Wavelet domain communication system: bit error sensitivity characterization for geographically separated transceivers (U)

• Published in: MILCOM 2002. Proceedings Vol. 2; pp. 1378 - 1382 vol.2
• Main Authors: Lee, M.J., Temple, M.A., Claypoole, R.L., Raines, R.A., Stephens, J.P.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 2002
• Subjects: Bit error rate; Computer languages; Electromagnetic spectrum; Interference; Mathematical model; Performance analysis; Transceivers; Wavelet domain; Wavelet packets; Wavelet transforms
• ISBN: 0780376250; 9780780376250
• DOI: 10.1109/MILCOM.2002.1179682

(U) The wavelet domain communication system (WDCS) uses a packet-based transform to estimate the electromagnetic spectrum. Adaptive thresholding and 'notching' is applied to remove sub-bands containing interference, effectively excising the interference. A unique communication basis function is generated from the notched spectral estimate and ideally contains no information bearing energy at notched locations. The basis function is data modulated prior to transmission. For this work, WDCS performance is evaluated for various scenarios using a MATLAB/spl reg/ model and simulation results. Bit error rate (P/sub B/) provides the key metric for analysis and performance evaluation. Bit error "sensitivity" is examined for geographically separated transceivers "observing" dissimilar electromagnetic environments. Previous WDCS research constrained transceivers to a localized geographical region such that nearly identical basis functions were created for communicating. Given sufficient transceiver separation, dissimilar basis functions are more likely to occur and P/sub B/ performance degrades. WDCS P/sub B/ robustness is addressed for scenarios including effects of: (1) Doppler variation between transceivers; and (2) operation over a fading channel. For binary, 4-Ary, and 8-Ary CSK data modulations, P/sub B/ sensitivity analyses indicate the WDCS communicates effectively under nonideal 'real-world' conditions while exhibiting average P/sub b/ improvements of 5.4, 5.1, and 5.8 dB, respectively, relative to systems having no interference suppression.