Design and Implementation of Least Mean Square Adaptive Filter Using Verilog
Sandu Ajay Kumar1, T. Satya Savithri2
1Sandu Ajay Kumar, Department of Electronics & Communication Engineering, Jawaharlal Nehru Technological University, Hyderabad (Telangana), India.
2Dr. T. Satya Savithri, Department of Electronics & Communication Engineering, Jawaharlal Nehru Technological University, Hyderabad (Telangana), India.
Manuscript received on 21 October 2024 | Revised Manuscript received on 10 December 2024 | Manuscript Accepted on 15 December 2024 | Manuscript published on 30 December 2024 | PP: 1-7 | Volume-12 Issue-12, December 2024 | Retrieval Number: 100.1/ijisme.C456614030225 | DOI: 10.35940/ijisme.C4566.12121224
Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Adaptive filters are ability of adaptation to an unknown environment. These filters have been used widely because of its capable of operating in an unknown system and low power implementation of hardware. Adaptive filters have great range of signal processing and control operations for the tracking time variations of input statistics and Robust to the noise immunity. These filters used various Areas like Noise cancelling (interface cancelling), system identification, inverse modelling and echo predication. Adaptive filters structures have the adaptive algorithms to perform the time variations of the input statics and Robust to the noise cancelling. The most popular algorithm is LMS (Least Mean Square) it produces the least mean square of error signal in the adaptive filter to minimize noise power. Adaptive filter structures follow the two algorithms RLS and LMS. RLS algorithms excellent performance with increased complexity and the filter coefficients that minimize waited linear least squares cost function relating to the input signals. It requires infinite memory for error signal. LMS algorithms are simplest to understand and describe the hardware of the system compare to the RLS (Recursive Least Square). LMS algorithms are follows the stochastic gradient descent method to minimize the error signal and de-nosing task. It estimates the gradient vector from the input data and LMS incorporates an iterative procedure that makes successive corrections to the weight vector in the direction of the negative of the gradient vector which leads to minimum mean square error. It doesn’t require correlation functions for calculations. The main aim of the project is to design the LMS algorithm based adaptive filter using Verilog HDL to reduce the power consumption, hardware complexity and improving the noise cancelling for the adaptive filter on the FGPA boards. An important challenge in the LMS adaptive filters design implementation of structural model in the Verilog HDL for image processing to target the noise cancelling, power and hardware complexity. Tool use for implementation the structural model of LMS filter is vivado tool and Xilinx software for FPGA board implementation.
Keywords: Adaptive Filters, FPGA, LMS, RLS, Increased Complexity.
Scope of the Article: VLSI Algorithms