Abstract | ||
---|---|---|
Generalized Fourier correlators imposing finite system space-bandwidth products are described, and a class of binary filters is proposed. In pattern classification and signal registration applications, it is shown that for a particular class of signals, the binary filters yield the same asymptotic performance as the matched filter. It is hence adduced that a dynamic range of a single bit in the filter suffices for classification purposes. The effects of statistical sidelobe fluctuations and a finite system space-bandwidth product are included in the analysis. It is demonstrated that performance improves in a natural fashion with increase in the space-bandwidth product for both the binary filter and the matched filter.<> |
Year | DOI | Venue |
---|---|---|
1989 | 10.1109/29.17550 | Acoustics, Speech and Signal Processing, IEEE Transactions |
Keywords | Field | DocType |
filtering and prediction theory,matched filters,pattern recognition,binary filters,finite system space-bandwidth products,matched filter,pattern classification,signal registration,statistical sidelobe fluctuations | Root-raised-cosine filter,Pattern recognition,Prototype filter,Filter bank,Adaptive filter,Artificial intelligence,Kernel adaptive filter,Matched filter,Nonlinear filter,Mathematics,Filter design | Journal |
Volume | Issue | ISSN |
37 | 4 | 0096-3518 |
Citations | PageRank | References |
3 | 0.82 | 3 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Santosh S. Venkatesh | 1 | 381 | 71.80 |
Demetri Psaltis | 2 | 431 | 209.24 |