Research Of Routing And Information Dissemination Mechanism Based On The Bio-inspired For MANET

As the MANET is independent of a fixed infrastructure and has unpredictabledynamic change in the topology, its route strategy has a complete transformation from thetraditional wired networks, and many routing protocols have been proposed in the past. Inthe MANET, the route node may not take the load balancing into account, which possiblyform local "hotspots" zone to lead to select an extremely congested node as the next-hop.The congested next-hop will result in frequently re-routing, trigger more route controlpackets, aggravate the network congestion, and consume a large amount of networkresource. Consequently, the congestion nodes possibly incur routing instability andre-routing. In the metabolism network of bacteria, the Host Escherichia Coli cell would toregulate its metabolic synthesis to keep the high cell growth rate in the dynamic nutrientenvironment. There is no explicit rule-based mechanism in the metabolism behavior, andit provides inspirations for mitigating the side-efforts of the congestion nodes. Inspired bythe adaptive metabolism behavior, a bio-inspired congestion-avoidance routing protocolATAR is proposed in this thesis to achieve that the route node can always adaptivelyselect one of light-load neighbor nodes with shorter path to forward the packet, and theATAR is provide a new solution for solving the route design problems in MANET.At first, this paper detailedly introduces the adaptive metabolism behavior and itsinspired mathematic biology model ARAS. Moreover, the randomicity of the noise-drivenstochastic approach in the original model ARAS is also introduced, and this drawbackwould lead to a non-effective node selection in the routing design. These works preparethe theoretical basis and references for the bio-inspired routing recovery algorithm and thedesign of adaptive routing strategy.Secondly, this paper concerns about the impact of congestion node while preservethe character of "Shortest Path", and utilizes the adaptability feature of the bio-inspiredmechanism ARAS in the routing design. The available queue buffer of node and the hopcount of feedback packet are integrated as a new bio-inspired routing decision metric, i.e.,the next-hop fitness, which can be denoted as the congestion level of node and the lengthof routing path. And then, this paper eliminates the ARAS randomicity by redefine themodel and the meanings of parameters, and present a new bio-inspired routing recoveryalgorithm based the improved ARAS. It can cut down the processing delay, and quickly and accurately recover from the failure without using additional broadcast controlpackets.At last, the route nodes would carry out the congestion aware and decide whether toforward the date or execute local recovery based on the next-hop fitness. In the localrecovery phase, the nodes would implement the bio-inspired routing recovery algorithmto attain the objective of route convergence. Through the proposed routing scheme, thenode can always adaptively select one of light-load neighbor nodes with shorter path toforward the packet, and it could achieve the higher successful-delivery ratio and the lowerETE average delay. Especially, it can efficiently mitigate the congestion and implementthe load-balancing.The main contribution and innovation of this article include the following: Firstly,based on the mathematic model ARAS, this paper utilizes a cross-layer design method topresent a new bio-inspired routing decision metric. This route metric denotes thecongestion level of node and the length of routing path. Secondly, this paper eliminatesthe ARAS randomicity, and presents a new bio-inspired routing recovery algorithm basedon the improved ARAS to quickly and accurately recover from the failure without usingadditional broadcast control packets. Finally, using the bio-inspired route metric in therouting-decision along with the bio-inspired routing recovery, this paper present a newadaptive congestion-avoidance routing protocol, and it allows the routing-decision nodecan always adaptively select the light-load node as the next-hop with shorter path.