Research And Development Of Vehicle Vibration Signal Processing System Based On FPGA

In automotive environment, the components of MEMS dual-axis inclinometer have its own characteristics and are vulnerable to the vehicle vibration, electromagnetic interference, so the true signal is subject to serious interference, and which made data analysis great difficulty and unable to get the expected results, thus a filter is needed. However, the existing DSP processor can't meet the real-time demand in speed and anti-interference characteristics. And the FPGA-based on DSP is widely used in the signal processing field, the adaptive filtering method is one of the most active subjects.In the paper, in order to achieve adaptive signal de-noising with high-speed, first of all, it presents a novel LMS adaptive filtering algorithm with variable step size which overcomes the key problem in the classical LMS algorithm. The new algorithm establishes the non-linear relationship between the step and the error. And a MATLAB simulation analysis is done to compare classic LMS algorithm and other improve algorithm. Secondly, it presents hardware implementation design with Verilog HDL language in the ISE11.1 development platform, and the design, logic synthesis and simulation on ModelSim is done. Finally, the vehicle acceleration signal filter is finished. After the vehicle acceleration switches from the interface RS232 to 422 for conversion and transmits data to the Virtex-II Pro XC2VP30 chip for digital filtering though the UART linking to the computer for checking filter effect. Which, in order to assist testing programs, the paper also implements the UART baud rate generator, UART transmitter and receiver based XC2VP30 device as well as the simulation host computer display software.Experiment results show that compared with traditional algorithms, this algorithm has the advantages of fast convergence and high computing speed. The algorithm and the filter implements on FPGA are good in real-time use, can meet the real time processing requirements for Car-axis inclinometer's data.