TY - GEN
T1 - Importance Sampling Estimator of Outage Probability under Generalized Selection Combining Model
AU - Ben Rached, Nadhir
AU - Botev, Zdravko
AU - Kammoun, Abla
AU - Alouini, Mohamed-Slim
AU - Tempone, Raul
N1 - KAUST Repository Item: Exported on 2021-02-19
PY - 2018/9/21
Y1 - 2018/9/21
N2 - We consider the problem of evaluating outage probability (OP) values of generalized selection combining diversity receivers over fading channels. This is equivalent to computing the cumulative distribution function (CDF) of the sum of order statistics. Generally, closed-form expressions of the CDF of order statistics are unavailable for many practical distributions. Moreover, the naive Monte Carlo method requires a substantial computational effort when the probability of interest is sufficiently small. In the region of small OP values, we propose instead an efficient, yet universal, importance sampling (IS) estimator that yields a reliable estimate of the CDF with small computing cost. The main feature of the proposed IS estimator is that it has bounded relative error under a certain assumption that is shown to hold for most of the challenging distributions. Moreover, an improvement of this estimator is proposed for the Pareto and the Weibull cases. Finally, the efficiency of the proposed estimators are investigated through various numerical experiments.
AB - We consider the problem of evaluating outage probability (OP) values of generalized selection combining diversity receivers over fading channels. This is equivalent to computing the cumulative distribution function (CDF) of the sum of order statistics. Generally, closed-form expressions of the CDF of order statistics are unavailable for many practical distributions. Moreover, the naive Monte Carlo method requires a substantial computational effort when the probability of interest is sufficiently small. In the region of small OP values, we propose instead an efficient, yet universal, importance sampling (IS) estimator that yields a reliable estimate of the CDF with small computing cost. The main feature of the proposed IS estimator is that it has bounded relative error under a certain assumption that is shown to hold for most of the challenging distributions. Moreover, an improvement of this estimator is proposed for the Pareto and the Weibull cases. Finally, the efficiency of the proposed estimators are investigated through various numerical experiments.
UR - http://hdl.handle.net/10754/630800
UR - https://ieeexplore.ieee.org/abstract/document/8462177
UR - http://www.scopus.com/inward/record.url?scp=85054243700&partnerID=8YFLogxK
U2 - 10.1109/ICASSP.2018.8462177
DO - 10.1109/ICASSP.2018.8462177
M3 - Conference contribution
AN - SCOPUS:85054243700
SN - 9781538646588
SP - 3909
EP - 3913
BT - 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
PB - Institute of Electrical and Electronics Engineers (IEEE)
ER -