Research On Dynamic Slot Allocation Algorithms Based On Path In TDMA Ad Hoc Networks

A Mobile Ad Hoc Network(MANET)is a kind of special wireless network,and it requires no fixed infrastructures and consists of a group of mobile network nodes with the same functions.Not only the nodes in a MANET act as information terminals that generate and process information,but do like the routers that are able to forward packets for enhancing the network coverage.Due to these characteristics,MANETs are suitable for emergency scenarios such as battlefields and rescues.In a MANET,network nodes share the wireless channel with their neighbors because of its broadcast feature,and it definitely results in channel contention and collision among these nodes.Therefore,how to efficiently implement allocation of channel resource is one of its core technologies,and is also the key to improve network performance.A MANET features distributed networking control,and all MAC protocols are run from the perspective of link level.Consequently,in a multi-hop scenario,it is difficult to guarantee end-to-end Qo S due to the contention and collision among nodes on the path.Aiming at the problems mentioned above,a path-level dynamic time slot allocation algorithm-PL-TDMA is proposed.In the PL-TDMA,the source node of the business flow obtains the information of business delivered by the application layer,including the arrival rate,the information of path,and the like.Then each node on the path of business summarizes its slot information along the path to the previous hop of the destination node which performs the calculation of the slot allocation.The allocation algorithm runs in turn to ensure that each node obtains the same number of time slots,and the number of time slots obtained by the bottleneck nodes is increased by one-step back-off in the allocation process.finally the result is distributed to these nodes to reserve and occupy the time slot.The algorithm improves the end-to-end throughput while improving the effective utilization of time slots and shorting the time to complete allocation of time slot.The end-to-end throughput and the time to complete allocation of time slot are used as performance metrics.The PL-TDMA algorithm proposed in this paper is modeled and simulated on the OPNET,and compared with the optimal solution of the global allocation and the USAP protocol.These simulation results show that the end-to-end throughput of PL-TDMA approaches the optimal solution.Compared with USAP protocol,it performswith higher end-to-end throughput,higher effective slot utilization,and shorter time to complete allocation of time slot.