A very high frequency acquisition system and method based on compressed sensing

Abstract

This invention discloses a UHF acquisition system and method based on compressed sensing. The system includes: a radio frequency front-end module configured to receive and condition UHF analog signals from a UHF sensor; a compressed sensing acquisition module configured to perform random sub-Nyquist sampling and linear projection on the conditioned signal in the analog domain to obtain low-dimensional compressed measurement values; and a signal reconstruction and processing module configured to reconstruct the original UHF signal based on the compressed measurement values ​​to estimate or extract signal features. This invention solves the problems of high cost, high power consumption, and large data volume caused by the high sampling rate requirements of traditional UHF acquisition systems, reducing the constraints on devices such as ADCs while ensuring signal characteristics.

Description

Technical Field

[0001] This invention belongs to the field of substation primary equipment condition monitoring technology, and relates to a system for acquiring ultra-high frequency signals, particularly an ultra-high frequency acquisition system and method based on compressed sensing theory that can sample at a rate much lower than the Nyquist rate. Background Technology

[0002] Ultra-high frequency (UHF) signal acquisition is crucial in fields such as partial discharge detection, wireless communication, and radio frequency identification (RFID). However, traditional acquisition systems based on the Nyquist sampling theorem face significant challenges. UHF signals typically range from 300MHz to 1.5GHz. To acquire signals in the GHz band without distortion, a high-speed ADC with a speed of at least 3GS / s is required. These devices are expensive, consume a lot of power, and generate massive amounts of data, placing enormous pressure on storage and transmission. This is particularly problematic in applications requiring the deployment of numerous acquisition points or long-term online monitoring (such as partial discharge monitoring in smart substations using GIS). Currently, frequency down-detection is the primary method, but this results in the loss of the original signal characteristics.

[0003] Compressed sensing theory states that if a signal is sparse or compressible in a certain transform domain, it can be acquired at a sampling rate much lower than the Nyquist rate, and then the original signal can be reconstructed with high probability from a small number of samples using a nonlinear optimization algorithm. Currently, compressed sensing technology is mostly used in image processing and post-processing algorithm research. How to innovatively apply it to real-time acquisition hardware for high-frequency analog signals such as UHF signals, solving the aforementioned engineering challenges while operating with low power consumption, is an important research direction.

[0004] Existing technologies already include solutions for applying compressed sensing to partial discharge signal acquisition. Chinese patent CN107561416A proposes an ultra-high frequency acquisition system based on compressed sensing. However, this document does not optimize the overall power consumption of the system for long-term online monitoring scenarios. Its acquisition unit is always active and continues to consume power even when no effective signal is detected. This is unacceptable for distributed sensing nodes that rely on battery power or require extremely low operating costs. Furthermore, no specific hardware design scheme is provided.

[0005] Chinese patent CN112508089A proposes an adaptive compressed sensing method for partial discharge signal compressed transmission. This method addresses the issue of a large amount of useless redundant information during partial discharge signal acquisition, which leads to significant waste of communication costs. It adaptively identifies the presence of partial discharge pulse signals in signal blocks, applying a high compression rate to signal blocks without pulse signals and a low compression rate to those containing partial discharge signals, thus achieving compressed sensing technology for partial discharge signals. This technology primarily addresses the problem of large data volumes in data transmission or storage, but does not mention reducing the sampling rate for sampling. Summary of the Invention

[0006] The purpose of this invention is to provide an ultra-high frequency (UHF) acquisition system and method based on compressed sensing, which can solve the problems of high cost, high power consumption, and large data volume caused by the high sampling rate requirements of traditional UHF acquisition systems. Under the premise of ensuring the integrity of signal characteristics, the sampling rate and the overall power consumption of the system are reduced by orders of magnitude.

[0007] To achieve the above objectives, the solution of the present invention is:

[0008] A UHF acquisition system based on compressed sensing includes,

[0009] The radio frequency front-end module is configured to receive and condition ultra-high frequency analog signals from ultra-high frequency sensors;

[0010] The event detection circuit is configured to acquire the signal energy of the conditioned signal and generate a trigger signal or a sleep signal based on a preset threshold.

[0011] The clock and control logic unit is configured to generate random binary sequences of values ​​+1 and -1 according to the trigger signal, the sequence rate being the sub-Nyquist sampling rate, and to control the system to enter a low-power state according to the sleep signal.

[0012] The compressed sensing acquisition module is configured to perform random sub-Nyquist sampling and linear projection on the conditioned signal based on the random binary sequence in the analog domain to obtain low-dimensional compressed measurements; and,

[0013] The signal reconstruction and processing module is configured to reconstruct the original UHF signal based on the compressed measurement values ​​or extract signal features.

[0014] The compressed sensing acquisition module includes,

[0015] An analog multiplier is configured to multiply a conditioned signal with a random binary sequence to obtain a product signal;

[0016] An integrator is configured to integrate the product signal over a fixed time window to obtain an integral result; and,

[0017] An analog-to-digital converter is configured to sample the integration result to obtain a compressed measurement value.

[0018] The radio frequency front-end module includes,

[0019] A tunable bandpass filter is configured to acquire signals in the target frequency band from UHF analog signals acquired by a UHF sensor; and,

[0020] A low-noise amplifier is configured to amplify the acquired signal.

[0021] The signal reconstruction and processing module works as follows:

[0022] Based on compression measurements , constitute measurement vector ;

[0023] For the random binary sequence Discretization yields the measurement matrix And select sparse basis matrices The optimization problem is solved using the orthogonal matching pursuit algorithm. This yields the sparse coefficient vector. ;

[0024] according to The original UHF signal was estimated by reconstruction. .

[0025] The signal reconstruction and processing module extracts signal features, including one or more of the following: pulse amplitude, phase, rise time, frequency domain energy distribution, and repetition rate.

[0026] The compressed measurement values ​​output by the compressed sensing acquisition module are transmitted to the signal reconstruction and processing module via wired or wireless communication.

[0027] A UHF acquisition method based on compressed sensing includes,

[0028] Conditioning of the UHF analog signals acquired by the UHF sensor;

[0029] The signal energy of the conditioned signal is acquired, and a trigger signal or a sleep signal is generated based on a preset threshold.

[0030] Based on the trigger signal, the conditioned signal is subjected to random sub-Nyquist sampling and linear projection in the analog domain based on a random binary sequence to obtain a low-dimensional compressed measurement value; wherein, the random binary sequence is a binary sequence with values ​​of +1 and -1, and its sequence rate is the sub-Nyquist sampling rate;

[0031] Based on the compressed measurement values, the original UHF signal is reconstructed and estimated or signal features are extracted.

[0032] In the analog domain, the conditioned signal is subjected to random sub-Nyquist sampling and linear projection based on a random binary sequence to obtain low-dimensional compressed measurements, including...

[0033] The conditioned signal is multiplied by a random binary sequence to obtain the product signal;

[0034] The product signal is integrated within a fixed time window to obtain the integration result;

[0035] The integral result is sampled to obtain a compressed measurement value.

[0036] This includes conditioning the UHF analog signals acquired by the UHF sensor, including...

[0037] Acquire the target frequency band signal from the UHF analog signal collected by the UHF sensor;

[0038] The acquired signal is amplified.

[0039] The original UHF signal estimate is reconstructed based on the compressed measurement values, including:

[0040] Based on compression measurements , constitute measurement vector ;

[0041] For the random binary sequence Discretization yields the measurement matrix And select sparse basis matrices The optimization problem is solved using the orthogonal matching pursuit algorithm. This yields the sparse coefficient vector. ;

[0042] according to The original UHF signal was estimated by reconstruction. .

[0043] The process involves extracting signal features, including one or more of the following: pulse amplitude, phase, rise time, frequency domain energy distribution, and repetition rate.

[0044] By adopting the above scheme, the most significant advantage of this invention compared to existing technologies lies in the synergistic innovation of "event detection and trigger control" and "analog domain compressed sensing sampling," simultaneously achieving a leap in system-level power consumption and breaking through hardware sampling rate constraints. This system directly performs data compression at the analog domain front end, greatly alleviating the pressure on data transmission and storage. Simultaneously, its low-power characteristics enable large-scale distributed deployment and long-term online monitoring. Ultimately, this invention achieves a paradigm shift from "brute-force acquisition" relying on expensive high-speed hardware to "precise sensing" relying on intelligent algorithms, laying the foundation for the widespread application of UHF technology.

[0045] The core of this invention lies in using control logic to generate a random basis sequence to control an analog switch. This allows for random sub-Nyquist sampling and projection calculation of the UHF signal conditioned by the RF front-end in the analog domain, directly outputting low-dimensional compressed measurement values. This eliminates the need for an ultra-high-speed ADC to capture signal information across the entire frequency band. The receiver employs a reconstruction algorithm based on orthogonal matched pursuit to reconstruct the original sparse signal from the compressed measurement values ​​with high probability. This invention significantly reduces the sampling rate, data volume, and system power consumption, making it particularly suitable for long-term online monitoring and diagnosis of transient UHF signals generated by transformers, GIS partial discharge, cable faults, etc. Attached Figure Description

[0046] Figure 1 This is an overall structural block diagram of the system of the present invention;

[0047] Figure 2 This is a schematic diagram of the system connection in an embodiment of the present invention;

[0048] Figure 3 This is a flowchart of the method in an embodiment of the present invention;

[0049] Figure 4 This is a schematic diagram illustrating the effect of an embodiment of the present invention. Detailed Implementation

[0050] like Figure 1 As shown, this invention provides an ultra-high frequency acquisition system based on compressed sensing, including a radio frequency front-end module, a compressed sensing acquisition module, a clock and control logic unit, and a signal reconstruction and processing module, which are described below.

[0051] The input terminal of the radio frequency front-end module is connected to an ultra-high frequency sensor to receive raw ultra-high frequency analog signals. Internally, it includes, in sequence:

[0052] A) Tunable bandpass filter: used to filter out signals in the target frequency band and suppress out-of-band noise interference; its center frequency and bandwidth can be configured by external commands to adapt to different application scenarios.

[0053] B) Low-noise amplifier: Used to amplify the filtered weak signal and improve the signal-to-noise ratio. The output of the RF front-end module outputs a conditioned analog signal. And connect it to the input end of the compressed sensing acquisition module.

[0054] The compressed sensing acquisition module is the core of the system, used to directly perform compressed sampling of signals in the analog domain. Internally, it includes:

[0055] A) Analog multiplier: One of its input terminals is connected to the output terminal of the RF front-end module to receive signals. The other input is connected to the clock and control logic unit to receive a random binary sequence. The analog multiplier is used to implement signal... With random sequences The multiplication operation outputs the product signal. .

[0056] B) Integrator: Its input is connected to the output of the analog multiplier, and it is used to perform product analysis on the product signal. In a fixed time window Integral operations are performed within the system.

[0057] C) Analog-to-digital converter: Its input is connected to the output of the integrator, and it is used to sample and quantize the output voltage value of the integrator at the end of each integration time window, and output a digital compressed measurement value. .

[0058] The clock and control logic unit provides timing control for the system and includes the following components:

[0059] A) Clock source: generates the system's master clock.

[0060] B) Pseudo-random binary sequence generator: driven by the master clock, it generates random sequences at a rate of f_cs. The sequence values ​​are +1 and -1. This rate f_cs is much lower than the Nyquist sampling rate of the ultra-high frequency signal, i.e., it is a sub-Nyquist sampling rate.

[0061] C) Control logic: Used to generate the integrator's reset signal and the ADC's start sampling signal, ensuring that the integration, sampling, and reset operations are strictly synchronized with the random sequence.

[0062] The signal reconstruction and processing module receives a series of compressed measurement values ​​output by the compressed sensing acquisition module. , constitute measurement vector This module is configured to perform one or all of the following operations:

[0063] A) Signal reconstruction: using the known measurement matrix (from random sequence) Discretization and pre-selected sparse basis matrices (e.g., discrete cosine transform matrix or wavelet transform matrix), the optimization problem is solved using the orthogonal matching pursuit algorithm. This yields the sparse coefficient vector. Ultimately passed Reconstructing the original signal estimate at a high sampling rate .

[0064] B) Feature extraction: Based on the compressed measurement vector Y or the reconstructed signal estimation The characteristic parameters of the signal are extracted, including one or more of the following: pulse amplitude, phase, rise time, frequency domain energy distribution, and repetition rate, for subsequent equipment status diagnosis.

[0065] The compressed measurement value output by the compressed sensing acquisition module is transmitted to the remote signal reconstruction and processing module via wired or wireless communication.

[0066] Specifically, the present invention may also include an event detection circuit, the input of which is connected to the output of the RF front-end module, and the output of which is connected to the clock and control logic unit. The event detection circuit is configured to continuously (or periodically wake up) monitor the energy or envelope of the analog signal x(t) with extremely low power consumption. This circuit typically includes an envelope detector and a threshold comparator. When the detected signal energy exceeds a preset threshold, it determines that a valid event (such as a partial discharge pulse) has been detected and immediately outputs a trigger signal to activate the clock and control logic unit and the compressed sensing acquisition module (triggered by the clock and control logic unit). When the signal energy is below the preset threshold, it determines that there is no valid event and outputs a sleep signal. This circuit is crucial for achieving ultra-low standby power consumption in this system, and its own power consumption is far lower than that of the subsequent signal processing chain.

[0067] The system's data flow and operating principle are as follows: The analog signal captured by the UHF sensor is filtered and amplified by the RF front-end module before being sent to the compressed sensing acquisition module. Simultaneously, a low-speed random sequence generated by the clock and control logic unit is also input to this module. In the analog multiplier, the analog signal is multiplied by the random sequence, and the result is integrated within a fixed time window in the integrator. This integrated result is sampled by a low-speed ADC to obtain a compressed measurement value. This process is repeated M times to obtain an M-dimensional compressed measurement vector Y. After this vector is uploaded through the communication interface, the signal reconstruction and processing module reconstructs the original signal using a nonlinear optimization algorithm or directly performs feature analysis.

[0068] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0069] This embodiment aims to address the industry pain points in transformer internal partial discharge monitoring, such as the difficulty and high cost of wiring in traditional wired UHF monitoring systems, and the high power consumption and inability of wireless sensing nodes to operate online for extended periods due to high-speed sampling.

[0070] Implementation environment: A 500kV transformer, which has one or more built-in UHF sensors embedded in its body.

[0071] 1. System Composition and Deployment

[0072] like Figure 2 As shown, this embodiment mainly includes a built-in sensing unit, a compressed acquisition unit, and a monitoring platform.

[0073] 1) Built-in sensing unit: It is a UHF sensor pre-installed by the transformer factory. Its effective frequency band is 300MHz~1500MHz. The sensed partial discharge pulse electromagnetic wave signal is led out to the sealed joint of the tank wall through a coaxial cable.

[0074] 2) Compressed Acquisition Unit: An explosion-proof, moisture-proof, low-power sealed metal housing, directly mounted on the outside of the transformer tank wall, near the sensor lead-out connector. Internally integrated are:

[0075] a) RF front-end: Utilizing an integrated tunable bandpass filter and low-noise amplifier module from Mini-Circuits, model ZAB-2000A+. The center frequency is set to 800MHz, the bandwidth to 500MHz, and the gain to 30dB. This configuration effectively suppresses low-frequency periodic interference generated by transformer operation.

[0076] b) Compressed acquisition: The analog multiplier AD834 and integrator circuit are used, and the ADC is TI's AD9268 (16-bit 100MS / s).

[0077] c) Control and processing unit: A Xilinx Artix-7 FPGA is used to implement pseudo-random sequence generation, timing control and data buffering.

[0078] d) Power supply and communication: DC24V power supply is used, and communication uses a custom UDP network module.

[0079] 3) Monitoring Platform

[0080] The monitoring platform uses embedded hardware, employing an FPGA to directly receive UDP network data and reconstruct the data.

[0081] 2. Work Process

[0082] 1) Signal Sensing and Conditioning: Partial discharge occurs inside the transformer, and the UHF sensor couples to the transient electromagnetic wave signal, outputting a weak analog signal. This signal is transmitted to the compressed acquisition node via a coaxial cable, where it is first filtered (to remove power frequency interference) and amplified by the radio frequency front end.

[0083] 2) Compressed sampling:

[0084] The FPGA generates a random sequence φ(t) at a rate of 50 MS / s.

[0085] The conditioned signal x(t) is mixed with φ(t) in an analog multiplier.

[0086] The mixing result is fed into the integrator, whose integration time window is set to 20ns (i.e. 1 / 50MHz).

[0087] At the end of each 20ns window, the integrator output voltage is sampled using an AD9268 chip to obtain a 16-bit compressed measurement value y[m].

[0088] This process is continuous, with the FPGA continuously collecting M=1000 measurement values ​​to form a compressed data packet Y.

[0089] 3) Data transmission: The FPGA sends the assembled data packets to the host computer via a custom UDP protocol. Compared to the massive amounts of data generated by traditional methods requiring a GS / s sampling rate, the transmission requirements are reduced by several orders of magnitude.

[0090] 4) Signal reconstruction and diagnosis:

[0091] After receiving data packet Y, the monitoring platform reconstructs the measurement matrix Φ based on the known random sequence. Then, it invokes the deployed OMP reconstruction algorithm, using Discrete Cosine Transform (DCT) as the sparse basis, to solve for... Finally, a time-domain pulse waveform with an equivalent sampling rate of 2.5 GS / s was reconstructed. .

[0092] 5) Intelligent Analysis:

[0093] The reconstructed waveform The data is input into a pre-trained deep learning model, which automatically determines the discharge type (such as levitation discharge or surface discharge) and calculates characteristic parameters such as discharge amplitude and frequency.

[0094] The system automatically generates diagnostic reports and sends alarm information to maintenance personnel when a serious discharge is detected.

[0095] 3. Key parameters and beneficial effects

[0096] The cost of the core ADC has dropped from several thousand yuan (GS / sADC) to the level of hundreds of yuan (MS / sADC), reducing the cost per point by about 70%, making it possible to deploy multiple monitoring nodes inside a single transformer, thereby enabling power supply location.

[0097] This embodiment not only reconstructs the waveform, but more importantly, it directly uses compressed data or reconstructed waveforms for intelligent diagnosis, transforming the data into state information that can be directly used for decision-making, greatly enhancing the practical value of the monitoring system.

[0098] Thus, this invention provides a complete, specific, and implementable solution to the core problem in ultra-high frequency monitoring of partial discharge in transformers. Figure 4 The effects of embodiments of the present invention are shown.

[0099] like Figure 3 The diagram shown is a flowchart of a UHF acquisition method based on compressed sensing according to the present invention, which includes the following steps:

[0100] S1, When a partial discharge occurs inside the transformer, the UHF sensor outputs the original analog signal;

[0101] S2, bandpass filtering and gain conditioning are performed on the original analog signal;

[0102] S3, mixes the conditioned signal with the random sequence;

[0103] S4, perform integration processing on the mixing result;

[0104] S5, perform ADC down-sampling on the integration result to obtain the compressed measurement value;

[0105] S6, a time-domain pulse waveform is reconstructed using the deployed OMP algorithm;

[0106] S7 inputs the reconstructed waveform signal into artificial intelligence for fault type identification.

[0107] This invention also provides another computer device, including a processor and a memory configured to store a computer program capable of running on the processor; wherein, when the processor is configured to run the computer program, it executes the method steps of the foregoing embodiments.

[0108] In practical applications, the aforementioned processor includes a Field-Programmable Gate Array (FPGA), and the processor can be a Central Processing Unit (CPU) or a Digital Signal Processor (DSP). It is understood that for different devices, the electronic devices used to implement the above-mentioned processor functions can also be other types, and this embodiment of the invention does not impose specific limitations.

[0109] The aforementioned memory can be volatile memory, such as random-access memory (RAM); or non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); or a combination of the above types of memory, and provides instructions and data to the processor.

[0110] In an exemplary embodiment, the present invention also provides a computer-readable storage medium for storing a computer program.

[0111] Optionally, the computer-readable storage medium can be applied to any of the methods in the embodiments of the present invention, and the computer program causes the computer to execute the corresponding processes implemented by the processor in the various methods of the embodiments of the present invention. For the sake of brevity, these will not be described in detail here.

[0112] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0113] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0114] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0115] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0116] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0117] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A compressive sensing based ultra-high frequency acquisition system, characterized in that: The method comprises the steps of: including, a radio frequency front-end module configured to interface and condition the ultra-high frequency analog signal received from the ultra-high frequency sensor; an event detection circuit configured to obtain the signal energy of the conditioned signal, and generate a trigger signal or a sleep signal based on a preset threshold; a clock and control logic unit configured to generate a random binary sequence with values of +1 and -1 according to the trigger signal, and the sequence rate is sub-Nyquist sampling rate, and control the system to enter a low-power state according to the sleep signal; a compressed sensing acquisition module configured to perform random sub-Nyquist sampling and linear projection on the conditioned signal based on the random binary sequence in the analog domain, and obtain low-dimensional compressed measurement values; and, a signal reconstruction and processing module configured to reconstruct the original ultra-high frequency signal estimate or extract signal features based on the compressed measurement values; wherein the compressed sensing acquisition module comprises, an analog multiplier configured to perform multiplication operation on the conditioned signal and the random binary sequence to obtain a product signal; an integrator configured to integrate the product signal in a fixed time window to obtain an integration result; and an analog-to-digital converter configured to sample the integration result to obtain the compressed measurement values; Based on the compressed measurements , the measurement vector ; to the random binary sequence discretization results in a measurement matrix and a sparse basis matrix is selected by solving the optimization problem with the orthogonal matching pursuit algorithm a sparse coefficient vector is obtained ; According to reconstructing the original UHF signal estimate .

2. The system of claim 1, wherein: wherein the working process of the signal reconstruction and processing module is, the radio frequency front-end module comprises, an adjustable band-pass filter configured to obtain the signal of the target frequency band in the ultra-high frequency analog signal collected by the ultra-high frequency sensor; and 3. The system of claim 1, wherein: a low-noise amplifier configured to amplify the obtained signal.

4. The system of claim 1, wherein: The signal reconstruction and processing module extracts signal features, including one or more of pulse amplitude, phase, rise time, frequency energy distribution, and repetition rate.

5. A method for ultra-high frequency acquisition based on compressed sensing, characterized in that: The compressed measurement values output by the compressed sensing acquisition module are transmitted to the signal reconstruction and processing module through wired or wireless communication. including, conditioning the ultra-high frequency analog signal collected by the ultra-high frequency sensor; obtaining the signal energy of the conditioned signal, and generating a trigger signal or a sleep signal based on a preset threshold; performing random sub-Nyquist sampling and linear projection on the conditioned signal based on a random binary sequence in the analog domain according to the trigger signal, and obtaining low-dimensional compressed measurement values; wherein the random binary sequence is a binary sequence with values of +1 and -1, and the sequence rate is sub-Nyquist sampling rate; reconstructing the original ultra-high frequency signal estimate or extracting signal features based on the compressed measurement values; wherein performing random sub-Nyquist sampling and linear projection on the conditioned signal based on a random binary sequence in the analog domain to obtain low-dimensional compressed measurement values comprises, performing multiplication operation on the conditioned signal and the random binary sequence to obtain a product signal; integrating the product signal in a fixed time window to obtain an integration result; sampling the integration result to obtain the compressed measurement values; Based on the compressed measurements , the measurement vector ; to the random binary sequence discretization results in a measurement matrix and a sparse basis matrix is selected by solving the optimization problem with the orthogonal matching pursuit algorithm a sparse coefficient vector is obtained ; According to reconstructing the original UHF signal estimate .

6. The method of claim 5, wherein: wherein reconstructing the original ultra-high frequency signal estimate based on the compressed measurement values comprises, conditioning the ultra-high frequency analog signal collected by the ultra-high frequency sensor comprises, obtaining the signal of the target frequency band in the ultra-high frequency analog signal collected by the ultra-high frequency sensor; amplifying the obtained signal.

7. The method of claim 5, wherein: Extracting signal features including one or more of pulse amplitude, phase, rise time, frequency content distribution, repetition rate.

Images