Enhancement of load balanced gateway selection in integrated Internet-MANET using genetic algorithm

Mobile Internet-based ad hoc networking is an interesting domain which involves challenging problems. The fundamental approach in connecting MANET with the internet is through intermediate Gateways. The gateway discovery and selection of an appropriate and suitable route in Integrated Internet MANETs are identified as critical issues. In existing approaches, these two issues are addressed as separate problems. In the proposed approach, a new method, Adaptive Steady Load Balancing Gateway Selection is implemented which is based on Path Load Balancing, and which is implemented by calculating the load along a path and takes into account the route queue length metric. The proposed protocol employs a genetic algorithm to optimize the gateway selection process. The fitness function of the genetic algorithm uses fuzzy logic based on three network parameters. Other salient features of the proposed protocol are: dynamic adjustment of gateway advertisement periodicity and range. The proposed solution is simulated using network simulator NS-2, and its results are compared with the existing approach.