Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Proceedings of the 10th international conference on Architectural support for programming languages and operating systems
Proceedings of the 2003 ACM symposium on Applied computing
Proceedings of the 3rd international conference on Embedded networked sensor systems
| Hi-index | 0.00 |
The current trend for monitoring different parameters in any facility is towards fully integrated surveillance systems using sensor network. For example, sensor network may be deployed over a military installation to monitor temperature, humidity, pressure and surveillance for various equipment or critical regions. Deploying individual sensors node for each parameter would be costly, especially if multiple parameters are to be measured at a single location. This calls for intelligent deployment of sensor nodes with multiple sensing interfaces. In this paper, we propose placement and interface selection algorithm (PISA) for deploying such sensor nodes and selecting their respective sensing interfaces. The inputs to PISA are the various individual regions where different types of parameters are to be measured. These regions are characterized with imaginary nodes (Inodes). PISA works in two phases; firstly, it identifies various clusters of Inodes that can be formed using heterogeneous clustering algorithm and then in the second phase it uses the cluster centers as candidate locations to identify the appropriate sensor node locations and their respective sensing interfaces with the help of genetic algorithm. We discuss the features of PISA and test various example cases via MATLAB simulations. The solutions obtained using PISA enable efficient parameter monitoring over the regions of interest with minimum number of sensor nodes and sensing interfaces.