Project Title: Digital system identification
Project Leader and Supervisor: Dr. Mohammad Kamrul Hasan,
Other Member(s): Zahidur R. Chowdhury, Rubyet Adnan, M. Maksudur Rahman Bhuiyan and M. Rezwan Khan
My role: Propose new concepts, Problem solving, design of experiments, computer simulation through coding and report writing.
Tools used: MATLAB / LATEX / Microsoft Office / Origin
Summary
- A new method was proposed for autoregressive (AR) parameter estimation from colored noise-corrupted observations using a damped sinusoidal model for autocorrelation function of the noise-free signal. The damped sinusoidal model parameters were first estimated using a least-squares based method from the given noisy observations. The AR parameters were then directly obtained from the damped sinusoidal model parameters. The performance of the proposed scheme was evaluated using numerical examples.
- Figure: Estimated poles of (left) the AR(3) system and (right) the AR(4) system at different SNR’s (o: true, +:ILS-CN (Improved Least-Squared for Colored Noise, x: proposed).
Publication(s)
M. Kamrul Hasan, Zahiudr R. Chowdhury and M. Rizwan khan, “Identification of autoregressive signals in colored noise using damped sinusoidal odel", IEEE Transaction on Circuit and Systems I, vol. 50(7), pp. 966-969, 2003. view / link
M. Kamrul Hasan, Zahidur R. Chowdhury, Rubyet Adnan, M. M. Rahman Bhuiyan, and M. Rezwan Khan, “A new method for parameter estimation of autoregressive signals in colored noise”, presented in 11th European Signal Processing Conference, pp. 173-176, France, 2002. view / link