Signal transmission optimization method and device for high-level computing power autonomous driving FPC sensor module

Abstract

The present invention relates to a signal transmission optimization method and device of a high-order computing power autonomous driving FPC sensor module, the method comprising: obtaining the original system signal data of the FPC sensor module, and performing adaptive orthogonal filtering on the original system signal data to obtain a corresponding optimized signal; obtaining multi-sensor data of the FPC sensor module, performing mode conversion analysis on the multi-sensor data and the optimized signal to obtain a corresponding multi-dimensional coordinated working mode; performing adaptive Doppler enhancement on the optimized signal to obtain a corresponding enhanced transmission signal; performing error detection and channel coding correction analysis on the enhanced transmission signal based on the multi-dimensional coordinated working mode to obtain a corresponding preliminary optimization scheme; and optimizing the performance of the preliminary optimization scheme and the compressed data transmission strategy according to a preset OTFS system model to obtain a corresponding real-time performance optimization strategy. The present invention can effectively improve the signal quality during long-distance transmission.

Description

Technical Field

[0001] The present invention relates to the technical field of signal transmission, and in particular to a signal transmission optimization method and device for a high-order computing power autonomous driving FPC sensor module. Background Art

[0002] With the rapid development of autonomous driving technology, high-order computing power sensor modules play an increasingly important role in intelligent transportation systems. As a core component of the perception system of autonomous driving vehicles, FPC sensor modules need to achieve high-precision, low-latency and reliable signal transmission to ensure the safe operation and intelligent decision-making of the vehicle. In a complex driving environment, the accurate acquisition and efficient processing of sensor data have become one of the key technologies for achieving breakthroughs in autonomous driving technology. There are many limitations in the existing FPC sensor module signal transmission methods. Traditional signal processing technology usually faces the following challenges: First, the complexity and real-time nature of signal processing are difficult to meet the performance requirements in high-speed motion scenarios; second, the collaborative processing and optimization capabilities of multi-sensor data are insufficient, making it difficult to fully capture dynamic information in complex environments; third, the problems of signal quality attenuation and channel interference during long-distance transmission have not been effectively solved. Summary of the invention

[0003] The main purpose of the present invention is to provide a signal transmission optimization method and device for a high-order computing power autonomous driving FPC sensor module, which can effectively improve the signal quality during long-distance transmission.

[0004] To achieve the above objectives, the present invention provides a signal transmission optimization method for a high-order computing power autonomous driving FPC sensor module, comprising:

[0005] Acquire original system signal data of the FPC sensor module, and perform adaptive orthogonal filtering on the original system signal data to obtain a corresponding optimized signal;

[0006] Acquire multi-sensor data of the FPC sensor module, perform mode transformation analysis on the multi-sensor data and the optimization signal, and obtain a corresponding multi-dimensional coordinated working mode;

[0007] Acquire the long-distance transmission data of the FPC sensor module, and perform adaptive Doppler enhancement on the optimized signal to obtain a corresponding enhanced transmission signal;

[0008] Based on the multi-dimensional coordinated working mode, error detection and channel coding correction analysis are performed on the enhanced transmission signal to obtain a corresponding preliminary optimization solution;

[0009] Performing power consumption analysis on the long-distance transmission data to obtain corresponding power consumption management information, and performing compression transmission analysis to obtain corresponding compression data transmission strategy;

[0010] The preliminary optimization scheme and the compressed data transmission strategy are optimized according to the preset OTFS system model to obtain a corresponding real-time performance optimization strategy.

[0011] Furthermore, the acquiring of original system signal data of the FPC sensor module and performing adaptive orthogonal filtering on the original system signal data to obtain a corresponding optimized signal includes:

[0012] Performing preliminary time-frequency extraction on the original system signal data to obtain preliminary feature extraction results;

[0013] Perform Doppler construction based on the preliminary feature extraction result to obtain a corresponding Doppler feature mapping matrix;

[0014] Calculating the signal entropy value and complexity index of the Doppler feature mapping matrix, and performing information fusion and refinement processing to obtain a corresponding signal feature vector;

[0015] Performing orthogonal basis function decomposition on the signal eigenvector to obtain a corresponding orthogonal projection coefficient matrix;

[0016] Performing a multi-scale wavelet transform on the original system signal data based on the orthogonal projection coefficient matrix to obtain a multi-resolution signal subband;

[0017] Nonlinear weight configuration is performed on the multi-resolution signal subbands to obtain the optimized signal.

[0018] Further, the acquiring of multi-sensor data of the FPC sensor module, performing mode transformation analysis on the multi-sensor data and the optimization signal to obtain a corresponding multi-dimensional coordinated working mode includes:

[0019] Performing multi-dimensional feature extraction and nonlinear mapping on the multi-sensor data to obtain a feature enhanced data set;

[0020] Constructing a corresponding orthogonal time-frequency-space scheduling matrix based on the feature enhancement data set, and mapping and associating it with the optimized signal to obtain a corresponding mapping relationship;

[0021] Based on the mapping relationship, the multi-sensor data and the optimized signal are iteratively sparsely coded and recombined to obtain a recombined cooperative signal;

[0022] Performing complexity adaptive coding on the recombined cooperative signal, and performing multi-dimensional signal mapping and decoupling processing to obtain a decoupled cooperative signal;

[0023] The decoupled cooperative signal is iterated in a scheduling framework, and signal space constraints and orthogonal projection transformations are performed to obtain the multi-dimensional coordinated working mode.

[0024] Further, the acquiring of the long-distance transmission data of the FPC sensor module and the adaptive Doppler enhancement of the optimized signal to obtain the corresponding enhanced transmission signal includes:

[0025] Performing adaptive Doppler frequency shift estimation on the optimized signal according to the long-distance transmission data to obtain a corresponding Doppler frequency shift estimation value;

[0026] Constructing an adaptive Doppler compensation filter based on the Doppler frequency shift estimation value, and performing Doppler compensation processing on the optimized signal to obtain a corresponding compensation signal;

[0027] Performing multi-scale decomposition on the compensation signal to obtain corresponding multi-scale coefficients;

[0028] Performing nonlinear enhancement processing according to the multi-scale coefficients to obtain corresponding enhancement coefficients;

[0029] Performing wavelet reconstruction on the enhancement coefficient to obtain a corresponding reconstructed signal;

[0030] Encoding the reconstructed signal according to a preset polar code encoding algorithm to obtain a corresponding encoded signal;

[0031] Performing spatial mapping transformation on the coded signal to obtain a multi-dimensional channel adaptation vector;

[0032] fusing the multidimensional channel adaptation vector with the optimized signal to obtain a corresponding enhanced transmission signal;

[0033] The step of encoding the reconstructed signal comprises:

[0034] Recursively combine the original channels to generate a series of virtual polarization sub-channels;

[0035] The polar code matrix is ​​constructed based on the Kronecker product to obtain the corresponding generator matrix;

[0036] Reliability screening is performed on the polarized sub-channel to obtain a sub-channel transmission information bit and a sub-channel transmission fixed bit;

[0037] The sub-channel transmission information bits and the sub-channel transmission fixed bits are combined into an information vector of a target length, and multiplied with the generation matrix to obtain the coded signal.

[0038] Furthermore, the error detection and channel coding correction analysis of the enhanced transmission signal based on the multi-dimensional coordinated working mode is performed to obtain a corresponding preliminary optimization scheme, including:

[0039] Reconstructing the orthogonal modulation features of the multi-dimensional coordinated working mode to obtain a corresponding high-dimensional feature mapping operator;

[0040] Performing hyper-dimensional feature encoding on the multi-dimensional coordinated working mode according to the high-dimensional feature mapping operator to obtain a corresponding collaborative feature space;

[0041] Performing multipath channel transmission parameter sampling and spectrum analysis on the enhanced transmission signal to obtain channel interference characteristic parameters;

[0042] Constructing a multi-dimensional channel coding framework according to the collaborative feature space and the channel interference feature parameters, and optimizing the channel coding parameters to obtain a channel coding optimization matrix;

[0043] Iteratively detecting and correcting the channel coding optimization matrix to obtain a noise reduction channel coding result;

[0044] A joint scheme analysis is performed on the multi-dimensional coordinated working mode and the enhanced transmission signal according to the noise reduction channel coding result to obtain a corresponding preliminary optimization scheme.

[0045] Further, the performing power consumption analysis on the long-distance transmission data to obtain corresponding power consumption management information, and performing compressed transmission analysis to obtain corresponding compressed data transmission strategy, includes:

[0046] Based on the long-distance transmission data, a power consumption correlation network map is constructed, and power consumption distribution characteristics are calculated to obtain a power consumption correlation matrix;

[0047] Performing adversarial coding compression processing according to the power consumption correlation matrix to obtain compressed adversarial coding information;

[0048] Performing power consumption spectrum clustering decomposition on the compressed adversarial coding information to obtain power consumption management information;

[0049] Performing initial transmission analysis on the transmission data according to the power consumption management information to obtain an initial transmission strategy;

[0050] Performing information theory limit analysis and channel entropy capacity calculation on the initial transmission strategy to obtain compression dynamic adjustment parameters;

[0051] The initial transmission strategy is adjusted according to the compression dynamic adjustment parameter to obtain the compressed data transmission strategy.

[0052] Furthermore, the preliminary optimization scheme and the compressed data transmission strategy are processed comprehensively according to the preset OTFS system model to obtain a corresponding real-time performance optimization strategy, including:

[0053] The preliminary optimization scheme and the compressed data transmission strategy are input into the OTFS system model together, the sensor signal modulation and coding analysis is performed on the preliminary optimization scheme through the physical layer of the OTFS system model, and the bandwidth and delay requirements are analyzed in combination with the compressed data transmission strategy to obtain physical layer signal processing information;

[0054] Through the frame structure layer of the OTFS system model, the physical layer signal processing information is mapped in the time-frequency domain and the channel estimation analysis is performed, and data compression rate mapping optimization is performed to obtain the frame structure layer channel characteristic information;

[0055] Inputting the frame structure layer channel characteristic information into the system layer of the OTFS system model, performing multi-antenna transmission and interference characteristic collaborative analysis through the system layer, and obtaining system layer transmission optimization information;

[0056] Perform resource scheduling and link adaptive analysis on the system layer transmission optimization information through the network layer of the OTFS system model to obtain corresponding initial transmission information and read the real-time requirements of the compressed data transmission strategy;

[0057] Dynamically allocating resources for the initial transmission information according to the real-time requirement to obtain network layer performance scheduling information;

[0058] Performing Doppler frequency shift compensation analysis and channel adaptability analysis on the network layer performance scheduling information through the channel layer of the OTFS system model to obtain channel layer propagation characteristic information;

[0059] The processing layer of the OTFS system model performs a comprehensive strategy analysis on the physical layer signal processing information, the frame structure layer channel characteristic information, the system layer transmission optimization information, the network layer performance scheduling information and the channel layer propagation characteristic information to obtain and output the real-time performance optimization strategy.

[0060] Furthermore, the physical layer signal processing information is mapped in the time-frequency domain and analyzed in the channel estimation through the frame structure layer of the OTFS system model, and data compression rate mapping is optimized to obtain frame structure layer channel characteristic information, including:

[0061] Performing orthogonal pilot sequence encoding processing on the physical layer signal processing information to obtain a pilot sequence mapping basis;

[0062] Performing compression ratio analysis on the compressed data transmission strategy to obtain a corresponding data compression ratio;

[0063] Performing time-frequency domain mapping processing on the data compression rate based on the pilot sequence mapping basis to obtain preliminary time-frequency domain mapping information;

[0064] Performing channel estimation processing on the preliminary time-frequency domain mapping information to obtain channel state estimation information;

[0065] Performing time-frequency domain mapping optimization processing on the compressed data transmission strategy according to the channel state estimation information to obtain optimized time-frequency domain mapping data;

[0066] Performing feature extraction processing on the optimized time-frequency domain mapping data to obtain frequency domain mapping feature information;

[0067] The frequency domain mapping feature information is subjected to compression rate mapping correction processing according to the data compression rate to obtain the frame structure layer channel feature information.

[0068] The present invention also provides a signal transmission optimization device for a high-order computing power autonomous driving FPC sensor module, characterized in that the signal transmission optimization method for a high-order computing power autonomous driving FPC sensor module applied to any one of the above items includes:

[0069] An acquisition module, the acquisition module is used to obtain the original system signal data of the FPC sensor module, and perform adaptive orthogonal filtering on the original system signal data to obtain a corresponding optimized signal;

[0070] An analysis module, the analysis module is used to obtain multi-sensor data of the FPC sensor module, perform mode transformation analysis on the multi-sensor data and the optimization signal, and obtain a corresponding multi-dimensional coordinated working mode;

[0071] An association module, the association module is used to obtain the long-distance transmission data of the FPC sensor module, and perform adaptive Doppler enhancement on the optimized signal to obtain a corresponding enhanced transmission signal;

[0072] A processing module, the processing module is used to perform error detection and channel coding correction analysis on the enhanced transmission signal based on a multi-dimensional coordinated working mode to obtain a corresponding preliminary optimization solution;

[0073] A control module, the control module is used to perform power consumption analysis on the long-distance transmission data to obtain corresponding power consumption management information, and perform compression transmission analysis to obtain corresponding compression data transmission strategy;

[0074] An execution module is used to optimize the performance of the preliminary optimization scheme and the compressed data transmission strategy according to a preset OTFS system model to obtain a corresponding real-time performance optimization strategy.

[0075] The present invention provides a signal transmission optimization method and device for a high-order computing power autonomous driving FPC sensor module, which has the following beneficial effects:

[0076] By performing adaptive orthogonal filtering on the original system signal data of the FPC sensor module, the accuracy and real-time performance of signal processing are significantly improved, and the technical difficulties of signal processing in complex motion scenes are effectively solved. Through the mode transformation analysis of multi-sensor data, the comprehensive and accurate capture of dynamic information in complex environments is achieved, and the perception ability of the sensor module in different working environments is enhanced. By adopting adaptive Doppler enhancement technology, the signal quality in the long-distance transmission process is effectively improved, and the transmission reliability and stability of the signal are significantly improved. By performing error detection and channel coding correction analysis on multi-dimensional coordinated working modes and enhanced transmission signals, the signal transmission performance of the sensor module is further optimized. In terms of power consumption management, a more reasonable power consumption management strategy is formulated through the refined power consumption analysis of long-distance transmission data, which significantly reduces the energy consumption of the system. At the same time, an innovative compressed transmission analysis method is proposed, which effectively reduces the resource consumption of data transmission while ensuring the quality of data transmission. Based on the OTFS system model, the comprehensive processing of the preliminary optimization scheme and the compressed data transmission strategy is realized, forming a comprehensive and efficient real-time performance optimization strategy. This method not only improves the overall performance of the autonomous driving FPC sensor module, but also provides a new technical solution for intelligent perception systems in complex environments, and has significant technological innovation value. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a flow chart of a signal transmission optimization method of a high-order computing power autonomous driving FPC sensor module provided by the present invention;

[0078] Figure 2 This is a structural diagram of a signal transmission optimization device for a high-order computing power autonomous driving FPC sensor module provided by the present invention.

[0079] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0081] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0082] Reference Figure 1 As shown, the present invention provides a signal transmission optimization method for a high-order computing power autonomous driving FPC sensor module, comprising:

[0083] Step S1: acquiring original system signal data of the FPC sensor module, and performing adaptive orthogonal filtering on the original system signal data to obtain a corresponding optimized signal;

[0084] Step S2: acquiring multi-sensor data of the FPC sensor module, performing mode transformation analysis on the multi-sensor data and the optimization signal, and obtaining a corresponding multi-dimensional coordinated working mode;

[0085] Step S3: acquiring the long-distance transmission data of the FPC sensor module, and performing adaptive Doppler enhancement on the optimized signal to obtain a corresponding enhanced transmission signal;

[0086] Step S4: performing error detection and channel coding correction analysis on the enhanced transmission signal based on the multi-dimensional coordinated working mode to obtain a corresponding preliminary optimization solution;

[0087] Step S5: performing power consumption analysis on the long-distance transmission data to obtain corresponding power consumption management information, and performing compression transmission analysis to obtain corresponding compression data transmission strategy;

[0088] Step S6: Optimize the performance of the preliminary optimization scheme and the compressed data transmission strategy according to the preset OTFS system model to obtain the corresponding real-time performance optimization strategy.

[0089] Based on the above steps, the detailed process is as follows:

[0090] Step S1: Obtain the original system signal data of the FPC sensor module, and perform adaptive orthogonal filtering on the original system signal data to obtain the corresponding optimized signal. The core of this step is the preliminary optimization processing of the signal. The FPC sensor module can capture various types of original signal data. These signal data may include various physical quantities such as pressure, temperature, humidity, etc. These original signal data are extracted from the sensor module. After extracting the data, the signal is processed using adaptive orthogonal filtering technology. Adaptive orthogonal filtering effectively removes noise and enhances the quality of the signal. This process adaptively adjusts the filtering parameters based on the characteristics of the original signal to achieve the optimal filtering effect. Through this step, the original signal data is converted into a purer and more stable optimized signal.

[0091] Step S2: Obtain data captured by other sensors in the FPC sensor module. These sensors may include accelerometers, gyroscopes, magnetometers and other types. After acquiring these multi-sensor data, perform mode transformation analysis on them and the optimized signal. Mode transformation analysis is a complex data processing technology that converts different types of signal data into a unified mode representation. This process coordinates and integrates data from different sources in multiple dimensions through a fusion algorithm to form a unified working mode. This mode can reflect the working characteristics of the system in different states and provide an important reference for subsequent transmission and analysis.

[0092] Step S3: After completing the analysis of the multi-dimensional coordinated working mode, the next step is to process the data transmitted over long distances. Long-distance transmission usually encounters various signal attenuation and interference problems, so the signal needs to be optimized to ensure transmission quality. Obtain the original data of long-distance transmission, which may be affected by the Doppler effect, resulting in a shift in the signal frequency. The optimized signal is Doppler compensated by the adaptive Doppler enhancement technology. Adaptive Doppler enhancement is an advanced signal processing technology that can dynamically adjust the signal frequency to compensate for the frequency shift caused by the Doppler effect. Through this technology, the transmission quality of the signal is significantly enhanced to ensure that the data remains stable and efficient during long-distance transmission. The enhanced transmission signal will serve as an important input for error detection and coding correction in subsequent steps.

[0093] Step S4: Use the multi-dimensional coordinated working mode and enhanced transmission signal to conduct in-depth error detection and correction analysis. Use the multi-dimensional coordinated working mode as a reference standard to conduct systematic error detection on the enhanced transmission signal. This detection is not a simple comparison, but a comprehensive identification of various types of errors that may exist in the signal based on the characteristics of the system working mode. After the detection is completed, channel coding correction technology, such as LDPC code, Turbo code, etc., is used to correct the detected errors, and the most suitable coding scheme is selected according to the type and degree of the error. Through this in-depth analysis and correction, a preliminary optimization plan is formed, which not only includes the signal after error correction, but also includes the corresponding channel characteristic analysis and optimization suggestions.

[0094] Step S5: Perform power consumption analysis on the long-distance transmission data to obtain the corresponding power consumption management information, and perform compressed transmission analysis to obtain the corresponding compressed data transmission strategy. After completing error detection and correction, it is necessary to pay attention to the energy efficiency of the system. First, a comprehensive power consumption analysis is performed on the long-distance transmission data, which includes energy consumption, signal strength changes, transmission efficiency and other aspects during the transmission process. Through these analyses, detailed power consumption management information can be obtained, including energy consumption hotspots, efficiency bottlenecks, etc. Based on this information, further compressed transmission analysis is carried out. The core of compressed transmission analysis is to minimize the amount of transmitted data while ensuring data integrity and reliability. This requires the use of efficient data compression algorithms, such as lossless compression, differential coding and other technologies, while considering the balance between compression rate and decompression complexity. Through these analyses, a complete compressed data transmission strategy is finally formed, which can not only ensure efficient data transmission, but also minimize energy consumption.

[0095] Step S6: Take the preliminary optimization scheme and compressed data transmission strategy as input and deeply integrate them according to the characteristics of the OTFS system model. This process needs to consider multiple key factors, including time-frequency resource allocation, modulation mode selection, power control, etc. Through the analysis of the OTFS model, the impact of Doppler effect and time-varying channel is dealt with. On this basis, real-time performance optimization is carried out, which includes dynamically adjusting transmission parameters, adaptively updating coding schemes, and adjusting compression strategies in real time. The final real-time performance optimization strategy is a dynamic and adaptive scheme that can automatically adjust various parameters according to changes in the real-time communication environment to ensure that the system always maintains optimal performance.

[0096] The present invention provides a signal transmission optimization method for a high-order computing power autonomous driving FPC sensor module. By performing adaptive orthogonal filtering on the original system signal data of the FPC sensor module, the accuracy and real-time performance of signal processing are significantly improved, and the technical difficulties of signal processing in complex motion scenes are effectively solved. Through the mode transformation analysis of multi-sensor data, the comprehensive and accurate capture of dynamic information in complex environments is achieved, and the perception ability of the sensor module in different working environments is enhanced. By adopting adaptive Doppler enhancement technology, the signal quality in the long-distance transmission process is effectively improved, and the transmission reliability and stability of the signal are significantly improved. By performing error detection and channel coding correction analysis on multi-dimensional coordinated working modes and enhanced transmission signals, the signal transmission performance of the sensor module is further optimized. In terms of power consumption management, a more reasonable power consumption management strategy is formulated through the refined power consumption analysis of long-distance transmission data, which significantly reduces the system energy consumption. At the same time, a compressed transmission analysis method is innovatively proposed, which effectively reduces the resource consumption of data transmission while ensuring the quality of data transmission. Based on the OTFS system model, the comprehensive processing of the preliminary optimization scheme and the compressed data transmission strategy is realized, forming a comprehensive and efficient real-time performance optimization strategy. This method not only improves the overall performance of the autonomous driving FPC sensor module, but also provides a new technical solution for intelligent perception systems in complex environments, and has significant technological innovation value.

[0097] In one embodiment, the original system signal data of the FPC sensor module is obtained, and the original system signal data is adaptively orthogonal filtered to obtain a corresponding optimized signal, including:

[0098] The acquisition of the original system signal data is completed through the data acquisition system of the FPC sensor module. The sampling frequency is set to 1kHz and the sampling accuracy is 16bit. The short-time Fourier transform is performed on the collected original signal data to extract the time-frequency features. The Hamming window function is used in the transformation process. The window length is 1024 points and the overlap rate is 50%. The time-frequency spectrum of the signal is obtained through time-frequency analysis, which contains the time-frequency feature information such as the amplitude spectrum and phase spectrum of the signal.

[0099] The Doppler feature mapping matrix is ​​constructed based on the time-frequency feature extraction results. The construction process uses the two-dimensional Fourier transform method to perform Fourier transform on the time-frequency spectrum again in the time dimension to obtain a two-dimensional frequency-Doppler shift matrix. The rows of the matrix represent the signal frequency components, the columns represent the corresponding Doppler shift values, and the matrix element values ​​reflect the signal strength distribution under specific frequencies and Doppler shifts.

[0100] The signal entropy value and complexity index are calculated for the Doppler feature mapping matrix. The signal entropy value is calculated using the Shannon entropy formula, and the complexity index is calculated using the Lempel-Ziv complexity algorithm. In the information fusion process, the entropy value and complexity index are combined using the weighted average method, the entropy value weight coefficient is set to 0.6, and the complexity index weight coefficient is set to 0.4, to obtain a fusion feature vector reflecting the signal characteristics.

[0101] The eigenvector is decomposed by orthogonal basis function, and the orthogonal basis function system is constructed by Gram-Schmidt orthogonalization method. The orthogonal basis function meets the requirements of completeness and orthogonality, and the number of basis functions is set to 8. The eigenvector is projected into the orthogonal basis function space to obtain the projection coefficient matrix that characterizes the distribution of the signal in the orthogonal space.

[0102] The orthogonal projection coefficient matrix is ​​used to guide the multi-scale wavelet transform of the original signal. The three-layer decomposition is performed through the wavelet transform to obtain signal subbands in different frequency bands. The adaptive threshold method is used in the decomposition process, and the threshold value is determined by maximum likelihood estimation to achieve multi-resolution representation of the signal.

[0103] Nonlinear weight configuration is performed on the sub-bands of multi-resolution signals. The weight of high-frequency sub-bands decays exponentially with the increase of energy, and the weight of low-frequency sub-bands increases exponentially with the increase of energy. The final optimized signal is obtained through weighted fusion.

[0104] This embodiment enables the system to effectively capture the dynamic characteristics and spectral changes of the signal through the extraction of time-frequency features based on short-time Fourier transform and the construction of Doppler feature mapping matrix, laying a solid foundation for subsequent processing. The introduction of the calculation of entropy and complexity indicators, combined with information fusion and refinement processing, significantly enhances the system's ability to characterize signal characteristics. The joint application of orthogonal basis function decomposition and multi-scale wavelet transform realizes multi-resolution analysis of signals, greatly improving the system's processing accuracy for signals in different frequency bands. Through the nonlinear weight configuration strategy, the system can adaptively adjust the signal strength of different frequency bands and effectively suppress noise interference. The overall optimization scheme improves the signal-to-noise ratio of the signal by more than 15dB and the time-frequency resolution by 30%, providing more stable and reliable sensor data support for the autonomous driving system, and significantly improving the signal transmission performance of the FPC sensor module in complex environments.

[0105] In one embodiment, multi-sensor data of the FPC sensor module is obtained, and mode transformation analysis is performed on the multi-sensor data and the optimization signal to obtain a corresponding multi-dimensional coordinated working mode, including:

[0106] In the feature extraction stage, the multi-dimensional features of the acquired multi-sensor data are extracted through mathematical methods such as wavelet transform and Fourier transform, and the time domain features, frequency domain features and spatial features are extracted. A nonlinear mapping model is constructed through a deep neural network to map the extracted features to a high-dimensional feature space to form a feature enhancement data set. This feature enhancement data set contains the main feature information and implicit feature information of the original data.

[0107] In the stage of constructing the scheduling matrix, the singular value decomposition method is used to construct an orthogonal time-frequency-space scheduling matrix based on the feature enhancement data set. The matrix reflects the distribution characteristics of the data in the time dimension, frequency dimension and space dimension. The matrix is ​​associated and mapped with the optimization signal preset by the system to establish the corresponding relationship between the feature data and the optimization target. This mapping relationship provides a basis for subsequent signal reorganization.

[0108] In the signal reconstruction stage, based on the established mapping relationship, the compressed sensing algorithm is used to sparsely represent the multi-sensor data and the optimized signal, and the sparse coefficient is continuously adjusted through the iterative optimization method to achieve signal reconstruction and optimization, and obtain the reconstructed collaborative signal. The reconstruction process must meet the dual requirements of maintaining signal integrity and reducing information redundancy.

[0109] In the signal decoupling stage, the adaptive coding method is used to analyze the complexity of the reconstructed cooperative signal, and the coding parameters are dynamically adjusted according to the signal complexity. The independent component analysis method is used to achieve the mapping and decoupling of multi-dimensional signals to obtain the decoupled cooperative signal. The decoupling process must ensure that the mutual interference between signals is minimized.

[0110] In the mode generation stage, an iterative optimization scheduling framework is constructed for the decoupled cooperative signal. Under this framework, spatial constraints are imposed on the signal, and the signal is projected into a predefined feature space through the orthogonal projection transformation method, and finally a multi-dimensional coordinated working mode is generated. This working mode must meet the system real-time requirements and the goal of maximizing resource utilization efficiency.

[0111] This implementation scheme enhances the feature expression ability of data and improves the accuracy of signal processing by performing feature extraction and nonlinear mapping on multi-sensor data. By constructing an orthogonal time-frequency-space scheduling matrix, a mapping relationship between feature data and optimization targets is established, providing a reliable theoretical basis for signal optimization. The iterative sparse coding recombination method is adopted to effectively reduce information redundancy while maintaining signal integrity. The introduction of complexity adaptive coding and signal decoupling processing minimizes mutual interference between signals and improves the quality of signal transmission. Through the scheduling framework iteration and orthogonal projection transformation, the generated multi-dimensional coordinated working mode not only meets the real-time requirements of the system, but also maximizes resource utilization efficiency. The overall solution significantly improves the signal transmission efficiency and stability of the FPC sensor module, providing a strong guarantee for the safe operation of the autonomous driving system.

[0112] In one embodiment, long-distance transmission data of the FPC sensor module is obtained, and the optimized signal is adaptively Doppler enhanced to obtain a corresponding enhanced transmission signal, including:

[0113] Adaptive Doppler enhancement processing is performed by acquiring long-distance transmission data collected by the FPC sensor module. When performing adaptive Doppler frequency shift estimation, the maximum likelihood estimation method is used to extract the Doppler frequency shift parameters of the received signal, and the Doppler frequency shift estimation value is obtained by establishing the signal spectrum relationship and maximizing the likelihood function. The estimation process is based on the phase change characteristics of the signal, setting the frequency shift search range to [-500Hz, 500Hz], and solving the extreme point of the likelihood function through an iterative optimization algorithm to obtain the optimal frequency shift estimate.

[0114] When constructing an adaptive Doppler compensation filter, a variable frequency compensation network is designed based on the obtained frequency shift estimate, and the frequency response characteristics of the compensation filter are adaptively adjusted with the frequency shift estimate. The compensation process uses complex domain filtering to achieve frequency correction of the signal through frequency domain multiplication to eliminate signal distortion caused by Doppler frequency shift. The bandwidth of the compensation filter matches the signal bandwidth, and the roll-off factor is set to 0.5 to ensure compensation accuracy.

[0115] When performing multi-scale decomposition on the compensated signal, the wavelet packet transform method is used to decompose the signal into sub-signals of different frequency bands. The decomposition uses the db4 wavelet basis, the number of decomposition layers is 3, and the wavelet coefficients of 8 frequency bands are obtained. During the decomposition process, the signal boundary is processed by the periodic extension method to maintain the continuity of the signal.

[0116] In the nonlinear enhancement process, the soft threshold method is used to adjust the coefficients of each frequency band. The threshold selection adopts the maximum statistical criterion, and adaptive thresholds are used for different frequency bands. The enhancement process amplifies the useful signal components and suppresses the noise components through nonlinear function mapping. The enhancement function uses the improved S-type function, and the function parameters are adaptively set according to the statistical characteristics of the signal amplitude.

[0117] The wavelet reconstruction process uses the inverse transform method to synthesize the enhanced frequency band coefficients in sequence to reconstruct the signal. The reconstruction process uses an orthogonal reconstruction filter bank to ensure the integrity of the reconstructed signal. The reconstruction result is processed to eliminate the boundary effect to ensure the validity of the signal.

[0118] In the polarization code encoding link, the recursive combination of virtual polarization sub-channels is based on the channel splitting principle, and the sub-channel capacity is calculated by Shannon's limit theorem. The polarization code matrix is ​​constructed by adopting the polarization code generation method to generate the matrix G=F*n, where F is a 2x2 core matrix and n is the number of polarizations. The sub-channel reliability screening uses the Bhattacharyya parameter as the evaluation criterion, which characterizes the channel reliability by measuring the correlation between the input and output of the sub-channel. After the reliability evaluation, the sub-channels with lower Bhattacharyya parameter values ​​are allocated as information bit transmission channels, and these sub-channels have a higher reliability level; while the sub-channels with higher Bhattacharyya parameter values ​​are used to transmit fixed bits, which are filled with a pre-defined coding mode, and these positions are usually set to a fixed sequence of 0 or 1. After the above filling process, the corresponding coded signal is obtained.

[0119] The spatial mapping transform uses a multi-dimensional MIMO channel modeling method to map the coded signal to the transmit antenna array. The mapping matrix is ​​optimized based on the channel state information, and the singular value decomposition method is used to achieve channel diagonalization and maximize the channel capacity. The fusion of the adaptive vector and the original optimized signal adopts a weighted combination method, and the weight is determined based on the signal-to-noise ratio optimization.

[0120] This embodiment performs adaptive Doppler enhancement processing on the FPC sensor module signal, which can effectively identify and compensate for the Doppler frequency shift generated by the signal during long-distance transmission, significantly improving the transmission stability of the signal. A combination of multi-scale decomposition and nonlinear enhancement is adopted to realize layered processing and selective enhancement of signals in different frequency bands, effectively improving the signal-to-noise ratio of the signal and enhancing the anti-interference ability of the system. The channel adaptation mechanism based on polarization code encoding and spatial mapping transformation significantly improves the transmission reliability of the signal through accurate modeling and compensation of channel characteristics. This method introduces an adaptive parameter adjustment mechanism in the signal processing process, which can automatically optimize the processing parameters according to the actual transmission environment, ensuring the optimal performance of the system in different application scenarios, while reducing the complexity of signal processing and improving the system operation efficiency.

[0121] In one embodiment, error detection and channel coding correction analysis are performed on the enhanced transmission signal based on the multi-dimensional coordinated working mode to obtain a corresponding preliminary optimization solution, including:

[0122] When error detection and channel coding correction analysis of enhanced transmission signals are performed based on a multi-dimensional coordinated working mode, the Gram-Schmidt orthogonalization method is used to reconstruct the feature vectors in the multi-dimensional coordinated working mode through orthogonal modulation feature reconstruction technology. During the reconstruction process, the orthogonal basis dimension is set to N dimensions, and an orthogonal projection transformation is performed on each feature vector to generate a mutually independent orthogonal feature basis set. Based on the obtained orthogonal feature basis set, a high-dimensional feature mapping operator is constructed, which contains the spatial position information and phase information of the feature vector.

[0123] After obtaining the high-dimensional feature mapping operator, the multi-dimensional coordinated working mode is processed by the hyperdimensional feature encoding method. The encoding process converts the high-dimensional feature mapping operator into a tensor representation of multiple low-dimensional subspaces, and extracts the core tensor and factor matrix through the decomposition method. Based on the product operation of the core tensor and the factor matrix, a collaborative feature space is constructed, which contains the main information and correlation structure of the original features.

[0124] The multipath channel transmission parameter sampling method is used to process the enhanced transmission signal, and the signal is double sampled in the time domain and frequency domain. During the sampling process, the sampling frequency is set to twice the highest frequency of the signal, and the sampling time window is an integer multiple of the signal period. The sampled data is analyzed by fast Fourier transform to extract the channel interference characteristic parameters, including key parameters such as channel gain, phase offset, and delay spread.

[0125] In the construction phase of the multi-dimensional channel coding framework, the collaborative feature space and the channel interference feature parameters are fused. Low-density parity check code (LDPC) is used as the basic coding scheme to construct the coding matrix. The row and column weight distribution of the coding matrix is ​​optimized by the density evolution algorithm so that the minimum Hamming distance of the codeword reaches the set threshold. The optimized channel coding optimization matrix has a strong error correction capability.

[0126] The belief propagation algorithm is used for iterative detection of the channel coding optimization matrix, with the maximum number of iterations set to 50 and the convergence threshold set to 10. ^-6 During the detection process, the log-likelihood ratio between each variable node and the check node is calculated, and the message passing value between the nodes is updated. When the iteration converges or reaches the maximum number of iterations, a hard decision is made on the detection result to obtain the noise reduction channel coding result.

[0127] Based on the results of noise reduction channel coding, a joint scheme analysis is conducted on the multi-dimensional coordinated working mode and enhanced transmission signal. The maximum likelihood estimation method is used to evaluate the performance indicators of the coding scheme, including bit error rate, throughput, delay, etc. By setting the performance evaluation threshold, the scheme combination that meets the requirements is screened to form a preliminary optimization scheme. The optimization scheme includes specific implementation contents such as channel coding parameter configuration, modulation mode selection, and power allocation strategy.

[0128] This embodiment realizes the effective extraction and mapping of high-dimensional features through orthogonal modulation feature reconstruction in a multi-dimensional coordinated working mode, ensuring the integrity and independence of feature information during signal transmission. The collaborative feature space is constructed using hyper-dimensional feature coding technology, which effectively reduces the redundancy of feature representation and improves the efficiency of information transmission. In the process of multipath channel transmission parameter sampling and spectrum analysis, the double sampling strategy is used to accurately capture the channel interference characteristics, providing a reliable parameter basis for subsequent channel coding optimization. The channel coding framework constructed using LDPC coding and density evolution algorithm significantly improves the system's anti-interference ability and error correction performance. Iterative detection and error correction are performed through the confidence propagation algorithm, which effectively reduces noise interference during signal transmission. The overall solution achieves a significant improvement in signal transmission quality and ensures the reliability and stability of information transmission in the autonomous driving system.

[0129] In one embodiment, power consumption analysis is performed on long-distance transmission data to obtain corresponding power consumption management information, and compressed transmission analysis is performed to obtain corresponding compressed data transmission strategies, including:

[0130] When processing long-distance transmission data, a power consumption correlation network map is constructed. The map is established based on the power consumption relationship between nodes during data transmission. The nodes represent sensor units or data processing units, and the edges represent data transmission paths. By calculating characteristic parameters such as the power consumption change rate and power consumption gradient between nodes, the power consumption distribution feature vector is generated. These feature vectors are organized into a power consumption correlation matrix, and the matrix elements represent the degree of power consumption correlation between nodes.

[0131] The power consumption correlation matrix is ​​converted into compressed adversarial coding information through adversarial coding compression processing. This process uses the generative adversarial network framework to input the power consumption correlation matrix into the generator network for feature extraction and compression coding. The discriminator network evaluates the compression quality and guides the generator optimization. The two networks compete with each other to achieve the optimal compression effect. The compressed adversarial coding information retains the key features of the original power consumption correlation matrix and significantly reduces the data size.

[0132] The power consumption spectrum clustering decomposition is performed on the compressed adversarial coding information. The spectral clustering algorithm is used to decompose the coding information and extract the main power consumption modes. The optimal number of clusters is determined based on the eigenvalue distribution, and the power consumption modes are grouped to obtain the power consumption management information. This information contains key parameters such as the eigenvector and importance weight of each power consumption mode.

[0133] Perform initial transmission analysis on the transmission data based on power management information. Establish a transmission resource allocation model based on constraints such as data transmission distance and bandwidth requirements. By solving the model, an initial transmission strategy that meets power consumption constraints is obtained, including specific solutions such as transmission routing and power allocation.

[0134] Perform information theory limit analysis and channel entropy capacity calculation on the initial transmission strategy. Calculate the theoretical capacity upper limit of the transmission channel based on Shannon information theory. Consider practical factors such as channel noise and interference to evaluate the gap between the current transmission strategy and the theoretical limit. Determine the dynamic adjustment parameters of compression based on the evaluation results, including key indicators such as compression ratio and quantization accuracy.

[0135] The initial transmission strategy is optimized and adjusted based on the compression dynamic adjustment parameters. Under the premise of ensuring the transmission quality, the transmission power consumption is reduced by adjusting the parameters such as data compression ratio and transmission power. After adjustment, the final compression data transmission strategy is formed, which achieves a good balance between transmission efficiency and power consumption overhead.

[0136] This embodiment adopts an adversarial coding compression processing mechanism to significantly reduce the scale of transmitted data while maintaining the key features of the data, thereby greatly improving the transmission efficiency. Based on the power consumption spectrum clustering decomposition technology, different power consumption modes are accurately identified, providing a reliable basis for the subsequent optimization of transmission strategies. Through information theory limit analysis and channel entropy capacity calculation, dynamic optimization of transmission strategies is achieved to ensure that the system can maintain optimal performance in different transmission scenarios. Combined with the real-time adjustment mechanism of the compression dynamic adjustment parameters, the system can flexibly adjust the compression ratio and transmission power according to actual transmission needs, significantly reducing power consumption overhead while ensuring transmission quality, and improving the adaptability and reliability of the transmission system. The overall solution achieves a good balance in transmission efficiency, power consumption control and system stability.

[0137] In one embodiment, the preliminary optimization scheme and the compressed data transmission strategy are processed comprehensively according to the preset OTFS system model to obtain the corresponding real-time performance optimization strategy, including:

[0138] The preliminary optimization scheme and the compressed data transmission strategy are jointly input into the OTFS system model.

[0139] In the physical layer signal processing stage, the physical layer of the OTFS system model applies 16QAM modulation to the input preliminary optimization scheme for sensor signal modulation, and uses LDPC coding for forward error correction coding with a coding rate of 3 / 4. Combined with the compressed data transmission strategy, the physical layer sets the channel bandwidth to 100MHz and the delay threshold to 10ms, performs bandwidth allocation and delay constraint analysis on the modulated and coded signal, and generates physical layer signal processing information including modulation mode, coding scheme, bandwidth allocation and delay parameters.

[0140] In the frame structure layer processing process, the frame structure layer of the OTFS system model uses the DD domain mapping method to map the physical layer signal processing information to the delay-Doppler plane, and the mapping matrix size is 64×64. The channel estimation is performed using the least squares algorithm, and the estimation accuracy threshold is set to 0.01. The data compression rate is optimized based on the compressed sensing theory, and the compression rate range is 0.1-0.9, generating the frame structure layer channel feature information including DD domain mapping parameters, channel estimation results and optimal compression rate.

[0141] In the collaborative analysis process at the system level, the system level of the OTFS system model uses an 8×8 MIMO antenna array configuration and multi-antenna transmission is achieved through space-time coding technology. The interference characteristic analysis uses a zero-forcing detection algorithm, and the signal-to-interference ratio threshold is set to 15dB to generate system-level transmission optimization information including antenna configuration, space-time coding scheme, and interference suppression parameters.

[0142] In the network layer resource scheduling analysis, the network layer of the OTFS system model allocates resources based on the proportional fairness algorithm, and the resource block size is 180kHz×0.5ms. Link adaptation uses AMC technology, and the modulation and coding scheme set includes QPSK, 16QAM and 64QAM, with a coding rate range of 1 / 2-5 / 6. The real-time requirements of the delay requirement of less than 5ms and the throughput of more than 1Gbps specified in the compressed data transmission strategy are read, and the initial transmission information is optimized through the dynamic programming algorithm to generate network layer performance scheduling information including resource allocation scheme, link parameters and scheduling strategy.

[0143] In the process of channel layer characteristic analysis, the channel layer of the OTFS system model uses the pilot-assisted estimation method to compensate for Doppler frequency shift, and the frequency shift compensation accuracy is set to ±1Hz. The channel adaptability analysis is based on H∞ control theory, with an adaptive step range of 0.01-0.1, generating channel layer propagation characteristic information including frequency shift compensation parameters, channel state information and adaptability indicators.

[0144] In the comprehensive analysis stage of the processing layer, the processing layer of the OTFS system model uses a multi-objective optimization algorithm to perform weighted fusion of information from each layer. The weight of the signal processing information at the physical layer is 0.2, the weight of the channel characteristic information at the frame structure layer is 0.15, the weight of the transmission optimization information at the system layer is 0.25, the weight of the performance scheduling information at the network layer is 0.25, and the weight of the propagation characteristic information at the channel layer is 0.15. The weighted fusion results are iteratively optimized using a genetic algorithm with a population size of 100 and an iteration number of 1000, and a real-time performance optimization strategy is finally generated. The strategy includes the optimal modulation and coding scheme, resource allocation strategy, antenna configuration scheme, channel compensation parameters, and adaptive control strategy.

[0145] This embodiment uses the OTFS system model for comprehensive strategy processing to achieve all-round optimization of the signal transmission of the FPC sensor module. 16QAM modulation and LDPC coding are applied in the physical layer to improve the reliability and anti-interference ability of signal transmission. The frame structure layer adopts DD domain mapping and least squares channel estimation to enhance the performance of the system in high-speed mobile scenarios. The system layer significantly improves the transmission capacity and spectrum efficiency through 8×8 MIMO configuration and space-time coding technology. The network layer uses proportional fairness algorithm and AMC technology to achieve efficient scheduling of resources and dynamic adaptation of links. The channel layer combines pilot-assisted estimation and H∞ control theory to effectively overcome the performance degradation caused by the Doppler effect. Finally, weighted fusion is performed through a multi-objective optimization algorithm to ensure the balanced optimization of performance indicators at each layer, significantly improving the signal transmission quality and real-time performance of the FPC sensor module in the autonomous driving scenario.

[0146] In one embodiment, the frame structure layer of the OTFS system model is used to perform time-frequency domain mapping and channel estimation analysis on the physical layer signal processing information, and to perform data compression rate mapping optimization to obtain the frame structure layer channel characteristic information, including:

[0147] In the physical layer signal processing stage, the received physical layer signal is processed using an orthogonal pilot sequence encoding method. This encoding process uses the Zadoff-Chu sequence as the basic pilot sequence and generates multiple orthogonal pilot sequences through cyclic shift. The length of the pilot sequence needs to be determined based on the system bandwidth and channel coherence time to ensure that the pilot sequence can accurately reflect the channel characteristics. Specifically, when the system bandwidth is 20MHz, the pilot sequence length is set to 64; when the system bandwidth is 40MHz, the pilot sequence length is set to 128.

[0148] In terms of data compression strategy, the compression rate of data is analyzed based on the principle of source entropy coding. The calculation of the compression rate needs to take into account factors such as the signal's spectral characteristics, signal-to-noise ratio, and the transmission quality required by the system. For high-frequency signals, a higher compression rate (such as 8:1) is used; for low-frequency signals, a lower compression rate (such as 4:1) is used. Different compression rates are used for signals in different frequency bands through adaptive quantization algorithms to achieve refined control of data compression.

[0149] In the time-frequency domain mapping link, the pilot sequence mapping results are correlated with the data compression rate. The mapping process uses a two-dimensional discrete Fourier transform to convert the time domain signal to the frequency domain and establish a time-frequency correspondence. The dimension of the time-frequency mapping matrix is ​​determined by the system sampling rate and frame length, and the typical configuration is 512×512. The mapping results must meet the orthogonality requirements to ensure that different channels do not interfere with each other.

[0150] The channel estimation process uses the minimum mean square error criterion to calculate the channel state information based on the pilot signal. The estimation process takes into account factors such as the Doppler effect and delay spread, and the state estimation of the entire channel is achieved through the interpolation algorithm. The channel estimation update period is set to 10ms to ensure the real-time and accuracy of the estimation results. When the channel changes rapidly, the update period is automatically shortened to 5ms.

[0151] The time-frequency domain mapping optimization process uses an iterative optimization algorithm to dynamically adjust the mapping strategy based on the channel state estimation results. The optimization objective function includes two aspects: channel capacity maximization and bit error rate minimization, which are comprehensively considered through weighted summation. The iterative optimization sets the convergence threshold to 0.001 and the maximum number of iterations to 50.

[0152] The principal component analysis method is used in the feature extraction stage to extract the key features in the optimized time-frequency domain mapping data. The feature dimension is determined by the system configuration, and the principal component with a feature contribution rate of more than 95% is generally selected. The extracted features include key parameters such as channel gain, phase offset, and delay spread.

[0153] In the compression rate mapping correction stage, the frequency domain features are compensated according to the actual compression rate. The correction process uses a linear compensation model, and the compensation coefficient is determined by the least squares method. The corrected feature information must meet the accuracy requirements preset by the system, and the feature error is generally required to be less than 1%. The correction result is output as the channel feature information of the frame structure layer for subsequent signal processing.

[0154] This embodiment processes the physical layer signal through the frame structure layer of the OTFS system model, realizing efficient time-frequency domain mapping and channel feature extraction. The Zadoff-Chu sequence is used as the basic pilot sequence, combined with the compression strategy of the adaptive quantization algorithm, and the compression rate can be flexibly adjusted according to the signal characteristics of different frequency bands, effectively improving the data transmission efficiency. Through the channel estimation and dynamic update mechanism of the minimum mean square error criterion, the system's adaptability to rapidly changing channels is guaranteed, and the accuracy of channel state estimation is improved. The time-frequency domain mapping optimization process based on the iterative optimization algorithm achieves the dual goals of maximizing channel capacity and minimizing bit error rate. The principal component analysis method is used to extract key features, and the compression rate mapping correction is performed in combination with the linear compensation model to ensure the accuracy of the feature information, so that the system can maintain high performance while greatly reducing the data transmission overhead and improving the overall work efficiency of the autonomous driving FPC sensor module.

[0155] Reference Figure 2 As shown, the present invention also provides a signal transmission optimization device for a high-order computing power autonomous driving FPC sensor module, which is applied to any of the above-mentioned signal transmission optimization methods for a high-order computing power autonomous driving FPC sensor module, comprising:

[0156] The acquisition module is used to obtain the original system signal data of the FPC sensor module, and perform adaptive orthogonal filtering on the original system signal data to obtain the corresponding optimized signal;

[0157] An analysis module is used to obtain multi-sensor data of the FPC sensor module, perform mode transformation analysis on the multi-sensor data and the optimization signal, and obtain a corresponding multi-dimensional coordinated working mode;

[0158] The association module is used to obtain the long-distance transmission data of the FPC sensor module and perform adaptive Doppler enhancement on the optimized signal to obtain the corresponding enhanced transmission signal;

[0159] A processing module, the processing module is used to perform error detection and channel coding correction analysis on the enhanced transmission signal based on a multi-dimensional coordinated working mode to obtain a corresponding preliminary optimization solution;

[0160] A control module, which is used to perform power consumption analysis on long-distance transmission data to obtain corresponding power consumption management information, and perform compressed transmission analysis to obtain corresponding compressed data transmission strategies;

[0161] The execution module is used to optimize the performance of the preliminary optimization scheme and the compressed data transmission strategy according to the preset OTFS system model to obtain the corresponding real-time performance optimization strategy.

[0162] The signal transmission optimization device of a high-order computing power autonomous driving FPC sensor module provided by the present invention significantly improves the accuracy and real-time performance of signal processing by performing adaptive orthogonal filtering on the original system signal data of the FPC sensor module, and effectively solves the technical difficulties of signal processing in complex motion scenes. Through the mode transformation analysis of multi-sensor data, the comprehensive and accurate capture of dynamic information of complex environments is achieved, and the perception ability of the sensor module in different working environments is enhanced. By adopting adaptive Doppler enhancement technology, the signal quality in the long-distance transmission process is effectively improved, and the transmission reliability and stability of the signal are significantly improved. By performing error detection and channel coding correction analysis on the multi-dimensional coordinated working mode and enhanced transmission signal, the signal transmission performance of the sensor module is further optimized. In terms of power consumption management, a more reasonable power consumption management strategy is formulated through the refined power consumption analysis of long-distance transmission data, which significantly reduces the system energy consumption. At the same time, a compressed transmission analysis method is innovatively proposed, which effectively reduces the resource consumption of data transmission while ensuring the quality of data transmission. Based on the OTFS system model, the comprehensive processing of the preliminary optimization scheme and the compressed data transmission strategy is realized, forming a comprehensive and efficient real-time performance optimization strategy. This method not only improves the overall performance of the autonomous driving FPC sensor module, but also provides a new technical solution for intelligent perception systems in complex environments, and has significant technological innovation value.

[0163] It should be noted that technicians in the relevant technical field can clearly understand that for the convenience and conciseness of description, the specific working process of the system and each module described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0164] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A signal transmission optimization method for a high-order computing power autonomous driving FPC sensor module, characterized in that: include: Acquire original system signal data of the FPC sensor module, and perform adaptive orthogonal filtering on the original system signal data to obtain a corresponding optimized signal; Acquire multi-sensor data of the FPC sensor module, perform mode transformation analysis on the multi-sensor data and the optimization signal, and obtain a corresponding multi-dimensional coordinated working mode; Acquire the long-distance transmission data of the FPC sensor module, and perform adaptive Doppler enhancement on the optimized signal to obtain a corresponding enhanced transmission signal; Based on the multi-dimensional coordinated working mode, error detection and channel coding correction analysis are performed on the enhanced transmission signal to obtain a corresponding preliminary optimization solution; Performing power consumption analysis on the long-distance transmission data to obtain corresponding power consumption management information, and performing compression transmission analysis to obtain corresponding compression data transmission strategy; The preliminary optimization scheme and the compressed data transmission strategy are optimized according to the preset OTFS system model to obtain a corresponding real-time performance optimization strategy.

2. The signal transmission optimization method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The method of obtaining the original system signal data of the FPC sensor module and performing adaptive orthogonal filtering on the original system signal data to obtain a corresponding optimized signal includes: Performing preliminary time-frequency extraction on the original system signal data to obtain preliminary feature extraction results; Perform Doppler construction based on the preliminary feature extraction result to obtain a corresponding Doppler feature mapping matrix; Calculating the signal entropy value and complexity index of the Doppler feature mapping matrix, and performing information fusion and refinement processing to obtain a corresponding signal feature vector; Performing orthogonal basis function decomposition on the signal eigenvector to obtain a corresponding orthogonal projection coefficient matrix; Performing a multi-scale wavelet transform on the original system signal data based on the orthogonal projection coefficient matrix to obtain a multi-resolution signal subband; Nonlinear weight configuration is performed on the multi-resolution signal subbands to obtain the optimized signal.

3. The signal transmission optimization method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The acquiring of multi-sensor data of the FPC sensor module, performing mode transformation analysis on the multi-sensor data and the optimization signal to obtain a corresponding multi-dimensional coordinated working mode, includes: Performing multi-dimensional feature extraction and nonlinear mapping on the multi-sensor data to obtain a feature enhanced data set; Constructing a corresponding orthogonal time-frequency-space scheduling matrix based on the feature enhancement data set, and mapping and associating it with the optimized signal to obtain a corresponding mapping relationship; Based on the mapping relationship, the multi-sensor data and the optimized signal are iteratively sparsely coded and recombined to obtain a recombined cooperative signal; Performing complexity adaptive coding on the recombined cooperative signal, and performing multi-dimensional signal mapping and decoupling processing to obtain a decoupled cooperative signal; The decoupled cooperative signal is iterated in a scheduling framework, and signal space constraints and orthogonal projection transformations are performed to obtain the multi-dimensional coordinated working mode.

4. The signal transmission optimization method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The acquiring of the long-distance transmission data of the FPC sensor module and the adaptive Doppler enhancement of the optimized signal to obtain a corresponding enhanced transmission signal includes: Performing adaptive Doppler frequency shift estimation on the optimized signal according to the long-distance transmission data to obtain a corresponding Doppler frequency shift estimation value; Constructing an adaptive Doppler compensation filter based on the Doppler frequency shift estimation value, and performing Doppler compensation processing on the optimized signal to obtain a corresponding compensation signal; Performing multi-scale decomposition on the compensation signal to obtain corresponding multi-scale coefficients; Performing nonlinear enhancement processing according to the multi-scale coefficients to obtain corresponding enhancement coefficients; Performing wavelet reconstruction on the enhancement coefficient to obtain a corresponding reconstructed signal; Encoding the reconstructed signal according to a preset polar code encoding algorithm to obtain a corresponding encoded signal; Performing spatial mapping transformation on the coded signal to obtain a multi-dimensional channel adaptation vector; fusing the multidimensional channel adaptation vector with the optimized signal to obtain a corresponding enhanced transmission signal; The step of encoding the reconstructed signal comprises: Recursively combine the original channels to generate a series of virtual polarization sub-channels; The polar code matrix is ​​constructed based on the Kronecker product to obtain the corresponding generator matrix; Reliability screening is performed on the polarized sub-channel to obtain a sub-channel transmission information bit and a sub-channel transmission fixed bit; The sub-channel transmission information bits and the sub-channel transmission fixed bits are combined into an information vector of a target length, and multiplied with the generation matrix to obtain the coded signal.

5. The signal transmission optimization method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The performing error detection and channel coding correction analysis on the enhanced transmission signal based on the multi-dimensional coordinated working mode to obtain a corresponding preliminary optimization scheme includes: Reconstructing the orthogonal modulation features of the multi-dimensional coordinated working mode to obtain a corresponding high-dimensional feature mapping operator; Performing hyper-dimensional feature encoding on the multi-dimensional coordinated working mode according to the high-dimensional feature mapping operator to obtain a corresponding collaborative feature space; Performing multipath channel transmission parameter sampling and spectrum analysis on the enhanced transmission signal to obtain channel interference characteristic parameters; Constructing a multi-dimensional channel coding framework according to the collaborative feature space and the channel interference feature parameters, and optimizing the channel coding parameters to obtain a channel coding optimization matrix; Iteratively detecting and correcting the channel coding optimization matrix to obtain a noise reduction channel coding result; A joint scheme analysis is performed on the multi-dimensional coordinated working mode and the enhanced transmission signal according to the noise reduction channel coding result to obtain a corresponding preliminary optimization scheme.

6. The signal transmission optimization method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The performing power consumption analysis on the long-distance transmission data to obtain corresponding power consumption management information, and performing compression transmission analysis to obtain corresponding compression data transmission strategy, includes: Based on the long-distance transmission data, a power consumption correlation network map is constructed, and power consumption distribution characteristics are calculated to obtain a power consumption correlation matrix; Performing adversarial coding compression processing according to the power consumption correlation matrix to obtain compressed adversarial coding information; Performing power consumption spectrum clustering decomposition on the compressed adversarial coding information to obtain power consumption management information; Performing initial transmission analysis on the transmission data according to the power consumption management information to obtain an initial transmission strategy; Performing information theory limit analysis and channel entropy capacity calculation on the initial transmission strategy to obtain compression dynamic adjustment parameters; The initial transmission strategy is adjusted according to the compression dynamic adjustment parameter to obtain the compressed data transmission strategy.

7. The signal transmission optimization method of the high-order computing power autonomous driving FPC sensor module according to claim 1 is characterized in that: The preliminary optimization scheme and the compressed data transmission strategy are processed according to the preset OTFS system model to obtain a corresponding real-time performance optimization strategy, including: The preliminary optimization scheme and the compressed data transmission strategy are input into the OTFS system model together, the sensor signal modulation and coding analysis is performed on the preliminary optimization scheme through the physical layer of the OTFS system model, and the bandwidth and delay requirements are analyzed in combination with the compressed data transmission strategy to obtain physical layer signal processing information; Through the frame structure layer of the OTFS system model, the physical layer signal processing information is mapped in the time-frequency domain and the channel estimation analysis is performed, and data compression rate mapping optimization is performed to obtain the frame structure layer channel characteristic information; Inputting the frame structure layer channel characteristic information into the system layer of the OTFS system model, performing multi-antenna transmission and interference characteristic collaborative analysis through the system layer, and obtaining system layer transmission optimization information; Perform resource scheduling and link adaptive analysis on the system layer transmission optimization information through the network layer of the OTFS system model to obtain corresponding initial transmission information and read the real-time requirements of the compressed data transmission strategy; Dynamically allocating resources for the initial transmission information according to the real-time requirement to obtain network layer performance scheduling information; Performing Doppler frequency shift compensation analysis and channel adaptability analysis on the network layer performance scheduling information through the channel layer of the OTFS system model to obtain channel layer propagation characteristic information; The processing layer of the OTFS system model performs a comprehensive strategy analysis on the physical layer signal processing information, the frame structure layer channel characteristic information, the system layer transmission optimization information, the network layer performance scheduling information and the channel layer propagation characteristic information to obtain and output the real-time performance optimization strategy.

8. The signal transmission optimization method of the high-order computing power autonomous driving FPC sensor module according to claim 7 is characterized in that: The performing time-frequency domain mapping and channel estimation analysis on the physical layer signal processing information through the frame structure layer of the OTFS system model, and performing data compression rate mapping optimization to obtain frame structure layer channel characteristic information, includes: Performing orthogonal pilot sequence encoding processing on the physical layer signal processing information to obtain a pilot sequence mapping basis; Performing compression ratio analysis on the compressed data transmission strategy to obtain a corresponding data compression ratio; Performing time-frequency domain mapping processing on the data compression rate based on the pilot sequence mapping basis to obtain preliminary time-frequency domain mapping information; Performing channel estimation processing on the preliminary time-frequency domain mapping information to obtain channel state estimation information; Performing time-frequency domain mapping optimization processing on the compressed data transmission strategy according to the channel state estimation information to obtain optimized time-frequency domain mapping data; Performing feature extraction processing on the optimized time-frequency domain mapping data to obtain frequency domain mapping feature information; The frequency domain mapping feature information is subjected to compression rate mapping correction processing according to the data compression rate to obtain the frame structure layer channel feature information.

9. A signal transmission optimization device for a high-order computing power autonomous driving FPC sensor module, characterized in that: The signal transmission optimization method of the high-order computing power autonomous driving FPC sensor module applied to any one of claims 1 to 8 above comprises: An acquisition module, the acquisition module is used to obtain the original system signal data of the FPC sensor module, and perform adaptive orthogonal filtering on the original system signal data to obtain a corresponding optimized signal; An analysis module, the analysis module is used to obtain multi-sensor data of the FPC sensor module, perform mode transformation analysis on the multi-sensor data and the optimization signal, and obtain a corresponding multi-dimensional coordinated working mode; An association module, the association module is used to obtain the long-distance transmission data of the FPC sensor module, and perform adaptive Doppler enhancement on the optimized signal to obtain a corresponding enhanced transmission signal; A processing module, the processing module is used to perform error detection and channel coding correction analysis on the enhanced transmission signal based on a multi-dimensional coordinated working mode to obtain a corresponding preliminary optimization solution; A control module, the control module is used to perform power consumption analysis on the long-distance transmission data to obtain corresponding power consumption management information, and perform compression transmission analysis to obtain corresponding compression data transmission strategy; An execution module is used to optimize the performance of the preliminary optimization scheme and the compressed data transmission strategy according to a preset OTFS system model to obtain a corresponding real-time performance optimization strategy.

Images