Research On DSP Test System With Error Correction And Intelligent Interface

DSP is in a digital signal to handle large amounts of information in the unique micro processor device. The working principle is turning the received analog signal to digital signal of 0 or 1, and then to adjust,weaken, strengthening of digital signal, and the digital data interpretation in the other system chip back to analog data format or actual environment.After decades of rapid development, the application of DSP products has expanded to people’s study, work and all sides of life. The resulting DSP test system design has become the mainstream in the current system design. The high integration of the system is composed of DSP chip not only play a prominent role in the field of communications, computer, and even in many product industry of people’s daily life in common use.In the practical application of test system, often appear as a result of the sensors work too narrow frequency band, lead to some useful signal frequency component deformity, finally appears the problem of distortion test data. Traditional SCM or integrated circuit or even a combination of both, errors are not current time correction of the above reasons. Visible, the instantaneous and the true response time of the received signal and the real parameters etc, all is one of the important index to study the system performance.In order to headle with current-time revise error, you need to design a data processing chip can have error correction function. Based onTMS320F2812 DSP chip and self-adaption neural network dynamic compensation algorithm, performance and structure characteristics of TMS320F2812 DSP chip characteristic, through the reasonable design of the peripheral hardware structure and combined with external expanding memory, this design has the function of error correction and intelligent interface of DSP system. The dynamic input, output signal is corrected, Construct Mathematics model. Meanwhile, the inverse modeling operation, where the signal output as compensation system input, the input signal as the output compensation system, using neural network fitting method to get the corresponding linear relationship of input and output. Experiments show that the testing system can correctly to store data, and can effective correction caused by the dynamic error of sensor simulator.The innovation points of this paper are as follows: the core components of TMS320F2812 DSP as the main control unit, and its functional properties based on the configuration of the peripheral circuit with good performance. Data processing, using adaptive neural network algorithm, the inverse modeling, effectively reduce the input signal is the disturbanceofinformation. Design in this paper:mainly focus on Performance Analysis of TMS320F2812 DSP chip specific, arrangement by corresponding peripheral circuit design,namely the relevant study on the corresponding data algorithm.