Abstract
In this paper, a quantitative analysis of signal-to-noise ratio (SNR) in the short-time Fourier transform domain for multicomponent signals in additive white noise is given. It is shown that the SNR is increased by the order of O(N/K), where K is the number of multicomponents of a signal, N/T is the sampling rate, and T is the window size. The SNR increase rate is optimal given K. For this result, the SNR definition is generalized, which is suitable for signals in the time domain as well as other domains. Numerical examples are presented to illustrate this theory.
| Original language | English |
|---|---|
| Pages (from-to) | 1101-1102 |
| Number of pages | 2 |
| Journal | IEEE Transactions on Signal Processing |
| Volume | 45 |
| Issue number | 4 |
| Publication status | Published - 1997 |
| Externally published | Yes |