Efficient Mobile Agent Path-search Techniques using Genetic Algorithm Processing

Background/Objectives: Although the efficiency of genetic algorithms improves with the creation of each generation, a number of generations are needed to obtain the desired results. In addition, when ad hoc unit increases are linked to a network, it may be necessary to compare all cases. Methods/Statistical Analysis: This requires the simultaneous generation of multiple algorithms at one time. Where a single process is used to manage all such algorithms, the overall efficiency of the network will decrease. Findings: The algorithm proposed in this thesis introduces router group cell units for use in the distributed processing of previous genetic algorithms. The experimental results showed that the proposed algorithm reduced path processing costs caused by the alternative path setup by approximately 27% when compared with Dijkstra's and the Munetomo algorithm. Operation time for the alternative path setup was approximately twice as fast as that of Dijkstra's algorithm. These results suggest that the algorithm proposed in this paper is more efficient than either Dijkstra's or the Munetomo algorithm in terms of alternative path setup during router failure. Application/Improvements: The study presents ways to reduce overall search delays across a network through the use of a cell-based genetic algorithm.