Abstract
This letter deals with the use of Importance Sampling (IS) techniques and the Mean-Square (MS) error in neural network training, for applications to detection in communication systems. Topics such as modifications of the MS objective function, optimal and suboptimal IS probability density functions, and adaptive importance sampling are presented. A genetic algorithm was used for the neural network training, having considered adaptive IS techniques for improving MS error estimations in each iteration of the training. Also, some experimental results of the training process are shown in this letter. Finally, we point out that the mean-square error (estimated by importance sampling) attains quasi-optimum training in the sense of minimum error probability (or minimum misclassification error).
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Smith, P. J., Shafi, M. and Gao, H.: Quick simulation: A review of importance sampling techniques in communications systems, IEEE J. Select. Areas Commun. 15 (1997), 597-613.
Chen, J.-C., Lu, D., Sadowsky, J. S. and Yao, K.: On importance sampling in digital communications-Part I: Fundamentals, IEEE J. Select. Areas Commun. 11 (1993), 289-299.
Wolfe, R. J., Jeruchim, M. C. and Hahn, P. M.: On optimum and suboptimum biasing procedures for importance sampling in communication simulation, IEEE Trans. Commun. 38 (1990), 639-647.
Al-Qaq, W. A., Devetsikiotis, M. and Townsend, J. K.: Stochastic gradient optimization of importance sampling for the efficient simulation of digital communication systems, IEEE Trans. Commun. 43 (1995), 2975-2985.
Al-Qaq, W. A. and Townsend, J. K.: A stochastic importance sampling methodology for the efficient simulation of adaptive systems in frequency nonselective Rayleigh fading channels, IEEE J. Select. Areas Commun. 15 (1997), 614-625.
Gerlach, K.: New results in importance sampling, IEEE Trans. Aerosp. Electr. Systems 35 (1999), 917-925.
Orsak, G. C.: A note on estimating false alarm rates via importance sampling, IEEE Trans. Commun. 41 (1993), 1275-1277.
Orsak, G. C. and Aazhang, B.: Constrained solutions in importance sampling via robust statistics, IEEE Trans. Inform. Theory 37 (1991), 307-316.
Sadowsky, J. S. and Bucklew, J. A.: On large deviation theory and asymptotically efficient Monte Carlo estimation, IEEE Trans. Inform. Theory 36 (1990), 579-588.
Sanz-González, J. L. and Andina, D.: Performance analysis of neural network detectors by importance sampling techniques, Neural Processing Letters 9 (1999), 257-269.
Sanz-González, J. L. and Andina, D.: Importance sampling techniques in neural detector training, Lectures Notes in Artificial Intelligence, LNAI 2167 (Machine Learning: ECML'01), Springer-Verlag (2001), 431-441.
Andina, D., Sanz-González, J. L. and Jiménez-Pajares, A.: A comparison of criterion functions for a neural network applied to binary detection, Proc. of IEEE Int. Conf. on Neural Networks (ICNN'95) 1 (1995), 329-333.
Poor, H. V.: An Introduction to Signal Detection and Estimation, Springer-Verlag, Berlin, (Second Edition), 1994.
Van Trees, H. L.: Detection, Estimation, and Modulation Theory, Part. I, John Wiley & Sons, New York, 1968.
Seijas, J. and Sanz-González, J. L.: Basic-evolutive algorithms for neural networks architecture configuration and training. Proc. IEEE Int. Symp. Circuits Syst. (ISCAS'95), 1 (1995), 125-130.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Sanz-González, J.L., Andina, D. & Seijas, J. Importance Sampling and Mean-Square Error in Neural Detector Training. Neural Processing Letters 16, 259–276 (2002). https://doi.org/10.1023/A:1021766820005
Issue Date:
DOI: https://doi.org/10.1023/A:1021766820005