A 92 ghz mimo roadside radar radio frequency system based on frequency conversion components

Abstract

The application relates to a 92GHz MIMO roadside radar radio frequency system based on a frequency conversion component, the radar radio frequency system comprising: a baseband processing layer comprising at least two cascaded millimeter wave system level SoC chips; a frequency conversion layer being a frequency conversion component FCM, the FCM comprising an oscillator synthesis unit, an upper mixer group and a lower mixer group; the oscillator synthesis unit is used for generating an oscillator signal of 13GHz; the input end of the upper mixer group is connected with the output end of the baseband processing layer; the oscillator end of the upper mixer group is connected with the oscillator synthesis unit; the output end of the lower mixer group is connected with the input end of the baseband processing layer; the oscillator end of the lower mixer group is connected with the oscillator synthesis unit, and is used for outputting echo signals of 92-94GHz received by an antenna array to the baseband processing layer after frequency conversion to 79-81GHz. The application can realize frequency conversion of 92-94GHz.

Description

Technical Field

[0001] This application relates to the field of radio frequency technology, and in particular to a 92GHz MIMO roadside radar radio frequency system based on a frequency conversion component. Background Technology

[0002] Millimeter-wave radar technology, with its advantages of all-weather operation, high precision, and non-contact measurement, has become a core sensor in intelligent transportation, autonomous driving, and industrial sensing. In roadside radar applications, millimeter-wave radar systems undertake key tasks such as traffic flow monitoring, target tracking, event detection, and safety warnings, playing a significant role in improving road safety and efficiency. To pursue higher sensing accuracy and resolution, the industry is gradually expanding to higher frequency bands. The 92GHz band, due to its shorter wavelength, can achieve higher angular resolution and a more compact antenna size, thus demonstrating enormous application potential and attracting widespread attention from the industry.

[0003] Currently, mainstream millimeter-wave radar systems generally adopt a highly integrated system-on-a-chip architecture, but their operating frequency bands are mostly concentrated in 77-81GHz.

[0004] However, existing architectures based on highly integrated SoCs in the 77-81GHz range, due to inherent technical limitations, struggle to simultaneously achieve high performance, low cost, and high reliability when extending to the 92GHz high-frequency band to address complex roadside applications. Therefore, a new technical solution is urgently needed. Summary of the Invention

[0005] Therefore, it is necessary to provide a 92GHz MIMO roadside radar radio frequency system based on a frequency conversion component to address the above-mentioned technical problems.

[0006] This application provides a 92GHz MIMO roadside radar radio frequency system based on a frequency conversion component. The radar radio frequency system includes a baseband processing layer, a frequency conversion layer, and an antenna array; wherein:

[0007] The baseband processing layer includes at least two cascaded millimeter-wave system-on-a-chip (SoC) chips for generating 79-81 GHz transmit signals and processing 79-81 GHz receive signals.

[0008] The frequency conversion layer is the frequency conversion component FCM, which includes a local oscillator synthesis unit, an up mixer group, and a down mixer group.

[0009] Local oscillator synthesis unit, used to generate a 13GHz local oscillator signal;

[0010] The input of the upmixer group is connected to the output of the baseband processing layer, and the local oscillator of the upmixer group is connected to the local oscillator synthesis unit, which is used to upconvert the 79-81GHz transmit signal to 92-94GHz and then output it to the antenna array.

[0011] The output of the downmixer group is connected to the input of the baseband processing layer, and the local oscillator of the downmixer group is connected to the local oscillator synthesis unit, which is used to downconvert the 92-94GHz echo signal received by the antenna array to 79-81GHz and then output it to the baseband processing layer.

[0012] In one embodiment, the millimeter-wave system-on-a-chip (SoC) is also configured to execute a frequency parameter adaptive adjustment algorithm to find the target operating point with minimal interference in the following manner:

[0013] Measure the initial interference power value at the current operating point, which is defined by a set of system operating parameters;

[0014] Adjust the value of each system operating parameter at the current operating point and measure the response interference power value after each adjustment;

[0015] Based on the difference between the adjusted response interference power value and the initial interference power value, determine the gradient vector of how the interference power changes with each system operating parameter;

[0016] Candidate operating points are determined based on the gradient vector;

[0017] Determine whether the candidate operating point meets the preset physical and protocol constraints;

[0018] If the conditions are met, control instructions are generated to distribute the values ​​of each system operating parameter contained in the candidate operating point to the corresponding system components for execution, and the candidate operating point is used as the target operating point.

[0019] If it does not meet the requirements, the candidate operating point will be corrected to the nearest point that meets the physical and protocol constraints, and control instructions will be generated to adjust the values ​​of each system operating parameter to the nearest point, and the nearest point will be used as the target operating point.

[0020] In one embodiment, the system operating parameters include one or more of the following: the output frequency of the local oscillator synthesis unit, the transmit power of the millimeter-wave system-on-a-chip (SoC), the receiver gain, and the frequency modulation slope of the linear frequency modulation signal.

[0021] In one embodiment, based on the difference between the adjusted response interference power value and the initial interference power value, a gradient vector of the interference power varying with each system operating parameter is determined, including:

[0022] For each system operating parameter to be optimized at the current operating point, perform the following operations:

[0023] Keeping other system operating parameters unchanged at the current operating point, the values ​​of the operating parameters of the system to be optimized are increased by a positive disturbance, and the first response disturbance power value is measured.

[0024] The difference between the first response interference power value and the initial interference power value is divided by the positive disturbance amount to calculate the estimated rate of change of the interference power with respect to the operating parameters of the system to be optimized.

[0025] Correspondingly, after all the operating parameters of the system to be optimized at the current operating point have been traversed, the obtained estimates of the rate of change are combined to obtain the gradient vector of the interference power as a function of each system operating parameter.

[0026] In one embodiment, the difference between the first response interference power value and the initial interference power value is divided by the positive disturbance amount to calculate an estimated rate of change of the interference power with respect to the operating parameters of the system to be optimized, including:

[0027] Keeping other system operating parameters unchanged at the current operating point, the values ​​of the operating parameters of the system to be optimized are increased by a negative disturbance, and the second response disturbance power value is measured.

[0028] The difference between the first response interference power value and the second response interference power value is divided by twice the disturbance amount to calculate the estimated rate of change of the interference power with respect to the operating parameters of the system to be optimized.

[0029] In one embodiment, the preset physical and protocol constraints include: the operating frequency band of the radio frequency signal after frequency conversion layer transformation is in the range of 92-94GHz, and the transmit power of the millimeter-wave system-on-a-chip (SoC) does not exceed the statutory limit.

[0030] In one embodiment, determining candidate operating points based on gradient vectors includes:

[0031] The calculation result is obtained by subtracting the product of the gradient vector and the preset learning rate factor from the coordinates of the current working point, and the calculation result is used as the candidate working point.

[0032] In one embodiment, the triggering condition for the cyclic execution of the adaptive adjustment algorithm for the triggering frequency parameter is: the magnitude of the gradient vector is less than a preset threshold, or the change in the initial interference power value is less than a preset tolerance in a consecutive preset number of iterations.

[0033] In one embodiment, the radar radio frequency system is configured to perform a self-calibration mode, in which the baseband processing layer can control the local oscillator synthesis unit and the up mixer group to generate a calibration signal with known power, and send it back to the baseband processing layer via the down mixer group for link gain measurement.

[0034] In one embodiment, the frequency conversion component FCM also includes a temperature sensor integrated internally.

[0035] The aforementioned 92GHz MIMO roadside radar RF system based on a frequency conversion component successfully upconverts the baseband signal of a mature 79-81GHz millimeter-wave system-on-a-chip (SoC) to 92-94GHz by introducing a frequency conversion layer (FCM). This combines the cost and reliability advantages of low-frequency chips with the beam accuracy and resolution advantages of high-frequency chips, avoiding the technical challenges and high costs of directly developing a 92GHz high-frequency SoC. This provides a solution for achieving high-performance and high-reliability roadside radar sensing. Attached Figure Description

[0036] 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.

[0037] Figure 1 This is a system architecture diagram of a 92GHz MIMO roadside radar radio frequency system based on a frequency conversion component in one embodiment;

[0038] Figure 2 This is a schematic diagram of the communication link of a 92GHz MIMO roadside radar radio frequency system based on a frequency conversion component in one embodiment;

[0039] Figure 3 This is a flowchart illustrating an adaptive frequency parameter adjustment algorithm in one embodiment.

[0040] Explanation of reference numerals in the attached figures:

[0041] 10. Baseband processing layer; 20. Frequency conversion layer; 30. Antenna array. Detailed Implementation

[0042] 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.

[0043] The 92GHz Multiple-Input Multiple-Output (MIMO) roadside radar radio frequency system is an advanced sensor system deployed at the edge of roads. Its core feature lies in its use of the millimeter-wave frequency band near 92GHz (e.g., 92-94GHz) for detection and the application of MIMO technology. This system simultaneously transmits and receives signals through multiple antennas, using signal processing to synthesize a virtual aperture. This improves the accuracy of distance, speed measurement, and angle resolution for targets such as vehicles and pedestrians in road scenes without significantly increasing hardware size.

[0044] In one exemplary embodiment, such as Figure 1 As shown, a 92GHz MIMO roadside radar RF system based on a frequency conversion component is provided. This radar RF system includes a baseband processing layer 10, a frequency conversion layer 20, and an antenna array 30. This system uses the frequency conversion layer 20 as a frequency band adaptation bridge to upconvert the 79-81GHz signal generated by the baseband processing layer to 92-94GHz for transmission. Then, it downconverts the 92-94GHz echo signal received by the antenna array back to 79-81GHz for baseband processing. This achieves the core objective of maintaining the baseband architecture while migrating the operating frequency band, combining the low-cost advantages of mature SoC chips with the high-resolution performance of the 92GHz band. Figure 1 This is a schematic diagram of the stacked structure of a 92GHz MIMO roadside radar radio frequency system based on a frequency conversion component:

[0045] The bottom layer is the baseband processing layer 10, which is connected to the external control / power supply module through connector 1 and connector 2. It integrates core circuits such as millimeter-wave system-on-a-chip (SoC) chips and provides signal processing and control functions for the upper layer.

[0046] The intermediate layer is the frequency conversion layer 20, which is the core functional layer for realizing the conversion between the 79-81GHz and 92-94GHz frequency bands. It is closely connected to the baseband processing layer 10 and the antenna array 30 through internal radio frequency traces / interfaces.

[0047] The top layer is antenna array 30, which includes a transmitting antenna (TX) and a receiving antenna (RX). TX is used to radiate 92-94GHz transmitted signals, and RX is used to receive 92-94GHz echo signals reflected from the target.

[0048] The following is a detailed introduction to each of the above components:

[0049] The baseband processing layer 10 includes at least two cascaded millimeter-wave system-on-a-chip (SoC) chips for generating 79-81 GHz transmit signals and processing 79-81 GHz receive signals.

[0050] Optionally, the baseband processing layer 10 includes at least two cascaded millimeter-wave system-on-a-chip (SoC) chips, each capable of integrating 3 transmit (Tx) / 4 receive (Rx) channels. Through a master-slave cascading mode (e.g., the master chip outputs a 40MHz reference clock, which is transmitted to the slave chip via the OSC_CLKOUT port to ensure synchronization between multiple chips), it can be expanded to a 6Tx / 8Rx (6T8R) MIMO channel configuration. Furthermore, a frame reference counter strictly controls the dual-chip ADC sampling window deviation within 5–10 ns, ensuring phase consistency of the multi-channel signals.

[0051] Its core functions are implemented as follows: Generating 79-81GHz transmit signals: The millimeter-wave system-on-a-chip (SoC) generates linear frequency modulated continuous wave (FMCW) signals through its built-in fractional-N type PLLHirp engine, and outputs them to the frequency conversion layer 20 via an internal power amplifier link. This 79-81GHz frequency band is the native operating frequency band of the millimeter-wave SoC, requiring no modification to the chip hardware architecture, and mature chip drivers and signal generation logic can be directly reused.

[0052] Processing 79-81GHz received signals: The 79-81GHz signals received from the frequency conversion layer 20 are pre-amplified by the chip's built-in low-noise amplifier (LNA), and then digitized by the analog-to-digital converter (ADC) with a sampling rate of 45Msps. Finally, the built-in digital signal processor (DSP) executes algorithms such as target detection, distance / velocity / angle calculation to complete the signal analysis and target information extraction.

[0053] Among them, the frequency conversion layer 20 is a frequency conversion module (FCM), which includes a local oscillator synthesis unit, an up mixer group, and a down mixer group.

[0054] The local oscillator synthesis unit is used to generate a 13GHz local oscillator signal.

[0055] Optionally, the local oscillator synthesis unit is based on an integrated PLL chip and an external balun design. Its core function is to generate a stable, low-noise 13GHz local oscillator signal. This frequency is determined based on the frequency conversion logic: transmit frequency band = baseband signal frequency + local oscillator frequency, receive baseband signal frequency = echo frequency band - local oscillator frequency (79-81GHz + 13GHz = 92-94GHz, 92-94GHz - 13GHz = 79-81GHz), ensuring no frequency shift during the conversion process. The phase noise of the 13GHz local oscillator signal is better than -110dBc / Hz at a 100kHz offset, and the output power is stable at 3.5dBm. When the local oscillator signal is distributed to the eight mixing channels via a power divider network, a λ / 4 impedance transformer and isolation resistors are used to optimize crosstalk between channels, with a crosstalk suppression capability ≥25dB, ensuring frequency stability during long-term operation.

[0056] like Figure 2 The diagram shown illustrates the signal processing flow of the frequency conversion layer 20, which includes:

[0057] Transmit Link (Upmixing Process): The 79-81GHz transmit signal output from the baseband processing layer (79-81GHz input arrow in the diagram) is input to the TX_IN1~TX_IN4 ports of the upmixer group; the 13GHz local oscillator signal generated by the local oscillator synthesis unit is synchronously input to the local oscillator terminal of the upmixer group; the upmixer group combines the 79-81GHz signal and the 13GHz local oscillator signal into a 92-94GHz signal through mixing operations. And output from ports TX_OUT1~TX_OUT4 to antenna array 30 to complete the up-band shift of the transmitted signal.

[0058] Receive link (downmixing process): The 92-94GHz echo signal received by the antenna array 30 is input to the RX_IN1~RX_IN4 ports of the downmixer group; the 13GHz local oscillator signal of the local oscillator synthesis unit is synchronously input to the local oscillator end of the downmixer group; the downmixer group combines the 92-94GHz echo signal and the 13GHz local oscillator signal into a 79-81GHz signal (92-94GHz-13GHz=79-81GHz) through mixing operations, and transmits it back to the baseband processing layer 10 from the RX_OUT1~RX_OUT4 ports, completing the downband shift of the echo signal.

[0059] The input of the upmixer group is connected to the output of the baseband processing layer 10, and the local oscillator of the upmixer group is connected to the local oscillator synthesis unit, which is used to upconvert the 79-81GHz transmission signal to 92-94GHz and then output it to the antenna array 30.

[0060] Optionally, the upmixer group adopts a 4-channel integrated design, corresponding one-to-one with the 4 transmit channels of the baseband processing layer 10. Its hardware connection strictly follows the modified claim 1: the input end is connected to the output end of the baseband processing layer 10, and the local oscillator end is connected to the local oscillator synthesis unit. Specifically, through upper sideband mixing technology, the 79-81GHz transmit signal output from the baseband processing layer and the 13GHz local oscillator signal provided by the local oscillator synthesis unit are nonlinearly combined in the mixer to extract the upper sideband component (i.e., 92-94GHz) of the baseband signal frequency + local oscillator frequency. After filtering out clutter by the internal filtering circuit, it is transmitted from the output end to the antenna array 30. The frequency conversion loss of the mixer is controlled at 8-10dB to ensure that the output signal power meets the radar detection range requirements (typical output power ≥ -5dBm).

[0061] The output of the downmixer group is connected to the input of the baseband processing layer 10, and the local oscillator of the downmixer group is connected to the local oscillator synthesis unit, which is used to downconvert the 92-94GHz echo signal received by the antenna array 30 to 79-81GHz and then output it to the baseband processing layer 10.

[0062] Optionally, the downmixer group also adopts a 4-channel integrated design, corresponding one-to-one with the 4 receiving channels of the baseband processing layer 10. The connection relationship is as follows: the output end is connected to the input end of the baseband processing layer 10, and the local oscillator end is connected to the local oscillator synthesis unit. Specifically: an image rejection mixing architecture is adopted to mix the 92-94GHz echo signal received by the antenna array 30 with the 13GHz local oscillator signal, and extract the lower sideband component (i.e., 79-81GHz) of the echo signal frequency minus the local oscillator frequency; at the same time, the image interference of the 79-81GHz intermediate frequency signal is eliminated by the built-in 90° phase shift network (image rejection ratio ≥30dB), and a low noise amplifier (LNA) is integrated at the front end of the mixer to control the noise figure of the receiving link to 3-4dB, thereby enhancing the ability to receive weak echo signals from long-distance, small targets.

[0063] Furthermore, antenna array 30 serves as the air interface for signal transmission and reception, connecting to the output of the upmixer group and the input of the downmixer group of the FCM via a standard WR10 waveguide port. The WR10 waveguide is a low-loss transmission structure for the 92-94GHz frequency band, with a transmission loss ≤0.5dB / m in this band, effectively reducing the attenuation of high-frequency signals during transmission and ensuring radar detection range (typical detection range ≥200m) and signal integrity. The working process of antenna array 30 is as follows: receiving the 92-94GHz transmitted signal output from the upmixer group and radiating it to the roadside detection area in the form of electromagnetic waves; simultaneously capturing the 92-94GHz echo signal reflected from the target and transmitting it to the input of the downmixer group through the WR10 waveguide, completing the signal transmission-reception closed loop.

[0064] In one exemplary embodiment, such as Figure 3 As shown, the millimeter-wave system-on-a-chip (SoC) is also configured to execute an adaptive frequency parameter adjustment algorithm to find the target operating point with minimal interference in the following manner. The specific implementation process of this algorithm is as follows:

[0065] S301. Measure the initial interference power value at the current operating point.

[0066] The current working point is defined by a set of system working parameters.

[0067] Optionally, the system operating parameters include one or more of the following: the output frequency of the local oscillator synthesizer, the transmit power of the millimeter-wave system-on-a-chip (SoC), the receiver gain of the millimeter-wave SoC, and the frequency modulation slope of the linear frequency modulated signal, represented by a vector as follows:

[0068]

[0069] in, The initial local oscillator frequency, This is the initial transmit power. This is the initial receiver gain. This represents the initial frequency modulation slope.

[0070] Furthermore, the initial interference power value is measured: the spectrum analysis module built into the millimeter-wave system-on-a-chip (SoC) is used to collect the current operating point data in real time. Interference signals (including environmental clutter, interference from other radio equipment, etc.) in the downlink are measured, and the initial interference power value is recorded as follows. This serves as a benchmark for subsequent parameter adjustments.

[0071] S302. Adjust the value of each system operating parameter at the current operating point and measure the response interference power value after each adjustment.

[0072] Optionally, based on the difference between the adjusted response interference power value and the initial interference power value, the gradient vector of the interference power changing with each system operating parameter is determined, including: for each operating parameter of the system to be optimized at the current operating point, the following operations are performed: keeping other system operating parameters at the current operating point unchanged, adding a positive disturbance to the value of the operating parameter of the system to be optimized, and measuring the first response interference power value; dividing the difference between the first response interference power value and the initial interference power value by the positive disturbance to calculate the estimated rate of change of the interference power with respect to the operating parameter of the system to be optimized.

[0073] Specifically, the difference between the first response interference power value and the initial interference power value is divided by a positive disturbance amount to calculate the estimated rate of change of the interference power with respect to the operating parameters of the system to be optimized. This includes: keeping other system operating parameters at the current operating point unchanged, adding a negative disturbance amount to the value of the operating parameters of the system to be optimized, and measuring the second response interference power value; the difference between the first response interference power value and the second response interference power value is divided by twice the disturbance amount to calculate the estimated rate of change of the interference power with respect to the operating parameters of the system to be optimized.

[0074] First, adjust the positive disturbance amount and measure the first response interference power value: add a positive disturbance amount (e.g., positive disturbance amount of the local oscillator frequency) to the value of the operating parameter of the system to be optimized. positive perturbation of transmit power To obtain new parameter values , (This is the positive disturbance value); Under this parameter configuration, the interference power value is measured and recorded as the first response interference power value.

[0075] Secondly, the negative disturbance adjustment and the measurement of the second response disturbance power value: keeping other system operating parameters unchanged, the value of the same operating parameter of the system to be optimized is increased by a negative disturbance (i.e., ), This is the positive disturbance amount; under this parameter configuration, the interference power value is measured and recorded as the second response interference power value;

[0076] Among them, the disturbance quantity The values ​​must be small and not exceed the safe range of parameters. For example, the disturbance of the local oscillator frequency should not exceed ±5MHz (to avoid exceeding 92-94GHz after frequency conversion), and the disturbance of the transmit power should not exceed ±1dBm (to avoid approaching the legal limit).

[0077] S303. Based on the difference between the adjusted response interference power value and the initial interference power value, determine the gradient vector of how the interference power changes with the operating parameters of each system.

[0078] Optionally, for each operating parameter of the system to be optimized, the first response interference power value can be used. With the second response interference power value Based on the difference, calculate the estimated rate of change of interference power relative to this parameter, using the following formula:

[0079]

[0080] This formula uses bidirectional disturbances to offset errors caused by the nonlinear characteristics of the system, making the estimated rate of change closer to the true value (the error can be reduced by more than 30% compared to unidirectional disturbances).

[0081] S304. Determine candidate operating points based on the gradient vector.

[0082] Optionally, after all the operating parameters of the system to be optimized at the current operating point have been traversed, the obtained estimates of the rates of change are combined to obtain the gradient vector of the interference power as a function of each system operating parameter, denoted as:

[0083]

[0084] in, This represents the gradient of the interference power P with respect to the system's operating parameter vector. It's a vector where each component represents the rate of change of the interference power in the corresponding parameter direction, and the direction of the gradient indicates the direction in which the interference power increases most rapidly. Therefore, in optimization algorithms, along its opposite direction (… P) Adjust the parameters to reduce the interference power. Unit: The unit for each component is [interference power unit] / [corresponding parameter unit], for example, dBm / MHz.

[0085] in, This is the first response interference power value. This is based on only the local oscillator frequency. Add a positive perturbation The measured interference power value, while keeping all other parameters constant, is expressed in dBm (decibel milliwatt) or other power units. This is the second response interference power value. This is based on only the local oscillator frequency. Add a negative perturbation - The measured interference power value, while keeping all other parameters constant, is expressed in dBm (decibels per milliwatt) or other power units.

[0086] in, Is it only transmitting power? Add a positive perturbation At that time, the measured interference power value; Is it only transmitting power? Add a negative disturbance The measured interference power value.

[0087] in, To only increase the receiver gain Add a positive perturbation At that time, the measured interference power value; To only increase the receiver gain Add a negative disturbance The measured interference power value.

[0088] in, To only adjust the frequency modulation slope Add a positive perturbation At that time, the measured interference power value; To only adjust the frequency modulation slope Add a negative disturbance — The measured interference power value.

[0089] Optionally, candidate operating points are determined based on the gradient vector, including: subtracting the product of the gradient vector and the preset learning rate factor from the coordinates of the current operating point to obtain the calculation result, and using the calculation result as a candidate operating point.

[0090] For example, using the gradient descent method, the parameters are adjusted along the direction of the fastest decrease in interference power. The calculation formula is as follows:

[0091]

[0092] in: These are the candidate operating points obtained through calculation; The preset learning rate factor (typically 0.1-0.3) is used to control the adjustment range of parameters. Too large a learning rate will lead to parameter overshoot (e.g., the transmit power exceeds the legal limit), while too small a learning rate will lead to slow algorithm convergence (increasing the number of iterations by more than 50%). In practical applications, it can be dynamically adjusted according to the magnitude of disturbance changes. (When the disturbance changes significantly, take the smaller value) Take the larger value when the disturbance is stable. ).

[0093] S305. Determine whether the candidate working point meets the preset physical and protocol constraints.

[0094] The preset physical and protocol constraints include: the operating frequency band of the radio frequency signal after conversion by the frequency conversion layer 20 is in the range of 92-94GHz, and the transmit power of the millimeter-wave system-on-a-chip does not exceed the legal limit.

[0095] Optionally, calculate candidate operating points. Corresponding radio frequency transmission band ( , (Baseband signal frequency), determine whether it meets the requirements. If the value exceeds this range, it indicates that the frequency band after frequency conversion does not meet the system design requirements and needs to be corrected.

[0096] The transmit power constraint is: to determine the transmit power among the candidate operating points. Does it meet the requirements? <<Statutory limits (e.g., the statutory limit for the transmit power of a 92GHz roadside radar is ≤23dBm); if the limit is exceeded, the transmit power must be forcibly corrected to the statutory limit.

[0097] Furthermore, the frequency conversion component FCM also includes an integrated temperature sensor.

[0098] Optionally, in conjunction with the temperature sensor integrated into the FCM, it is necessary to supplement the constraint judgment after temperature compensation: read the real-time reading T of the temperature sensor inside the FCM, and based on the pre-stored thermal drift model (such as... , For temperature coefficient, Using the reference temperature, calculate the temperature drift of each system's operating parameters; use the temperature drift to determine candidate operating points. Compensation is performed to obtain the candidate operating point after temperature compensation. Reassess Whether it complies with the above physical and protocol constraints, ensuring that temperature changes (such as the temperature of roadside equipment rising to 60°C in summer) do not cause parameter deviations.

[0099] S306. If the conditions are met, control instructions are generated to distribute the values ​​of each system operating parameter contained in the candidate operating point to the corresponding system components for execution, and the candidate operating point is used as the target operating point.

[0100] Optionally, if the candidate operating point meets the constraints: generate control instructions to... The system operating parameter values ​​contained therein are distributed to the corresponding system components (local oscillator synthesis unit, millimeter-wave system-on-a-chip transmit / receive link, signal modulation module) for execution. This is the target working point.

[0101] S307. If not, the candidate operating point is corrected to the nearest point that meets the physical and protocol constraints, and control instructions are generated to adjust the values ​​of each system operating parameter to the nearest point, and the nearest point is used as the target operating point.

[0102] Optionally, if the candidate operating point does not meet the constraints: use the projection method to... Corrected to the nearest point that meets the constraints, for example: if the RF band exceeds 94GHz, the local oscillator frequency will be adjusted. Revised to If the transmission power exceeds the limit, The value is corrected to the legal limit; after correction, a control command is generated to adjust the system parameters to the nearest point, which is the target operating point.

[0103] In one exemplary embodiment, the radar radio frequency system is configured to perform a self-calibration mode, in which the baseband processing layer 10 can control the local oscillator synthesis unit and the up mixer group to generate a calibration signal with known power, and send it back to the baseband processing layer 10 via the down mixer group for link gain measurement.

[0104] Understandably, this mode is independent of the frequency parameter adaptive adjustment algorithm and can be triggered on demand (such as when the system is powered on or periodically every 24 hours). Its core purpose is to correct the gain drift of the receiving link and ensure the accuracy of signal measurement.

[0105] The specific implementation process is as follows:

[0106] First, calibration signal generation and injection: The baseband processing layer 10 sends control commands to the local oscillator synthesis unit and the up mixer group to control the local oscillator synthesis unit to generate a 13GHz calibration local oscillator signal. The up mixer group mixes this signal with a fixed frequency (e.g., 80GHz) signal generated by the baseband to generate a calibration signal with known power (typical power of -10dBm, power accuracy ±0.2dBm). Through an internal coupling link, the calibration signal is injected into the input of the down mixer group (analog echo signal path).

[0107] Secondly, calibration signal feedback and gain measurement: The down-mixer group down-converts the calibration signal to 79-81GHz (consistent with the normal echo signal frequency band) and sends it back to the receiver of the baseband processing layer 10; the baseband processing layer measures the power of the feedback calibration signal using a built-in power meter, denoted as . .

[0108] Then, the link gain is calculated based on the injected calibration signal power. With measurement calibration signal power Calculate the net gain of the current receiving link. ;Will The gain drift is obtained by comparing it with the rated gain of the receiving link (typically 20dB). This drift can be used for parameter compensation in subsequent frequency parameter adaptive adjustment algorithms (such as correcting receiver gain). ).

[0109] Understandably, the self-calibration mode can control the gain error of the receiving link within ±0.5dB, avoiding the impact of gain drift caused by long-term operation (such as component aging) on ​​the accuracy of interference power measurement.

[0110] Optionally, the triggering condition for the adaptive adjustment algorithm of the trigger frequency parameter to be executed cyclically is: the magnitude of the gradient vector is less than a preset threshold, or the change in the initial interference power value is less than a preset tolerance in a preset number of consecutive iterations.

[0111] Based on the example above, the magnitude of the gradient vector is less than a preset threshold: ( The preset threshold (typically 0.1 dBm / parameter unit) indicates that the impact of parameter adjustment on interference power is minimal, and the system tends to stabilize; the change in the initial interference power value is less than the preset tolerance in a preset number of iterations: in N consecutive iterations (N is the preset number, typically 5-8 times), , The preset tolerance (typically 0.5 dBm) indicates that the interference power has stabilized at a low level and no further adjustment is needed. When any of the above conditions are met, the algorithm stops looping, and the system continues to operate at the current target operating point; if not, it returns to S301 to re-execute the iteration.

[0112] 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.

[0113] 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 92GHz MIMO roadside radar radio frequency system based on a frequency conversion component, characterized in that, The radar radio frequency system includes a baseband processing layer (10), a frequency conversion layer (20), and an antenna array (30); wherein: The baseband processing layer (10) includes at least two cascaded millimeter-wave system-on-a-chip (SoC) chips for generating 79-81 GHz transmit signals and processing 79-81 GHz receive signals. The frequency conversion layer (20) is a frequency conversion component FCM, which includes a local oscillator synthesis unit, an up mixer group and a down mixer group; The local oscillator synthesis unit is used to generate a 13GHz local oscillator signal; The input end of the upmixer group is connected to the output end of the baseband processing layer (10), and the local oscillator end of the upmixer group is connected to the local oscillator synthesis unit, which is used to upconvert the 79-81GHz transmission signal to 92-94GHz and then output it to the antenna array (30). The output of the downmixer group is connected to the input of the baseband processing layer (10), and the local oscillator of the downmixer group is connected to the local oscillator synthesis unit, which is used to downconvert the 92-94GHz echo signal received by the antenna array (30) to 79-81GHz and then output it to the baseband processing layer (10). The millimeter-wave system-on-a-chip (SoC) is also configured to execute an adaptive frequency parameter adjustment algorithm to find the target operating point with minimal interference in the following ways: Measure the initial interference power value at the current operating point, which is defined by a set of system operating parameters; Adjust the value of each system operating parameter at the current operating point, and measure the response interference power value after each adjustment; Based on the difference between the adjusted response interference power value and the initial interference power value, the gradient vector of the interference power as a function of each system operating parameter is determined. Based on the gradient vector, candidate operating points are determined; Determine whether the candidate operating point meets the preset physical and protocol constraints; If the conditions are met, control instructions are generated to distribute the values ​​of each system operating parameter contained in the candidate operating point to the corresponding system components for execution, and the candidate operating point is used as the target operating point. If the candidate operating point does not meet the requirements, the candidate operating point is corrected to the nearest point that meets the physical and protocol constraints, and control instructions are generated to adjust the values ​​of each system operating parameter to the nearest point, and the nearest point is used as the target operating point.

2. The system according to claim 1, characterized in that, The system operating parameters include one or more of the following: the output frequency of the local oscillator synthesis unit, the transmit power of the millimeter-wave system-on-a-chip (SoC), the receiver gain, and the frequency modulation slope of the linear frequency modulation signal.

3. The system according to claim 2, characterized in that, Based on the difference between the adjusted response interference power value and the initial interference power value, a gradient vector is determined for how the interference power varies with each system operating parameter, including: For each system operating parameter to be optimized at the current operating point, perform the following operations: Keeping other system operating parameters of the current operating point unchanged, the value of the operating parameter of the system to be optimized is increased by a positive disturbance, and the first response disturbance power value is measured. The difference between the first response interference power value and the initial interference power value is divided by the positive disturbance amount to calculate the estimated rate of change of the interference power with respect to the operating parameters of the system to be optimized. Accordingly, after all the operating parameters of the system to be optimized at the current operating point have been traversed, the obtained estimates of the rate of change are combined to obtain the gradient vector of the interference power as a function of each system operating parameter.

4. The system according to claim 3, characterized in that, The difference between the first response interference power value and the initial interference power value is divided by the positive disturbance amount to calculate an estimated rate of change of the interference power with respect to the operating parameters of the system to be optimized, including: Keeping other system operating parameters of the current operating point unchanged, the value of the operating parameter of the system to be optimized is increased by the negative disturbance amount, and the second response disturbance power value is measured. The difference between the first response interference power value and the second response interference power value is divided by twice the disturbance amount to calculate the estimated rate of change of the interference power with respect to the operating parameters of the system to be optimized.

5. The system according to claim 1, characterized in that, The preset physical and protocol constraints include: the operating frequency band of the radio frequency signal after transformation by the frequency conversion layer (20) is in the range of 92-94GHz, and the transmit power of the millimeter-wave system-on-a-chip does not exceed the statutory limit.

6. The system according to claim 1, characterized in that, Based on the gradient vector, candidate operating points are determined, including: The product of the gradient vector and the preset learning rate factor is subtracted from the coordinates of the current working point to obtain the calculation result, and the calculation result is used as the candidate working point.

7. The system according to claim 1, characterized in that, The triggering condition for the cyclic execution of the frequency parameter adaptive adjustment algorithm is: the magnitude of the gradient vector is less than a preset threshold, or the change in the initial interference power value is less than a preset tolerance in a preset number of consecutive iterations.

8. The system according to claim 1, characterized in that, The radar radio frequency system is configured to perform a self-calibration mode. In the self-calibration mode, the baseband processing layer (10) can control the local oscillator synthesis unit and the up mixer group to generate a calibration signal with known power, and send it back to the baseband processing layer (10) via the down mixer group for link gain measurement.

9. The system according to claim 1, characterized in that, The frequency conversion component FCM also includes an integrated temperature sensor.

Images