INITIAL TASK PLACEMENT FOR MACRO-PROGRAMMED WIRELESS SENSOR NETWORK

Vitalijus Martusevičius, Asta Martusevičienė, Egidijus Kazanavičius, Vaidas Jukavičius

Abstract


One of the most important issues in wireless sensor network programming is to assign a set of tasks to a set of nodes with limited energy resources in order to minimize energy consumption. In this paper, we develop a task assignment model where cost function is formulated as a sum of computation and communication energy spent in the network. The model employs task placement constraints that ensure expected lifetime of individual node. We propose an efficient genetic algorithm with repair operator to obtain a minimal cost solution. The repair operator guarantees valid task assignments that meet model constraints as candidate solutions are generated during the process of evolution. Experiments reveal that provided heuristic takes a reasonable amount of time to produce near optimal results.

http://dx.doi.org/10.5755/j01.itc.40.4.984


Keywords


wireless sensor network; task mapping; genetic algorithm; macro-programming

Full Text: PDF

Print ISSN: 1392-124X 
Online ISSN: 2335-884X