Browsing by Author "Varshney, PK"

Now showing 1 - 2 of 2

Asymptotic performance analysis for minimum-hamming-distance fusion
(2005)
Chen, PN
;
Wang, TY
;
Han, YSS
;
Varshney, PK
;
Yao, C
Distributed (M-ary) detection and fault-tolerance have been considered as two fundamental functions in the context of large-scale sensor networks. Distributed multiclass classification fusion using error correcting codes (DCFECC) has been proposed to provide good fault-tolerance capability in wireless sensor networks. Minimum Hamming distance fusion is an essential part of the DCFECC approach. In this paper, we study the asymptotic performance of minimum Hamming distance fusion for both fault-free and faulty situations when the number of sensors tends to infinity. We conclude that the error probability vanishes asymptotically as long as the minimum Hamming distance d(min) of the DCFECC code approaches infinity, and the probabilities of correct local classification for all hypotheses are greater than one half. In case d(min)/2, normalized by the number of sensors, can be made larger than the largest local classification error, an explicit expression for the error exponent of the DCFECC system in terms of the Kullback-Leibler divergence can be established. A converse where the DCFECC decoding error is bounded away from zero is also addressed.
Distributed fault-tolerant classification in wireless sensor networks
(2005-04-01)
Wang, TY
;
Han, YSS
;
Varshney, PK
;
Chen, PN
;
電信工程研究所
Fault-tolerance and data fusion have been considered as two fundamental functions in wireless sensor networks. In this paper, we propose a novel approach for distributed multiclass classification using a fault-tolerant fusion rule for wireless sensor networks. Binary decisions from local sensors, possibly in the presence of faults, are forwarded to the fusion center that determines the final classification result. Classification fusion in our approach is implemented via error correcting codes to incorporate fault-tolerance capability. This new approach not only provides an improved fault-tolerance capability but also reduces computation time and memory requirements at the fusion center. Code matrix design is essential for the design of such systems. Two efficient code matrix design algorithms are proposed in this paper. The relative merits of both algorithms are also studied. We also develop sufficient conditions for asymptotic detection of the correct hypothesis by the proposed approach. Performance evaluation of the proposed approach in the presence of faults is provided. These results show significant improvement in fault-tolerance capability as compared with conventional parallel fusion networks.