Abstract
This paper proposes a robust signal detection procedure based on generalised patterns in time-frequency space. The paper derives a non-linear time domain correlation process which is equivalent to a time-frequency based detection scheme. The performance of this robust detection scheme in the additive white Gaussian noise case is evaluated with the aid of receiver operating curves. Some examples are given which show that the proposed procedure under nominal conditions does not perform as well as standard detection scheme (ie matched filter) but gives better performance when the measured signal deviates from its nominal model.
Similar content being viewed by others
References
S. A. Kassam, and H. V. Poor, “Robust Techniques for Signal Processing,” Proceeding of the IEEE, vol. 73, no. 3, 1985.
L. Cohen, “Time–Frequency Distributions—A Review,” Proc. IEEE, vol. 77, no. 7, 1989, pp. 941–981.
O. P. Kenny, and L. B. White, “Robust Detection in Time–Frequency Space,” Signals Analysis Discipline Report, SA/95/06U, 1995.
S. Kotz, N. L. Johnson, and D. W. Boyd, “Series Representation of Distributions of Quadratic Forms in Normal Variables II. Non–Central Case,” Ann. of Statistical Math., 1966.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Kenny, O.P., White, L.B. Statistical Performance of a TFD based Detection Scheme. Multidimensional Systems and Signal Processing 9, 403–409 (1998). https://doi.org/10.1023/A:1008466625808
Issue Date:
DOI: https://doi.org/10.1023/A:1008466625808