STUDY OF LINEARLY CONSTRAINED LMS FILTER FOR CANCELLING NOISE IN RADAR SIGNAL PROCESSING

Abstract

In electronic warfare, accurate estimation of signal parameters plays an important role in helping to correctly identify radiation sources and determine their operating modes. On the other hand, the signal at the receiver consists of echo and noisy signals. Therefore, to improve accuracy when estimating signal parameters, this article studies the linearly constrained last mean square algorithm (LC-LMS) for reducing noise in signal processing and verifies with a generated radar signal by generator E8267D. The effectiveness of the LC-LMS filter is evaluated through two main steps: the first step provides a simulation of the LC-LMS and compares it with LMS and its variants, such as normalized LMS (NLMS) and leaky LMS (LLMS), with simulated radar signals such as continuous waveforms and radar pulses in a MATLAB environment. The second step provides verification of the LC-LMS filter by generating radar signals from vector signal generator E8267D. Simulation results show that the LC-LMS filter has better noise filtering than the LMS and its variants, such as the NLMS and LLMS. Additionally, the validation results confirm that the LC-LMS filter can be applied to real-time radar signals with a signal-to-noise ratio (SNR) greater than 0 dB.