Multi-dimensional information space view of wireless sensor networks with optimization applications

Publication Type:
Conference Proceeding
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2012, 6928 LNCS (PART 2), pp. 146 - 152
Issue Date:
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2011001679OK.pdf1 MB
Adobe PDF
This paper presents an optimization example using a new paradigm for viewing the work of Wireless Sensor Networks. In our earlier paper [1] the Observed Field (OF) is described as a multi-dimensional "Information Space" (ISp). The Wireless Sensor Network is described as a "Transformation Space" (TS), while the information collector is a single point consumer of information, described as an "Information Sink" (ISi). Formal mathematical descriptions were suggested for the OF and the ISp. We showed how the TS can be formally thought of as a multi-dimensional transform function between ISp and ISi. It can be aggregated into a notional multi-dimensional value between {0,1}. In this paper, this formal mathematical description is used to create a genetic algorithm based optimization strategy for creating routes through the TS, using a cost function based on mutual information. The example uses a connectivity array, a mutual information array and the PBIL algorithm. © 2012 Springer-Verlag.
Please use this identifier to cite or link to this item: