A novel approach for joint estimation of time delay and scale factor with applications to the M-wave analysis

In this paper we propose a method for joint estimation of delay and scale factor between two deterministic and unknown signals. The method is based on the separation of the two parameters. The scale factor is estimated from the autocorrelation functions (ACFs) of the two signals. The ACFs are independent on the delay while they maintain the scale factor. The delay is estimated by scaling one of the two signals by the estimated scale factor, so that the delay estimation is not biased by the scale factor. The method proposed is compared with the maximum likelihood joint estimation and provides better performance for a rather large range of signal to noise ratios. Potential applications of the technique In surface EMG signal analysis are discussed and results related to the M-wave processing are reported.