Dynamic multi-stage tandem queue modeling-based congestion adaptive routing for MANET

Abstract

In Mobile Ad hoc Network (MANET), Congestion is considered as the significant issue of concern as they influence router based network functionalities achieved by the intermediate nodes participating in routing. Congestion generally occurs due to limited availability of network resources such as bandwidth and energy which lead to significant packet loss, end-to-end delay and communication overhead. Hence, the core objective of congestion control algorithms was to utilize the allocated network resources and to maintain the load below an acceptable threshold capacity. This paper proposes a Dynamic multi-stage Tandem Queue modeling-based Congestion Adaptive Routing (DTQCAR) based on the estimations of average threshold level of congestion. The decision of congestion policing is achieving the current congestion level and sending warning messages to all their neighbors for dynamic adjusting of packets forwarding. The estimation of average threshold level is based on Dynamic multi-stage Tandem Queue modeling that inspires stochastic probability. Depending on the elucidated stochastic factors, neighbors attempt to locate a congestion-free alternative path to the destination. The simulation results also confirm that DTQCAR on average showed a better performance of 13%, 17% and 19% than the existing LBCAR, DCDR and EDAPR congestion control algorithms in terms of packet delivery ratio. Keywords: Stochastic factors, Multi-stage tandem queue, Congestion, Average queue length, Stochastic probability