Optimal and suboptimal detection of Gaussian signals in noise: asymptotic relative efficiency

The performance of Bayesian detection of Gaussian signals using noisy observations is investigated via the error exponent for the average error probability. Under unknown signal correlation structure or limited processing capability it is reasonable to use the simple quadratic detector that is optimal in the case of an independent and identically distributed (i.i.d.) signal. Using the large deviations principle, the performance of this detector (which is suboptimal for non-i.i.d. signals) is compared with that of the optimal detector for correlated signals via the asymptotic relative efficiency defined as the ratio between sample sizes of two detectors required for the same performance in the large-sample-size regime. The effects of SNR on the ARE are investigated. It is shown that the asymptotic efficiency of the simple quadratic detector relative to the optimal detector converges to one as the SNR increases without bound for any bounded spectrum, and that the simple quadratic detector performs as well as the optimal detector for a wide range of the correlation values at high SNR.

[1]  H. Vincent Poor,et al.  Sensor configuration and activation for field detection in large sensor arrays , 2005, IPSN 2005. Fourth International Symposium on Information Processing in Sensor Networks, 2005..

[2]  B. Bercu,et al.  Large deviations for quadratic forms of stationary Gaussian processes , 1997 .

[3]  C. Baker.  Optimum Quadratic Detection of a Random Vector in Gaussian Noise , 1966 .

[4]  Fred C. Schweppe,et al.  Evaluation of likelihood functions for Gaussian signals , 1965, IEEE Trans. Inf. Theory.

[5]  H. Vincent Poor,et al.  A large deviations approach to sensor scheduling for detection of correlated random fields , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[6]  H. Vincent Poor,et al.  An Introduction to Signal Detection and Estimation , 1994, Springer Texts in Electrical Engineering.

[7]  H. Vincent Poor,et al.  An introduction to signal detection and estimation (2nd ed.) , 1994 .

[8]  U. Grenander,et al.  Toeplitz Forms And Their Applications , 1958 .

[9]  Venugopal V. Veeravalli,et al.  Design of sensor networks for detection applications via large-deviation theory , 2004, Information Theory Workshop.

[10]  A. Dembo,et al.  Large Deviations for Quadratic Functionals of Gaussian Processes , 1993 .

[11]  Gerald R. Benitz,et al.  Large deviation rate calculations for nonlinear detectors in Gaussian noise , 1990, IEEE Trans. Inf. Theory.

[12]  S. Varadhan,et al.  Large deviations for stationary Gaussian processes , 1985 .

[13]  H. Poor,et al.  A reduced‐complexity quadratic structure for the detection of stochastic signals , 1985 .

[14]  H. Chernoff.  A Measure of Asymptotic Efficiency for Tests of a Hypothesis Based on the sum of Observations , 1952 .

[15]  W. Bryc,et al.  On the large deviation principle for a quadratic functional of the autoregressive process , 1993 .

[16]  H. Vincent Poor,et al.  Neyman-Pearson Detection of Gauss-Markov Signals in Noise: Closed-Form Error Exponent and Properties , 2005, ISIT.