Octavia A. Dobre
Department of Electrical Engineering, NJIT
Time: 1:00 – 2:00 PM, Thursday, April 15, 2004
Place: New Jersey Institute of Technology (NJIT), Room 202, ECE Center, Newark, NJ

Signal intelligence systems are required in several military applications, such as electronic surveillance and threat analysis. An important component of such a system is a modulation classifier, which identifies the modulation of the incoming signal. Once the correct modulation format is known, other tasks such as signal demodulation and information extraction can be performed. In a non-cooperative environment, with no knowledge of the transmitted data and many unknown parameters at the receiver, such as the signal power, carrier frequency and phase offsets, etc., blind identification of the modulation format is a difficult task.

In this talk, blind pattern recognition algorithms that employ cyclic cumulant (CC)-based features are presented to identify linear modulations. First, the modulation recognition problem is briefly formulated and the concept of cyclostationarity introduced. Then, the selection of CC-based features is discussed, emphasizing the tradeoff between their discriminating capability and robustness to modeling errors. In general, CC-based features are robust to carrier phase and synchronization errors. Furthermore, by properly choosing fourth-, sixth- and eighth-order CC-based features, a classifier robust to carrier frequency offset and phase noise is obtained for the recognition of quadrature amplitude modulation (QAM) signals.

Octavia A. Dobre received the M. Sc. and Ph. D. degrees in Electrical Engineering from “Politehnica” University of Bucharest (PUB), Romania, in 1991 and 1998, respectively. Between 1998 and 2001 she has been an Assistant Professor at Department of Remote Control and Electronics in Transports, PUB. In 2000 she was the recipient of a British Royal Society fellowship. She then joined Wireless Information Systems Engineering Laboratory at Stevens Institute of Technology in Hoboken, as a Fulbright fellow until 2002. Currently she is a Research Associate at Department of Electrical and Computer Engineering, New Jersey Institute of Technology. Her current research interests include automatic modulation classification algorithms, ARQ for wireless communications, simulation of communication systems, statistical signal processing for communications and multicarrier modulation techniques.