Laboratory Experiments to Measure the Effects of Signal Bandwidth and System Sample Rate on Processing Gain.
Abstract
Most of the conclusions of this memorandum have been stated in previous sections, but a brief summary will be inserted here; (1) The combination of a 100 Hz difference frequency and a 200 Hz sample rate was an unfortunate choice for the original DELTIC correlator. In general, a condition to be avoided is fs = 2 fH where fs is the system sample frequency and fH is the heterodyne difference frequency, (2) The convention of measuring filter bandwidths at the -3 dB points is misleading in a nonlinear system. Selecting the -20 dB points seems to be a better measure of filter bandwidth, (3) Care should be used when assessing statements concerning processing gain because the definition (or at least the measurement) of processing gain is ambiguous. The two definitions which are used in this memorandum result in a measured difference of 1 dB in favor of Processing Gain II, and (4) The processor's envelope detected output noise does not follow a Rayleigh distribution. The actual distribution is biased toward the lower noise values. Several interesting problems remain to be investigated when time and equipment permit. Of primary interest is the comparison of PN and FM signal types. This comparison should permit verification (or rejection) of the theory concerning the definition of filter bandwidth because an FM signal has a near-ideal wideband frequency spectrum.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 21, 1967
- Accession Number
- ADA054050
Entities
People
- P. H. Hawkes
Organizations
- Naval Information Warfare Systems Command