Modulation Recognition Algorithms for Intentional Modulation on Pulse (IMOP) Applications
Abstract
In this report, the problem of signal modulation classification is investigated. A modulation recognition algorithm for classifying different signal modulation types and noise is described. The modulation type includes unmodulated CW, narrow-band FM, wide-band FM, triangular FM, BPSK, DSB-SC and AM The algorithm involves a combination of decision theoretic and pattern recognition techniques. The decision theoretic technique is based on the calculation of a number of statistics of the input sequence to be classified. It is used to separate noise from the signals, the constant envelope waveforms from the varying-envelope waveforms, the unmodulated CW from the waveforms with phase information, and the two varying-envelope waveforms from one another. The pattern recognition technique is used to distinguish the three FM and BPSK waveforms. The technique is based on the use of a linear discriminant. The discriminant is trained by feature vectors generated from the residual-phase histogram Finally, computer simulations are used to demonstrate the performance of the proposed modulation recognition algorithm, and an extensive analysts is also included.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 2001
- Accession Number
- ADA640156
Entities
People
- Stephen Sung
- Yifeng Zhou
Organizations
- Defence Research and Development Canada