Basic concepts in information theory and coding: the adventures of secret agent 00111
Basic concepts in information theory and coding: the adventures of secret agent 00111
Population-Based Incremental Learning: A Method for Integrating Genetic Search Based Function Optimization and Competitive Learning
Ant-based topology convergence algorithms for resource management in VANETs
EUROCAST'07 Proceedings of the 11th international conference on Computer aided systems theory
Hi-index | 0.00 |
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.