Estimating the frequency of a noisy sinusoid by linear regression
IEEE Transactions on Information Theory
| Hi-index | 0.00 |
A novel algorithm for tracking the parameters of slowly time-varying multiple sine waves in noise is proposed. The estimates of the instant frequencies are updated directly using a simple property of the one-step increment of the phases. The algorithm is based on the recursive least squares method with exponential weighting factor. The rise-time and the noise rejection characteristics of the algorithm are studied theoretically in the single complex sine wave case, assuming that the SNR is sufficiently high. The proposed algorithm is easy to implement, and computer simulations show that it yields good results.