Answers are grounded in this paper and cite their sources.
Time delay estimation (TDE) is applied in many areas. Its estimation performance plays an important role in many actual systems, such as malfunction sound location. In this paper, estimation performances of three TDE algorithms, correlation, covariance, and fractional lower order covariance, are compared. Traditional, additive noises in the actual collected signals are described by Gaussian distribution. However, they have often impulsiveness in practice, and are modeled as α-stable distribution. First, correlation, covariance, and fractional lower order covariance method are analyzed in theory. Then, computer simulation experiments are carried out. Computer sound card records pure audio signals, different pulse intensity noises added to simulate actual environments. Next, results of three algorithms for time delay estimation were obtained in different signal to noise ratio (SNR) conditions. Under the same conditions, estimated RMS (root-mean-square) errors of three algorithms are analyzed and compared. Experimental results show that under low SNR and strong impulsive noise environments, fractional lower order covariance method indicates best estimation performance.
No takes yet. Share an insight, caveat, or question.
Junhao Li (2017) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: