A proactive management algorithm for self‐healing mobile ad hoc networks

Abstract

SUMMARYThe ability to proactively manage mobile ad hoc networks (MANETs) is critical for supporting complex services such as quality of service, security and access control in these networks. This paper focuses on the problem of managing highly dynamic and resource-constrained MANET environments through the proposal of a novel proactive management algorithm (PMA) for self-healing MANETs. PMA is based on an effective integration of autonomous, predictive and adaptive distributed management strategies. Proactive management is achieved through the distributed analysis of the current performance of the mobile nodes utilizing an optimistic discrete event simulation method, which is used to predict the mobile nodes' future status, and execution a proactive fault-tolerant management scheme. PMA takes advantage of distributed parallel processing, flexibility and intelligence of active packets to minimize the management overhead, while adapting to the highly dynamic and resource-constrained nature of MANETs. The performance of the proposed architecture is validated through analytical performance analysis and comparative simulation with the Active Virtual Network Management Protocol. The simulation results demonstrate that PMA not only significantly reduces management control overhead, but also improves both the performance and the stability of MANETs.