Application of adaptive filters in desired signal extraction

An application of Adaptive Filters (AF) in Desired Signal Extraction (DSE) embedded in Unknown Dynamic System (UDS) is explored in the current research. Least Mean Square Algorithm (LmSA) is utilized in proposed DSE. The UDS is synthesized by adopting Pseudo Random Binary Sequence (PrBS) excitation. To start with the LMSA is educated to converge and adapt its coefficients to the prevailing PRBS signal. These coefficients are subsequently utilized in DSE already embedded in the random noise. Carried out simulation studies by software implementation using `C' language with different filter orders, step sizes and arrived at an optimum solution in respect of different synthesized signals. Computational complexities associated with Weiner filter for simulation applications are also included in this paper. The results obtained during simulation for different filter orders are tabulated and also depicted pictorially. The LMSA implementation methodology is shown in a block diagram as well explained in terms of a flow chart. Finally the LMSA is implemented in realistic environment and the desired signal extracted is also depicted pictorially.