Time Varying Autoregressive Model Parameters Estimation using Discrete Energy Separation Algorithm

Abstract

Time Varying Autoregressive (TVAR) model for the Amplitude and Frequency modulated (AM-FM) signal is presented   In this paper. TVAR parameters of AM-FM signal are estimated using Discrete Energy Separation (DESA) Algorithm. The performance of DESA method is shown to be comparable to the existing basis function method for AM, FM, AM-FM signal models. The proposed method is simpler to execute in hardware and consumes considerably less computational resources compared to the method using Adaptive and the Basis function methods. .It is demonstrated that the proposed technique based on DESA has certain distinct advantages over the conventional method employing basis functions. Another advantage is that the present method works well with quickly varying signals