Markov chain models for genetic algorithm based topology control in MANETs
EvoCOMNET'10 Proceedings of the 2010 international conference on Applications of Evolutionary Computation - Volume Part II
| Hi-index | 0.00 |
This proposal will formulate a model representing key, multi-attribute Quality of Service (QoS) parameters for wireless Mobile Ad-hoc Networks (MANETs), consistent with Institute of Electrical and Electronics Engineers 802.11 standards. Designing and implementing MANETs can be quite complex due to the inherent nature of wireless communications and adverse effects, such as channel capacity variations, routing failures, and erratic power control, can wreak havoc with these implementations. The developed model, built on scalable, real-world scenarios from academia, government, and industry, will determine a compromise solution for these attributes and their non-linear trade-offs. After depicting representative models, network simulations and subsequent empirical data compilations will be analyzed for optimization purposes. Since solving the optimization problem will be highly complex, characterizing the salient QoS parameters using classical gradient-approach methods were considered unsuitable and impractical. Rather, a Non-Polynomial-complete model using evolutionary programming tools was conceived. The corresponding fitness function will represent these parameters through the use of computation intelligence techniques, preferably fuzzy logic associative memories and genetic algorithms. The outcome is expected to benefit design and implementation of data link protocol and network layer routing algorithm improvements for QoS control.