at time t - t', used to adjust the phase of the signal to match the desired phase characteristics.

[0092] This formula describes the signal processing process. First, the components that match the desired signal characteristics are enhanced through the matched filter Z(t - t'), and then the signals at different time points are weighted by the adaptive weight function w(t - t'). The final result is the received signal s MF (t), which is applicable to radar applications such as target detection, tracking, and recognition. This processing process helps to improve the signal quality and system performance.

[0093] Example 7: The detection threshold set by the analysis unit is expressed using the following formula:

[0094]

[0095] where μ is the mean of the received signal; σ is the standard deviation of the received signal; ɑ is the adaptive threshold coefficient, which is a set value with a value range of 1 to 3.

[0096] Specifically, the principle of this threshold setting is based on the statistical characteristics of the signal, especially the mean and standard deviation. By taking the mean and standard deviation into account, the position of the detection threshold can be determined according to the volatility of the signal. The adaptive threshold coefficient α allows adjusting the sensitivity of the threshold according to the application requirements. If α is smaller, the threshold will be more strict, and a stronger signal is required to be detected, which can reduce false alarms. If α is larger, the threshold will be more relaxed, and weaker signals can also be detected, but it may increase the risk of false alarms.

[0097] Example 8: The target motion estimation unit uses the following formula to calculate the time-domain differential signal from the received signals of multiple receiving channels:

[0098] di r(t) = r i r(t, r) - r i r(t - Δt, r)

[0099] where r i r(t, r) is the received signal of the i-th channel. We calculate the time-domain differential signal; d i d(t) is the time-domain differential signal; Δt is the time interval; The time-domain differential signals of multiple channels are combined into a vector d(t) = [d1(t), d2(t), …, d m (t)], where m is the number of channels; Through multi-channel pulse-Doppler processing, the velocity vector v(t) and the direction vector u(t) are estimated.

[0100] Specifically, the calculation of the time-domain differential signal d i (t) is based on the following principle: The difference between signals collected at different time points can be used to represent the movement of the target. By calculating the difference between the received signals at two moments t and t - Δt, the velocity information of the target can be captured. This is because the movement of the target will cause changes in the phase and amplitude of the signal, resulting in measurable changes in the differential signal. Multi-channel pulse-Doppler processing is a signal processing technique commonly used in radar systems for estimating the velocity and direction of the target. It is based on the radar pulse-Doppler principle, which uses the spectral characteristics of the signal to analyze the velocity information of the target. By combining the time-domain differential signals of multiple channels into a vector d(t), the information of multiple channels can be considered simultaneously to improve the accuracy of velocity estimation. Multi-channel processing allows the estimation of the velocity vector v(t) and the direction vector u(t) of the target by analyzing the phase and frequency information of different channels.

[0101] By calculating the time-domain differential signal d i(t), Embodiment 8 can capture the speed information of the target. The movement of the target causes changes in the received signal over time, and these changes can be quantified through time-domain differentiation. Therefore, an important function of Embodiment 8 is to estimate the speed of the target, which is crucial for tracking and locating the target. In addition to speed, Embodiment 8 can also be used to estimate the direction of the target. The multi-channel pulse-Doppler processing technique allows for the analysis of the signal phase and frequency information of different channels, thereby obtaining the direction vector of the target. This is very useful for determining the direction and position of the target relative to the radar. By combining the differential signals of multiple channels, Embodiment 8 can improve the resolution of target detection. Multi-channel processing allows for more precise differentiation between target signals and spurious signals, thereby reducing false alarms and improving the reliability of target detection. Multi-channel processing also helps to suppress multipath interference, which is caused by the reflection and propagation of signals along multiple paths. By analyzing the signals of multiple channels, the system can better distinguish the main target signal and the multipath reflection signal, thereby reducing the impact of multipath interference. By providing the speed and direction information of the target, Embodiment 8 enhances the ability of target tracking and positioning. The radar system can use these estimated values to predict the future position and trajectory of the target, achieving more precise target tracking and positioning.

[0102] The principle of multi-channel pulse-Doppler processing is based on the fact that the signals received by a multi-channel radar contain multiple echoes of the target, and these echoes correspond to different target motion speeds and directions. By analyzing the spectral information of these echoes, the speed vector and direction vector of the target can be estimated. First, the radar system transmits a pulse signal and receives the signal reflected by the target. These received signals contain target echoes from different distances, and the frequencies of these echoes are affected by the target motion. Using the pulse-Doppler processing technique, these echo signals can be converted into spectral information to obtain the speed information of the target. The multi-channel radar system simultaneously receives signals from multiple channels. Each channel may be located at a different position or have different receiving characteristics, so the received signals may vary slightly. For the signals received by each channel, pulse-Doppler processing is performed to convert them into spectral information. Since different channels may receive different target echoes, the spectral information will be different. The spectral information of different channels is combined together to obtain a comprehensive spectrogram. This comprehensive spectrogram contains information from different channels and can be used to estimate the speed and direction of the target. By analyzing the comprehensive spectrogram, the target speeds corresponding to different frequency components can be determined. Different frequency components represent targets with different speeds, so the speed vector of the target can be estimated. By analyzing the direction information of the comprehensive spectrogram, the motion direction of the target can be determined. The motion of the target in different directions will be manifested as different angles in the spectrogram.

[0103] Embodiment 9: The method for the target motion estimation unit to estimate the target motion process using a Kalman filter includes: decomposing the direction vector into direction components in three directions, namely x(t), y(t), and z(t); decomposing the velocity vector into velocity components in three directions, namely and Thus, the state vector of the target is represented as Using the following formula, the state prediction is performed using a Kalman filter:

[0104]

[0105] Using the following formula, the state update is performed using a Kalman filter:

[0106]

[0107] where P pred (t) is the state prediction covariance matrix; K(t) is the Kalman gain; z(t) is the observation residual; x pred (t) is the state prediction vector; F(t) is the state transition matrix, which is represented using the following formula:

[0108]

[0109] P(t) is the state covariance matrix, which is represented using the following formula

[0110]

[0111] H(t) is the observation matrix, which is represented using the following formula:

[0112]

[0113] Specifically, state prediction is one of the key steps of the Kalman filter and is used to estimate the expected value of the target state at the next moment. This step mainly involves the update of the state transition matrix and the state covariance matrix. The first part of state prediction involves the state transition matrix F(t). The state transition matrix describes the dynamic evolution of the system between two consecutive moments. Usually, F(t) is a linear matrix that maps the current state x(t - Δt) to the state prediction x pred (t) at the next moment t. The state covariance matrix P pred(t) represents the uncertainty of state prediction. It describes the accuracy of the estimation of the target state in the absence of new observation information. The Kalman filter estimates the uncertainty of the state estimation at the next moment by predicting the state covariance matrix. The main role of state prediction is to estimate the target state at the next moment through the system dynamic model. This is a prediction based on prior information without considering new observations. The result of state prediction will be used in the subsequent state update step to fully utilize the observation information to correct the state estimation and reduce the estimation uncertainty.

[0114] The observation matrix H(t) describes the mapping relationship between the observations and the state. It defines how to transform from the state space to the observation space. Usually, H(t) is a linear matrix. The observation noise is usually a random variable with known statistical properties. The Kalman filter assumes that the observation noise is zero-mean, independent and identically distributed Gaussian noise, and describes its intensity and correlation through the covariance matrix of the observation noise. The observation residual z(t) is the difference between the observation value and the state prediction value. It represents the measurement error between the actual observation value and the predicted value. The Kalman gain K(t) is a key parameter used to balance the weights of the predicted value and the observation value. It is calculated through the state covariance matrix P pred (t), the observation matrix H(t), and the covariance matrix of the observation noise. The calculation of the Kalman gain can minimize the mean square error of the state estimation.

[0115] The main role of Example 9 is to use the Kalman filter to track the movement of the target. The Kalman filter can estimate the position and velocity of the target according to the prediction model (state transition matrix) and the observation model (observation matrix), so as to track the movement trajectory of the target in real time. Through the state update step, the Kalman filter corrects the previous state estimation according to the actual observation value. This means that if new observation values are available, the Kalman filter will use this information to more accurately estimate the state of the target. This is a key advantage in real-time systems because it allows the filter to provide accurate target position and velocity while continuously updating. Through state prediction and update, the Kalman filter can not only provide the state estimation of the target, but also provide a measure of the uncertainty of the state estimation, that is, the state covariance matrix. Through state update, the filter can minimize the uncertainty to provide the optimal state estimation. Example 9 also includes the calculation of the Kalman gain, and the value of the Kalman gain is determined according to the uncertainty of the state estimation and the intensity of the observation noise. This makes the Kalman filter adaptable and can adjust the weight of the state estimation according to different environments and observation conditions. The Kalman filter in this example is multi-dimensional and can simultaneously estimate the position (three directions) and velocity (three directions) of the target. This makes it applicable to various multi-dimensional motion tracking tasks, such as aircraft navigation, vehicle autonomous driving, etc.

[0116] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An electronic fence based on ultra-wideband radar technology, characterized in that, Including: A signal transmitting unit, a signal propagation model building unit, a target object model building unit, a received signal model building unit, an analysis unit, and a target motion estimation unit; The signal transmitting unit is configured to generate an ultra-wideband signal, modulate the ultra-wideband signal through a pulse modulation function to obtain a modulated signal, and then transmit the modulated signal; The signal model building unit is configured to build a propagation model of the modulated signal in space to describe the process of the modulated signal propagating from the radar emission point to the target and then returning; The target object model building unit is configured to build an anisotropic scattering model of the target according to the geometric shape, electromagnetic characteristics, and material parameters of the target, so as to calculate the radar cross section of the target; The received signal model building unit is configured to build a received signal model according to the radar cross section of the target to describe the process of the received signal returning from the target to the radar receiver; The analysis unit is configured to process the received signal using a matched filtering technique to enhance the characteristics of the received signal, apply a beamforming technique, sum and superimpose the received signals of multiple receiving channels through weighting to improve the target resolution of the received signal, adopt an adaptive signal processing algorithm to suppress multipath interference and noise, and at the same time enhance the detection ability of the target in the received signal, and set a detection threshold to determine whether the target exists in the received signal; The target motion estimation unit is configured to calculate a time-domain differential signal from the received signals of multiple receiving channels, combine the time-domain differential signals into a vector, estimate the speed and direction of the target based on multi-channel adaptive beamforming, and use a Kalman filter to estimate the target motion process.

2. The electronic fence based on ultra-wideband radar technology according to claim 1, characterized in that, The modulated signal transmitted by the signal transmitting unit is represented by the following formula: s(t) = A·cos(2πf c t)·g(t)·p(t); where s(t) is the modulation signal; A·cos(2πf c t) represents the ultra-wideband signal; A is the amplitude of the ultra-wideband signal; f c is the center frequency of the ultra-wideband signal; g(t) is the pulse modulation function used to determine the time-domain characteristics of the signal; p(t) is the modulation pulse, expressed as: where N is the number of subcarriers in the modulation pulse; α i is the amplitude of each subcarrier; f i is the frequency of each subcarrier; φ i is the phase of each subcarrier.

3. The electronic fence based on ultra-wideband radar technology according to claim 2, characterized in that, The propagation model built by the signal model building unit is represented by the following formula: where N p is the order of polynomial expansion; φ n is the phase of multipath reflection, taking into account the interference of different paths; p(t,r) represents the complex amplitude of the modulated signal at time t and position r; 4πR is the propagation path length of the modulated signal, where R represents the distance from the radar emission point to the target or reflection point; represents the coefficients of the polynomial expansion, where k is the wave number, R is the distance, and n is the order number of the multipath is the phase term of the signal, c is the speed of light, and φ n represents the phase of the multipath reflection.

4. The electronic fence based on ultra-wideband radar technology according to claim 2, characterized in that, The anisotropic scattering model built by the target object model building unit is represented by the following formula: Where: RCS(θ, φ) is the radar cross section of the target object, representing the reflection intensity of the target to the modulated signal. It is a function related to the incident angle and azimuth angle, and is used to describe the reflection characteristics of the target in different directions; |F(θ, φ)| 2 represents the anisotropic scattering model of the target, which is obtained by modeling according to the geometric shape, electromagnetic characteristics and material parameters of the target; σ(θ, φ) is the local radar cross section of the target, which represents the cross section values of the target in different directions; θ is the direction angle of the target relative to the radar antenna. When θ is 0°, it means the target is directly in front of the radar antenna. Positive values represent angles in the clockwise direction, and negative values represent angles in the counterclockwise direction; φ is the phase of the modulated signal.

5. The electronic fence based on the ultra-wideband radar technology according to claim 4, characterized in that, The received signal model built by the received signal model building unit is represented by the following formula: where r(t,r) represents the received signal; ∫ 目标区域 is an integral symbol, indicating integration over the entire region where the target is located. It is used to integrate various reflected signals within the region where the target is located to calculate the received signal; dΩ is the solid angle element, used to integrate the reflected signals in different directions; j is the imaginary symbol.

6. The electronic fence based on ultra-wideband radar technology according to claim 5, characterized in that, The process of the analysis unit processing the received signal through a matched filtering technique, a beamforming technique, and an adaptive signal processing algorithm is represented by the following formula: where s MF (t) represents the received signal processed by beamforming technology and adaptive signal processing algorithms; is an integral symbol, representing the integration over the entire time domain; s(t') represents the received signal at time t'; Z(t - t') is the matched filter; w(t - t') is the adaptive weight function, which is expressed using the following formula: w(t - t') = w0·e jφ(t-t′) ; Where, w0 is the adaptive amplitude, which is a set value; φ(t - t') represents the phase at time t - t'.

7. The electronic fence based on ultra-wideband radar technology according to claim 6, characterized in that, The detection threshold set by the analysis unit is represented by the following formula: Where, μ is the mean value of the received signal; σ is the standard deviation of the received signal; α is the adaptive threshold coefficient, which is a set value, and the value range is from 1 to 3.

8. The electronic fence based on the ultra-wideband radar technology according to claim 7, characterized in that, The target motion estimation unit calculates the time-domain differential signal from the received signals of multiple receiving channels using the following formula: d i r(t) = r i r(t, r) - r i r(t - Δt, r) where r i (t, r) is the received signal of the i-th channel. We calculate the time-domain differential signal; d i (t) is the time-domain differential signal; Δt is the time interval; The time-domain differential signals of multiple channels are combined into a vector d(t) = [d1(t), d2(t), …, d m (t)], where m is the number of channels; Through multi-channel pulse-Doppler processing, the velocity vector v(t) and the direction vector u(t) are estimated.

9. The electronic fence based on ultra-wideband radar technology according to claim 8, wherein The target motion estimation unit, and the method for estimating the target motion process using a Kalman filter includes: decomposing the direction vector into direction components in three directions, namely x(t), y(t), and z(t); decomposing the velocity vector into velocity components in three directions, namely and Thereby obtaining the state vector representation of the target as Through the following formula, use the Kalman filter for state prediction: The state is updated using a Kalman filter through the following formula: Among them, P pred (t) is the state prediction covariance matrix; K(t) is the Kalman gain; z(t) is the observation residual; x pred (t) is the state prediction vector; F(t) is the state transition matrix, which is expressed by the following formula: P(t) is the state covariance matrix, which is represented by the following formula H(t) is the observation matrix, which is represented by the following formula:

Images