Efficient implementation of adaptive filters using TMS320C6713 DSP platform

This paper presents a methodology for accelerated development of solution associated to adaptive filtering using Matlab/Simulink, Code Composer Studio and DSK 6713 digital signal processing (DSP) technologies. The purpose of this methodology is to provide an efficient and rapid method to develop and test adaptive filters in DSPs. The methodology development represents a very important tool for the engineer in charge of design-simulation-implementation of adaptive filters. Another important benefit is that it avoids low level hardware work that can be tedious and time consuming. An example of application is presented in this paper in order to illustrate the feasibility of this methodology. The methodology is applied in order to implement an adaptive filter in the DSP platform. The project steps are discussed in details, including the methods of transforming the Matlab code into DSP code. The results are analyzed in terms of accuracy and convergence speed. The TMS320C6713 development kit supplied by Spectrum Digital Inc., that includes a float point DSP, was used to implement the adaptive filter.