Coding Theory for Reliable Signal Processing外文学习资料.pdf

2014 International Conference on Computing, Networking and Communications, Invited Position Papers Coding Theory for Reliable Signal Processing ∗ † ‡ ∗ Aditya Vempaty , Yunghsiang S. Han , Lav R. Varshney , Pramod K. Varshney ∗Department of EECS, Syracuse University, Syracuse, NY †Department of EE, National Taiwan University of Science and Technology, Taiwan, ROC ‡IBM Thomas J. Watson Research Center, Yorktown Heights, NY avempaty@syr.edu, yshan@mail.ntust.edu.tw, varshney@alum.mit.edu, varshney@syr.edu Abstract—With increased dependence on technology in daily proposed by Wang et. al in [3] which serves as the mathe- life, there is a need to ensure their reliable performance. There matical basis for the material presented in this paper. Sec. III are many applications where we carry out inference tasks assisted considers the problem of distributed classification in Wireless by signal processing systems. A typical system performing an inference task can fail due to multiple reasons: presence of a com- Sensor Networks (WSNs) in the presence of sensors which ponent with permanent failure, a malicious component providing may possibly have stuck-at faults and discuss how DCFECC corrupt information, or there might simply be an unreliable can be used to ensure reliable classification. We then focus on component which randomly provides faulty data. Therefore, it is the target localization problem in the presence of malicious important to design systems which perform reliably even in the sensors called Byzantines [4]