A radio frequency conversion assembly

Abstract

The application relates to the radio frequency technical field, in particular to a radio frequency conversion assembly, which comprises: a baseband processing unit adopting a millimeter wave SoC chip, which is used for generating and processing initial transmission signals of 79-81 GHz and initial reception signals of 79-81 GHz; a frequency conversion unit comprising an up-conversion subunit and a down-conversion subunit, and a local oscillator synthesis unit used for generating and providing local oscillator signals of a fixed frequency; a mode switching and processing system used for selectively conveying echo signals or target radio frequency signals from external non-cooperative illumination sources to the baseband processing unit after down-conversion processing in two working modes; the working modes include an active mode and a passive mode.

Description

Technical Field

[0001] This application relates to the field of radio frequency technology, and in particular to a radio frequency conversion component. Background Technology

[0002] Radio frequency conversion components are key components of modern electronic information systems (especially in the fields of radar, communication, sensing and electronic warfare), and their performance has a significant impact on the overall effectiveness of terminal equipment.

[0003] Existing traffic millimeter-wave radars are mainly based on the 76-81GHz frequency band and rely on automotive-grade millimeter-wave SoC chips to realize roadside radar functions.

[0004] However, the lack of readily available millimeter-wave system-on-a-chip (SoC) solutions in the 92-94GHz frequency band means that existing roadside radar systems cannot be directly migrated to this frequency band, facing technical bottlenecks in frequency band switching and system compatibility. Summary of the Invention

[0005] Therefore, it is necessary to provide a radio frequency conversion component capable of frequency conversion in the 92-94GHz band to address the aforementioned technical problems.

[0006] This application provides a radio frequency conversion component, which includes:

[0007] The baseband processing unit, using a millimeter-wave SoC chip, is used to generate and process the initial transmit signal and the initial receive signal in the 79-81GHz range.

[0008] The frequency conversion unit includes an up-conversion subunit and a down-conversion subunit. The up-conversion subunit is used to up-mix the initial transmit signal output by the baseband processing unit with a fixed-frequency local oscillator signal to output a 92-94GHz radio frequency transmit signal. The down-conversion subunit is used to down-mix the received 92-94GHz echo signal with the local oscillator signal to output an initial receive signal to the baseband processing unit.

[0009] The local oscillator synthesis unit is used to generate and provide a local oscillator signal at a fixed frequency;

[0010] The mode switching and processing system is used to selectively down-convert echo signals or target radio frequency signals from external non-cooperative irradiation sources and then transmit them to the baseband processing unit in two operating modes; the operating modes include active mode and passive mode.

[0011] In one embodiment, the mode switching and processing system includes:

[0012] The signal processing unit is used to receive target radio frequency signals through a signal input interface. The target radio frequency signals include direct path signals and reflected path signals. The signal processing unit is equipped with a down-conversion channel for down-converting the target radio frequency signals into intermediate frequency signals.

[0013] The switch has a first port connected to the output of the frequency conversion unit to receive the echo signal, a second port connected to the output of the signal processing unit to receive the intermediate frequency signal, and a common output connected to the baseband processing unit.

[0014] The control and power supply unit is used to control the selection state of the switching switch according to the working mode, and to control the power supply to the up-conversion subunit, down-conversion subunit and signal processing unit.

[0015] In active mode: The control and power supply unit controls the switching switch to select the first port, so that the echo signal is transmitted to the baseband processing unit through the switching switch, and supplies power to the up-conversion subunit and the down-conversion subunit.

[0016] In passive mode: The control and power supply unit controls the switching switch to select the second port, so that the intermediate frequency signal is transmitted to the baseband processing unit through the switching switch and power the signal processing unit.

[0017] In one embodiment, in passive mode, the RF conversion component further includes:

[0018] The waveform feature extraction unit, connected to the baseband processing unit, is used to receive the intermediate frequency signal from the switching switch and extract the signal waveform features of the non-cooperative irradiation source from the intermediate frequency signal through digital signal processing.

[0019] The coherent processing unit, connected to the waveform feature extraction unit, is used to dynamically reconstruct the local reference signal based on the extracted signal waveform features, and to perform coherent processing on the reflected path signal using the local reference signal to calculate the time delay difference and Doppler frequency shift of the reflected path signal relative to the direct path signal.

[0020] The information fusion unit, connected to the coherent processing unit, is used to fuse time delay difference and Doppler frequency shift information to output target detection results.

[0021] In one embodiment, the signal processing unit is a multi-channel parallel architecture, configured with multiple independent down-conversion channels. These multiple down-conversion channels can synchronously receive and down-convert signals from multiple non-cooperative irradiation sources of different frequency bands and different standards.

[0022] In one embodiment, the waveform feature extraction unit is further configured with a pattern recognition subunit, which is used to match the extracted signal waveform features with the pre-stored signal waveform feature databases of each non-cooperative irradiation source in order to identify the identity and signal type of the non-cooperative irradiation source.

[0023] In one embodiment, the radio frequency conversion component further includes a distributed synchronization unit connected to the waveform feature extraction unit and / or the coherent processing unit, for exchanging signal waveform features or target detection results among multiple radio frequency conversion components, and using a cross-correlation algorithm to keep the time and frequency synchronized among the radio frequency conversion components.

[0024] In one embodiment, the information fusion unit is further configured with a micro-Doppler analysis subunit, which is used to analyze the micro-Doppler effect in the reflection path signal, and reconstruct the structural feature micro-motion of the target or generate a three-dimensional image based on the micro-Doppler effect, and output this information as part of the target detection result.

[0025] In one embodiment, the radio frequency conversion assembly further includes a waveform prediction unit connected to the information fusion unit, for:

[0026] Based on the target detection results and historical signal waveform feature data, the motion trajectory and signal reflection characteristics of the detected target are predicted through a pre-built deep learning network.

[0027] Based on the motion trajectory and signal reflection characteristics, the transmitted waveform parameters are determined. These parameters are used by the baseband processing unit to adjust the modulation method, bandwidth, or transmission timing of subsequent transmitted signals.

[0028] In one embodiment, historical signal waveform feature data is extracted and stored by a waveform feature extraction unit after processing previously received non-cooperative irradiation source signals.

[0029] In one embodiment, the signal input interface includes an impedance matching network and an anti-interference filter for preprocessing the received target radio frequency signal.

[0030] The aforementioned radio frequency conversion component, by integrating a baseband processing unit and a frequency conversion unit based on a mature millimeter-wave SoC chip, and introducing a mode switching and processing system, effectively solves the technical bottleneck of the lack of readily available SoC chips in the 92-94GHz frequency band. This enables existing roadside radar systems to operate in this frequency band without changing the core processing architecture, while also having the ability to flexibly switch between active detection and passive listening modes, significantly improving the system's adaptability and versatility. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the various units in a radio frequency conversion component in one embodiment;

[0033] Figure 2 This is a chip-level implementation block diagram of a radio frequency conversion component in one embodiment;

[0034] Figure 3 This is a schematic diagram of the module cross-section and signal flow of a radio frequency conversion component in one embodiment;

[0035] Figure 4 This is a system block diagram of a radio frequency conversion component in one embodiment;

[0036] Figure 5 This is a schematic diagram of the module assembly of a radio frequency conversion component in one embodiment;

[0037] Figure 6 This is a block diagram of the mode switching and processing system in one embodiment;

[0038] Explanation of reference numerals in the attached figures:

[0039] 10. Baseband processing unit; 20. Frequency conversion unit;

[0040] 30. Local oscillator synthesis unit; 40. Control and power supply unit;

[0041] 50. Signal processing unit; 60. Waveform feature extraction unit;

[0042] 61. Pattern recognition subunit; 70. Coherence processing unit;

[0043] 80. Information fusion unit; 81. Micro-Doppler analysis subunit;

[0044] 90. Switch; 100. Distributed synchronization unit; 110. Waveform prediction unit. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0046] In one exemplary embodiment, a radio frequency conversion component is provided. This component features dual-mode operation (active / passive) as its core characteristic and achieves flexible switching between the two operating modes through a mode switching and processing system. Figure 1 As shown, the core components of this RF conversion assembly include a baseband processing unit 10, a frequency conversion unit 20, a local oscillator synthesis unit 30, and a mode switching and processing system; wherein:

[0047] The baseband processing unit 10 uses a millimeter-wave SoC chip to generate and process initial transmit and receive signals of 79–81 GHz.

[0048] The frequency conversion unit 20 includes an up-conversion subunit and a down-conversion subunit. The up-conversion subunit is used to up-mix the initial transmit signal output from the baseband processing unit 10 with a fixed-frequency local oscillator signal to generate a 92–94 GHz radio frequency transmit signal; the down-conversion subunit is used to down-mix the received 92–94 GHz echo signal with the local oscillator signal to output a 79–81 GHz initial receive signal to the baseband processing unit 10.

[0049] The local oscillator synthesis unit 30 is used to generate and provide the local oscillator signal at the fixed frequency.

[0050] like Figure 2 As shown, the electrical connections and signal transmission paths of the core functional chips of the component are illustrated from a chip-level perspective, demonstrating the physical carriers of each unit and their interaction logic; among which:

[0051] The physical carrier of the baseband processing unit 10 is the ADT7880 millimeter-wave SoC chip, which serves as the signal processing core of the entire component. Its transmit signal output is connected to the GWB-TX1 and GWB-TX2 chips (the physical carriers of the upconversion subunit) in the frequency conversion unit 20 via two independent signal links, respectively, for outputting the initial transmit signal of 79–81 GHz; its receive signal input is connected to the GWB-RX1 and GWB-RX2 chips (the physical carriers of the downconversion subunit) via two independent signal links, respectively, for receiving the initial receive signal of 79–81 GHz after down-mixing processing.

[0052] The upconversion subunits (GWB-TX1, GWB-TX2) and downconversion subunits (GWB-RX1, GWB-RX2) in the frequency conversion unit 20 are both connected to the local oscillator chip in the local oscillator synthesis unit 30, receiving the fixed-frequency local oscillator signal output from it. The power supply terminals of GWB-TX1 and GWB-TX2 are connected to LDO2 via switches 1 and 2, respectively; the power supply terminals of GWB-RX1 and GWB-RX2 are connected to LDO3 via switches 3 and 4, respectively. These switches (switches 1–4) and the voltage regulators (LDO1–LDO5) together constitute the control and power supply unit 40, used to control the power supply to the upconversion and downconversion subunits according to the operating mode. The power supply terminal of the ADT7880 chip is connected to LDO1, and the power supply terminal of the local oscillator chip is connected to LDO4.

[0053] like Figure 3 As shown, the three-layer stacked structure of the component (bottom layer - baseband layer, middle layer - frequency conversion / local oscillator layer, top layer - antenna layer) is displayed from a physical cross-section perspective, and the transmission paths of signals in different frequency bands are marked; among them:

[0054] The bottom layer is the baseband layer, integrating the ADT7880 chip, which serves as the signal processing base for the entire component. It interacts with the middle layer via vertical microstrip lines. In active mode, the initial 79–81 GHz transmit signal output by the ADT7880 is transmitted upwards through the microstrip lines to the GWB-TX chip in the middle layer; simultaneously, the initial 79–81 GHz receive signal, after down-conversion processing by the GWB-RX chip, is transmitted downwards through the microstrip lines to the ADT7880.

[0055] The middle layer is the frequency converter / local oscillator layer, employing a high-frequency Rogers board to reduce millimeter-wave signal transmission loss. GWB-TX1 and GWB-TX2 chips are arranged vertically on the left, and GWB-RX1 and GWB-RX2 chips are arranged vertically on the right, with the local oscillator chip positioned in the center. The local oscillator chip transmits a fixed-frequency local oscillator signal to the four frequency converter chips via four horizontal microstrip lines, ensuring frequency consistency during the mixing process.

[0056] The top layer is the antenna layer, made of metal. It has a TX transmitting slot on the left and an RX receiving slot on the right. The metal waveguide cavities within the slots guide the directional transmission of 92–94 GHz signals. In active mode, the 92–94 GHz RF transmit signal output by the GWB-TX chip is transmitted to the TX transmitting slot via a horizontal microstrip line and radiated outwards. External 92–94 GHz echo signals are received through the RX receiving slot and transmitted to the GWB-RX chip via a horizontal microstrip line for processing.

[0057] This three-layer stacked structure achieves a combination of vertical and horizontal signal transmission, ensuring low-loss transmission of millimeter-wave signals while making the layout of each functional unit compact and reasonable.

[0058] like Figure 4 As shown, the core workflow of the component is simplified from a functional module level perspective, with a focus on presenting the signal processing closed loop in active mode:

[0059] The baseband processing layer on the left is the signal generation and processing center, capable of generating initial transmit signals in the 79–81 GHz range and receiving and processing initial receive signals in the 79–81 GHz range. Its transmit signal output interface is connected to the upconversion unit of the intermediate frequency conversion layer, while its receive signal input interface is connected to the downconversion unit.

[0060] The intermediate frequency conversion layer is divided into two parts: an up-conversion unit and a down-conversion unit. The up-conversion unit receives the initial 79–81 GHz transmit signal from the baseband processing layer and the fixed-frequency local oscillator signal from the local oscillator chip. After up-mixing, it outputs a 92–94 GHz radio frequency transmit signal and sends it to external space through the right-side interface. The down-conversion unit receives the 92–94 GHz echo signal and the local oscillator signal from external space. After down-mixing, it generates the initial 79–81 GHz receive signal and sends it back to the baseband processing layer.

[0061] The intermediate local oscillator synthesis unit provides a stable fixed-frequency local oscillator signal for the frequency conversion layer, which is fed into the up-conversion unit and the down-conversion unit respectively to ensure that the mixing reference is consistent.

[0062] The above modules are connected through a signal link to form a complete active mode signal closed loop: baseband signal generation → up-conversion transmission → external reflection → down-conversion reception → baseband processing.

[0063] like Figure 5 As shown, the mechanical fixing and electrical connection methods between components and key external parts are illustrated from a system-level assembly perspective; among them:

[0064] The top-layer waveguide antenna is made of aluminum alloy with a silver-plated surface to reduce signal loss. The bottom features TX transmit and RX receive slots corresponding to the mid-layer frequency conversion component, and threaded holes at the four corners. The signal interface on the top of the frequency conversion component (integrating GWB-TX, GWB-RX, and a local oscillator chip) is precisely aligned with the transmit and receive slots of the waveguide antenna and mechanically secured with four M2 stainless steel screws, ensuring that the positional deviation between the two does not exceed 0.1mm to prevent leakage during 92–94GHz signal transmission.

[0065] The bottom processing board uses an FR4 epoxy glass cloth substrate. The top integrates the ADT7880 chip (baseband processing unit 10) and related circuits of the control and power supply unit 40, with solder pads arranged along the edges. The surface mount solder joints on the bottom of the frequency conversion component correspond one-to-one with the solder pads on the processing board, achieving electrical interconnection through SMT surface mount technology to establish a stable signal and power transmission channel.

[0066] This three-layer assembly structure, consisting of a waveguide antenna, a frequency conversion component, and a processing board, ensures mechanical stability through screw connections and low-loss electrical connections through surface mount soldering, guaranteeing reliable transmission of baseband, radio frequency, power supply, and control signals. The bottom of the processing board also features external power and control signal interfaces for interaction with external systems.

[0067] In addition, the mode switching and processing system is used to selectively transmit the echo signal or the target radio frequency signal from the external non-cooperative irradiation source to the baseband processing unit 10 after down-conversion processing in two operating modes; the operating modes include active mode and passive mode.

[0068] In one exemplary embodiment, such as Figure 6 As shown, the RF conversion component also includes a mode switching and processing system, which includes a signal processing unit 50, a switching switch 90, and a control and power supply unit 40; in passive mode, it may optionally include a waveform feature extraction unit 60, a coherent processing unit 70, and an information fusion unit 80. Wherein:

[0069] The signal processing unit 50 is used to receive target radio frequency signals from an external non-cooperative irradiation source through a signal input interface. The signals include direct path signals and reflected path signals. The signal processing unit 50 is configured with a down-conversion channel for down-converting the target radio frequency signals into intermediate frequency signals.

[0070] The first port of the switch 90 is connected to the output of the frequency conversion unit 20 to receive the echo signal, the second port is connected to the output of the signal processing unit 50 to receive the intermediate frequency signal, and the common output is connected to the baseband processing unit 10.

[0071] The control and power supply unit 40 is used to control the selection state of the switching switch 90 according to the operating mode, and to control the power supply to the up-conversion subunit, the down-conversion subunit, and the signal processing unit 50.

[0072] In active mode: the control and power supply unit 40 controls the switching switch 90 to select the first port, so that the echo signal is transmitted to the baseband processing unit 10 through the switching switch 90 and supplies power to the up and down frequency conversion subunits;

[0073] Figure 2Under the management of the control and power supply unit 40, the LDO5 provides operating voltage to the switch 90. The specific selection state of the switch 90 (first port or second port) is directly controlled by the control signal generated by the control and power supply unit 40. Furthermore, the initial received signals output by GWB-RX1 and GWB-RX2 are connected to the first port of the switch 90 via a aggregation link, in addition to being connected to the ADT7880; the intermediate frequency signal output by the signal processing unit 50 in the mode switching and processing system is connected to the second port of the switch 90. The common output terminal of the switch 90 is ultimately connected to the intermediate frequency signal input terminal of the ADT7880, providing a path for signal transmission in passive mode.

[0074] In passive mode: the control and power supply unit 40 controls the switching switch 90 to select the second port, so that the intermediate frequency signal is transmitted to the baseband processing unit 10 through the switching switch 90 and power the signal processing unit 50.

[0075] It can be understood that the signal processing unit 50 is the core module for receiving external non-cooperative signals and performing down-conversion. It is used to receive radio frequency signals from non-cooperative illumination sources such as civilian communication signals and radar signals, and convert them into intermediate frequency signals for subsequent processing. This unit includes a signal input interface and a down-conversion channel.

[0076] Optionally, the target radio frequency signal (including direct path signal and reflected path signal) of the non-cooperative illumination source is first received through a broadband radio frequency antenna array (such as a microstrip antenna or Vivaldi antenna).

[0077] In this embodiment, the signal input interface includes an impedance matching network and an anti-interference filter, which are used to preprocess the received target radio frequency signal.

[0078] In this embodiment, the signal processing unit 50 is a multi-channel parallel architecture, configured with multiple independent down-conversion channels, capable of simultaneously receiving and down-converting signals from multiple non-cooperative illumination sources of different frequency bands and standards. Accordingly, the signal enters the multi-channel RF front-end module, undergoes initial amplification by a low-noise amplifier (LNA), and then selects the target signal channel via a multi-channel RF switch. It then passes through an impedance matching network (to achieve 50Ω impedance matching and reduce reflections) and an anti-interference filter (to suppress out-of-band interference) to obtain a clean RF signal. Each channel operates independently: after amplification by the LNA, the RF signal is mixed by a broadband mixer with the local oscillator signal generated by a tunable local oscillator (TLO) synthesizer to achieve down-conversion; the level is then adjusted by an intermediate frequency amplifier, and finally converted into a digital intermediate frequency signal by a high-speed ADC for subsequent use.

[0079] In passive mode, the RF conversion component also includes:

[0080] The waveform feature extraction unit 60 is connected to the baseband processing unit 10 and is used to receive the intermediate frequency signal from the switching switch 90 and extract the signal waveform features of the non-cooperative irradiation source through digital signal processing.

[0081] The waveform feature extraction unit 60 is a module that performs preprocessing and feature analysis on the intermediate frequency signal. It extracts the signal fingerprint of the non-cooperative irradiation source through digital signal processing (such as filtering, time-frequency analysis, etc.).

[0082] In this embodiment, the waveform feature extraction unit 60 is further configured with a pattern recognition subunit 61, which is used to match the extracted signal waveform features with a pre-stored non-cooperative irradiation source signal feature database to identify the identity and signal type of the irradiation source.

[0083] Optionally, the signal power (average power, peak power) is monitored in real time, and the gain is adjusted through a digital / analog variable gain amplifier to prevent the signal from being too strong and causing ADC saturation or too weak and causing insufficient signal-to-noise ratio; based on the real-time spectrum analysis results, the FIR or IIR filter coefficients are dynamically loaded to retain the target frequency band and suppress out-of-band noise and adjacent channel interference; the signal is decomposed through discrete wavelet transform, and low amplitude coefficients are thresholded for noise reduction, and high energy components are selectively reconstructed to reduce the amount of data and retain key features.

[0084] Based on the preprocessed signal, digital signal processing algorithms are used to analyze the features: the signal modulation type (such as QPSK, 16QAM, OFDM) is identified by means of cyclic stationary analysis, higher-order spectrum analysis, and time-frequency analysis (such as short-time Fourier transform), and parameters such as symbol rate, carrier frequency, and phase are estimated; combined with the pattern recognition subunit 61, the extracted features are matched with the pre-stored illumination source feature database (containing known communication and radar signal features) to confirm the identity and signal type of the non-cooperative illumination source.

[0085] The coherent processing unit 70 is connected to the waveform feature extraction unit 60. It is used to dynamically reconstruct the local reference signal based on the extracted signal waveform features, and use the reference signal to perform coherent processing on the reflected path signal to calculate the time delay difference and Doppler frequency shift of the reflected path signal relative to the direct path signal.

[0086] The coherent processing unit 70 is the core module for realizing coherent matching and parameter measurement of non-cooperative signals. Based on the signal parameters provided by the waveform feature extraction unit 60, it generates a local reference signal that is consistent with the direct path signal. Then, it calculates the time delay difference (corresponding to the target distance) and Doppler frequency shift (corresponding to the target radial velocity) between the reflected path signal and the direct path signal through coherent matching.

[0087] Optionally, based on the parameters such as carrier frequency, modulation type, symbol rate, and chip sequence output by the waveform feature extraction unit 60, an initial reference signal is generated through digital signal synthesis technology (such as digital up-conversion and digital modulation); the frequency phase deviation is calibrated through carrier recovery (such as Costas loop), the alignment timing is restored through symbol timing, and the propagation loss is compensated through channel equalization, finally obtaining a local reference signal that is highly coherent with the direct path signal.

[0088] Furthermore, a broadband cross-correlation operation is performed between the local reference signal and the digital intermediate frequency signal of the reflection path (using FFT to achieve efficient convolution) to calculate the cross-correlation function; its peak position corresponds to the time delay difference, and the changes in peak phase and amplitude correspond to the Doppler frequency shift; parabolic interpolation or spline interpolation is used to optimize the time delay estimation accuracy, and multi-pulse phase coherent accumulation is used to optimize the Doppler estimation accuracy. At the same time, dynamic channel calibration and phase compensation are performed to reduce the influence of non-ideal factors.

[0089] The information fusion unit 80, connected to the coherent processing unit 70, is used to fuse time delay difference and Doppler frequency shift information to output target detection results. The information fusion unit 80 is a module that integrates measurement data and outputs target results. It receives time delay difference (corresponding to target distance) and Doppler frequency shift (corresponding to target radial velocity), and uses a multi-dimensional data fusion algorithm to estimate and track the target state, ultimately outputting target detection results (such as position, velocity, trajectory, etc.).

[0090] Optionally, a state estimation algorithm (such as Kalman filtering, extended Kalman filtering, or unscented Kalman filtering) is employed, using the distance converted from time delay difference and the radial velocity converted from Doppler frequency shift as input, to estimate and update the target state (position, velocity, acceleration) in real time, suppressing measurement noise, smoothing the target trajectory, and ensuring tracking stability. The final output includes core detection results such as the target's real-time position, velocity vector, heading, and continuous trajectory.

[0091] In this embodiment, the information fusion unit 80 is further configured with a micro-Doppler analysis subunit 81, which is used to analyze the micro-Doppler effect in the reflected path signal, thereby reconstructing the structural micro-motion characteristics of the detected target or generating a three-dimensional image, and outputting this information as part of the target detection result. Micro-Doppler information (such as frequency modulation generated by the movement of target components) can be further fused to reconstruct the target's micro-motion characteristics (such as UAV rotor rotation, human gait), enhancing the target recognition capability.

[0092] Furthermore, such as Figure 6 As shown, the waveform feature extraction unit 60 is further configured with a preprocessing subunit for performing at least one of the following operations on the intermediate frequency signal before extracting the signal waveform features:

[0093] (1) Dynamic gain control based on signal power statistics.

[0094] This operation dynamically adjusts the signal gain by monitoring the power statistics characteristics of the intermediate frequency signal (such as average power and peak power) in real time, ensuring that the signal amplitude is within the optimal dynamic range required for subsequent processing, and avoiding signal distortion or signal-to-noise ratio reduction due to power anomalies.

[0095] Optionally, the power data of the digital intermediate frequency signal can be acquired in real time, and statistical indicators such as average power, peak power and root mean square power can be calculated to determine whether it exceeds the preset optimal operating range. Based on the statistical results, the gain can be adjusted through a digital or analog variable gain amplifier: reduce the gain when the signal power is too high to prevent ADC saturation; increase the gain when the power is too low to avoid insufficient signal-to-noise ratio.

[0096] This operation can prevent ADC saturation or insufficient signal-to-noise ratio caused by abnormal signal power, ensure the integrity of the intermediate frequency signal, and keep the signal in the optimal signal-to-noise ratio range.

[0097] (2) Use configurable filters for digital filtering of specific frequency bands of the illumination source.

[0098] Based on the real-time spectral characteristics and identity information of the non-cooperative illumination source (provided by the pattern recognition subunit 61), the operation dynamically configures the digital filter parameters (such as center frequency, bandwidth, and filter type), retaining only the target signal frequency band and filtering out out-of-band noise and adjacent channel interference.

[0099] Optionally, by combining the real-time spectrum analysis results with the illumination source type identified by the pattern recognition subunit 61, the target parameters of the digital filter are determined; the matching filter coefficients are dynamically loaded into the digital filter (FIR or IIR structure) by the DSP or FPGA; the intermediate frequency signal is input into the configured filter, and a clean target frequency band signal is output.

[0100] This operation can accurately filter out out-of-band interference, reduce the impact of noise on feature extraction, and improve the accuracy of waveform feature extraction.

[0101] (3) Bandwidth compression preprocessing based on wavelet transform.

[0102] This operation uses wavelet transform to perform time-frequency decomposition on the intermediate frequency signal, selectively retaining high-energy feature components while removing noise, thereby achieving signal bandwidth compression and data simplification without losing core feature information.

[0103] Optionally, a multi-level discrete wavelet transform is used to decompose the digital intermediate frequency signal, breaking the signal down into wavelet coefficients of different scales; a noise threshold is set, low amplitude coefficients are set to zero, and high energy coefficients are retained; high energy coefficients are selected to reconstruct the signal, and noise coefficients are discarded, thereby achieving bandwidth compression and data reduction.

[0104] This operation can significantly reduce the amount of data, alleviate the computational burden on the waveform feature extraction unit 60, and improve the processing efficiency of the component; at the same time, it suppresses noise during compression, retains key features, and ensures the accuracy of feature extraction.

[0105] Furthermore, such as Figure 6 As shown, the radio frequency conversion component also includes a distributed synchronization unit 100, which is connected to the waveform feature extraction unit 60 and / or the coherent processing unit 70. It is used to exchange signal waveform features or target detection results among multiple radio frequency conversion components, and to use a cross-correlation algorithm to achieve time and frequency synchronization between the components.

[0106] The distributed synchronization unit 100 is the core module for enabling multiple distributed deployment components to work together. Through information exchange and time-frequency synchronization functions, it ensures data communication between multiple components, unifies time and frequency references, and supports multi-station collaborative detection and coherent accumulation.

[0107] Optionally, key signal features (such as carrier frequency, modulation method, and symbol rate) of the non-cooperative illumination source are obtained from the waveform feature extraction unit 60, and processing results such as target delay difference and Doppler frequency shift are obtained from the coherent processing unit 70; low-latency and high-bandwidth data interaction between components is realized through a high-speed data communication interface; the collected signal features and processing results are distributed to all distributed components to ensure global data consistency.

[0108] Furthermore, using GPS or BeiDou timing signals as a coarse synchronization reference, the time and frequency of each component are initially aligned; the signal characteristics of non-cooperative illumination sources (such as the starting point of periodic signals and carrier frequency) are shared, and the cross-correlation function of the same signal received by different components is calculated through a cross-correlation algorithm, with the peak position corresponding to the time offset and the peak phase change corresponding to the frequency offset; based on the estimated time and frequency offset, the local clock and frequency deviation of each component are corrected through adaptive filtering (such as Kalman filtering) to achieve high-precision synchronization.

[0109] In this embodiment, through data sharing and time-frequency synchronization, multiple distributed components can form a distributed array, expanding the detection range and accuracy, and supporting target detection at greater distances and in wider areas.

[0110] In one exemplary embodiment, such as Figure 6 As shown, the RF conversion component also includes a waveform prediction unit 110, connected to the information fusion unit 80, for:

[0111] Based on the target detection results and historical signal waveform feature data, the motion trajectory and signal reflection characteristics of the detected target are predicted through a pre-set deep learning model;

[0112] The transmitted waveform parameters are determined based on the motion trajectory and signal reflection characteristics. These parameters are used by the baseband processing unit 10 to adjust the modulation method, bandwidth, or transmission timing of subsequent transmitted signals.

[0113] Among them, the historical signal waveform feature data is extracted and stored by the waveform feature extraction unit 60 after processing the previously received non-cooperative irradiation source signals.

[0114] The waveform prediction unit 110 is an intelligent control module that optimizes the transmitted signal based on the target's dynamic and historical data. It uses a pre-set deep learning model to predict the target's reflection characteristics and adapt the transmission parameters, and outputs the transmitted waveform parameters to the baseband processing unit 10, so that the subsequent transmitted signal is more in line with the target's state, thereby improving detection efficiency and accuracy.

[0115] Among them, the target detection results come from the real-time target data (position, velocity, trajectory, etc.) of the information fusion unit 80; the historical signal waveform feature data are the features extracted by the waveform feature extraction unit 60 from previous non-cooperative signals (such as modulation type, symbol rate, reflection intensity, etc.); the pre-set deep learning model is a neural network trained on historical data, used to learn the correlation between the target motion state and the signal reflection characteristics; the transmitted waveform parameters include modulation method, signal bandwidth, transmission timing, etc.

[0116] Optionally, real-time target detection results (position, velocity, acceleration, trajectory trend, etc.) can be obtained from the information fusion unit 80; historical data (such as reflection intensity, frequency shift, attenuation characteristics, etc.) matching the current target type can be extracted.

[0117] Real-time target data and matched historical feature data are input into a pre-built deep learning model; based on the learned correlation between target motion and reflection characteristics, the model predicts the target's motion trajectory and corresponding signal reflection characteristics over a future period of time, such as changes in reflected signal intensity, frequency response, and attenuation trend.

[0118] Based on the predicted target trajectory and reflection characteristics, the transmission signal parameters are dynamically adapted: if the predicted reflection signal is weak (e.g., long distance, small reflection cross-section), a stronger anti-interference modulation method (e.g., pulse compression coding) is adopted or the bandwidth is increased; if the predicted target speed is fast, the transmission pulse interval is shortened; if the predicted frequency offset is large, the modulation bandwidth is adjusted.

[0119] The determined transmission waveform parameters are sent to the baseband processing unit 10, which adjusts the generation logic of subsequent transmission signals to achieve dynamic optimization of the transmission signals.

[0120] In this embodiment, by predicting the target's reflection characteristics, the transmitted signal parameters are matched with the target's state, avoiding the problem that fixed parameters cannot adapt to the target's dynamic changes, and improving the echo signal-to-noise ratio and detection performance.

[0121] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0122] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A radio frequency conversion component, characterized in that, The radio frequency conversion component includes: The baseband processing unit (10) uses a millimeter-wave SoC chip to generate and process the initial transmit signal and the initial receive signal of 79-81GHz. The frequency conversion unit (20) includes an up-conversion subunit and a down-conversion subunit. The up-conversion subunit is used to up-mix the initial transmit signal output by the baseband processing unit (10) with a local oscillator signal of a fixed frequency to output a radio frequency transmit signal of 92-94 GHz. The down-conversion subunit is used to down-mix the received 92-94 GHz echo signal with the local oscillator signal to output the initial receive signal to the baseband processing unit (10). Local oscillator synthesis unit (30) is used to generate and provide the fixed frequency local oscillator signal; A mode switching and processing system is used to selectively down-convert the echo signal or the target radio frequency signal from an external non-cooperative irradiation source and then transmit it to the baseband processing unit (10) in two operating modes; the operating modes include active mode and passive mode. The mode switching and processing system includes: The signal processing unit (50) is used to receive the target radio frequency signal through a signal input interface. The target radio frequency signal includes a direct path signal and a reflected path signal. The signal processing unit (50) is configured with a down-conversion channel for down-converting the target radio frequency signal into an intermediate frequency signal. The switching switch (90) has a first port connected to the output of the frequency conversion unit (20) to receive the echo signal, a second port connected to the output of the signal processing unit (50) to receive the intermediate frequency signal, and a common output connected to the baseband processing unit (10). The control and power supply unit (40) is used to control the selection state of the switching switch (90) according to the working mode, and to control the power supply on and off of the up-conversion subunit, the down-conversion subunit and the signal processing unit (50); In the active mode: the control and power supply unit (40) controls the switching switch (90) to select the first port, so that the echo signal is transmitted to the baseband processing unit (10) via the switching switch (90) and supplies power to the upconversion subunit and the downconversion subunit; In the passive mode: the control and power supply unit (40) controls the switching switch (90) to select the second port, so that the intermediate frequency signal is transmitted to the baseband processing unit (10) via the switching switch (90) and supplies power to the signal processing unit (50).

2. The radio frequency conversion component according to claim 1, characterized in that, In the passive mode, the radio frequency conversion component further includes: A waveform feature extraction unit (60) is connected to the baseband processing unit (10) and is used to receive the intermediate frequency signal from the switching switch (90) and extract the signal waveform features of the non-cooperative irradiation source from the intermediate frequency signal through digital signal processing. The coherent processing unit (70) is connected to the waveform feature extraction unit (60) and is used to dynamically reconstruct the local reference signal based on the extracted signal waveform features, and to perform coherent processing on the reflection path signal using the local reference signal to calculate the time delay difference and Doppler frequency shift of the reflection path signal relative to the direct path signal. The information fusion unit (80) is connected to the coherent processing unit (70) and is used to fuse the time delay difference and the Doppler frequency shift information to output the target detection result.

3. The radio frequency conversion component according to claim 2, characterized in that, The signal processing unit (50) is a multi-channel parallel architecture, configured with multiple independent down-conversion channels. The multiple down-conversion channels can synchronously receive and down-convert signals from multiple non-cooperative irradiation sources of different frequency bands and different standards.

4. The radio frequency conversion component according to claim 2, characterized in that, The waveform feature extraction unit (60) is further configured with a pattern recognition subunit (61) for matching the extracted signal waveform features with the pre-stored signal waveform feature databases of the non-cooperative irradiation sources to identify the identity and signal mode of the non-cooperative irradiation source.

5. The radio frequency conversion component according to claim 2, characterized in that, The radio frequency conversion component further includes a distributed synchronization unit (100), which is connected to the waveform feature extraction unit (60) and / or the coherent processing unit (70) for exchanging the respective signal waveform features or target detection results among the multiple radio frequency conversion components, and using a cross-correlation algorithm to keep the time and frequency synchronized among the radio frequency conversion components.

6. The radio frequency conversion component according to claim 2, characterized in that, The information fusion unit (80) is also equipped with a micro-Doppler analysis subunit (81) for analyzing the micro-Doppler effect in the reflection path signal, and reconstructing the structural features of the target or generating a three-dimensional image based on the micro-Doppler effect, and outputting this information as part of the target detection result.

7. The radio frequency conversion component according to claim 2, characterized in that, The radio frequency conversion component further includes a waveform prediction unit (110), which is connected to the information fusion unit (80) and is used for: Based on the target detection results and historical signal waveform feature data, the motion trajectory and signal reflection characteristics of the detected target are predicted through a pre-set deep learning network. Based on the motion trajectory and the signal reflection characteristics, the transmission waveform parameters are determined. The transmission waveform parameters are used by the baseband processing unit (10) to adjust the modulation method, bandwidth or transmission timing of the subsequent transmission signal.

8. The radio frequency conversion component according to claim 7, characterized in that, The historical signal waveform feature data is obtained by the waveform feature extraction unit (60) after processing the previously received non-cooperative irradiation source signals and storing them.

9. The radio frequency conversion component according to claim 1, characterized in that, The signal input interface includes an impedance matching network and an anti-interference filter for preprocessing the received target radio frequency signal.

Images