Spectral Estimation Using Constrained Autoregressive (CAR) Model

In this work, a spectral estimation technique using a novel autoregressive model, constrained autoregressive (CAR) model, is proposed. CAR model is based on constraining one of the model parameters of an autoregressive model. This helps obtain a modified or desired AR spectrum for the signal. Constraining different AR parameters or changing the values of a particular parameter results in dissimilar AR spectrum for the signal. The value of this constrained parameter can be used for externally controlling the gain or improving the spectral resolution between two peaks in the spectrum. By constraining the M^th parameter, a_M, in a M-order model the resolution between two closely spaced peaks present in the signal spectrum can be improved. Similarly, by constraining the a₀ parameter and assigning it different values the spectral gain can be controlled