Distributed detection in wireless sensor networks using complex field network coding

Eritmen, Kayhan and Keskinöz, Mehmet (2013) Distributed detection in wireless sensor networks using complex field network coding. In: 16th International Conference on Information Fusion (FUSION 2013), Istanbul, Turkey

Full text not available from this repository. (Request a copy)

Abstract

The signal transmission and the information fusion in the wireless sensor networks (WSN) are conventionally assumed to operate over orthogonal channels, which becomes bandwidth and throughput inefficient. To remedy this inefficiency and to improve the detection performance simultaneously, in this work, we first propose to use relaying with complex field network coding (CFNC), which operates over non-orthogonal channels. This provides not only the diversity in space but also diversity in time as well. In this method, each sensor is assigned a unique pre-determined signature and this provides robustness against the multi-access interference. Secondly, we derive the optimal likelihood ratio test (LRT) based fusion rule for the proposed system. Thirdly, we devise a way to choose the sensor signatures. Finally, we evaluate the detection performance of the proposed scheme and compare it with the performance of the conventional method. The simulation results show that we can obtain up to a 105% improvement using distributed detection with CFNC rather than employing the conventional method in literature. This suggests that the distributed detection with CFNC is a promising technique that can be used in next generation WSNs.
Item Type: Papers in Conference Proceedings
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101-6720 Telecommunication
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Electronics
Faculty of Engineering and Natural Sciences
Depositing User: Mehmet Keskinöz
Date Deposited: 22 Jan 2014 11:22
Last Modified: 26 Apr 2022 09:12
URI: https://research.sabanciuniv.edu/id/eprint/22724

Actions (login required)

View Item
View Item