Poisson Spectral Estimation using Hardlimited Samples.
Research rept. Oct 77-Sep 78,
NAVAL OCEAN SYSTEMS CENTER SAN DIEGO CA
Pagination or Media Count:
In this report, the problem of reconstructing the spectral density of a Gaussian signal process from hardlimited observations taken at Poisson sampling instants was considered. The estimate was shown to be asymptotically unbiased as the number of observation points approached infinity. In addition, the covariance of the estimate was also asymptotically bounded as the number of observations tended to infinity. The focus of future work will be to obtain tighter bounds on the covariance of the estimate and to show that these bounds tend to zero as the number of observations tends to infinity. Author
- Statistics and Probability