Research On Multi-constrained QoS Routing Mechanisms For Wireless Multimedia Sensor Networks

In recent years, multimedia sensor technology is developing rapidly, demand for wirelesssensor networks with multimedia function has become increasingly urgent, so wireless multimediasensor networks come into being. Multimedia data can provide more extensive information tomonitor region, but propose requirements for the network QoS mechanism. In applications ofwireless multimedia sensor network, besides delay and bandwidth, link reliability, the pathoverhead costs and so on should be considered for parameters of QoS constraints. Therefore, theresearch of multiple constrained QoS routing mechanisms for Wireless Multimedia SensorNetworks would be more challenging. The problem of multi-constrained QoS routing isNon-deterministic Polynomial complete problem, and using heuristic algorithm to solve is aneffective way.In this paper, firstly, ant colony QoS routing algorithm of wireless multimedia sensor networksis researched and algorithm performance of various parameters is analyzed, we found out that thealgorithm is of slow convergence and expensive cost. To solve these problems, a clustering antcolony QoS routing algorithm is proposed. Algorithm uses the relationship between the number ofnodes and the number of ants, and introduces the clustering ideas, to reduce the cost of thealgorithm and increase the convergence speed of algorithm by reducing the number of nodes in thenetwork and the number of ants. Additionally, transition probability matrix and pheromoneevaporation rate also have been improved to avoid local optima by clustering ant colony QoSrouting algorithm. Simulation results show that clustering ant colony QoS routing algorithm is moreeffective than basic ant colony algorithm.Then, to solve the problem that the designed multi-constrained QoS routing mechanism needconsider the limit of max life cycle in wireless multimedia sensor network, an genetic QoS routingalgorithm based on optimal energy is proposed. The algorithm can establish a QoS routing byadaptively selecting nodes of larger residual energy in the nodes of multi-QoS constrained, and thusgreatly extend the maximum life cycle of network.Finally, establishing multi-constrained QoS routing would take some cost of time, so researchfor it should consider the establishment speed of routing. In this paper, an improved ant colonygenetic QoS routing algorithm is proposed by using the characteristics of high efficiency in searchof genetic algorithm, and combining it with ant colony QoS routing algorithm. Simulation resultsshow that the algorithm improves the total convergence speed of the algorithm by adjusting thenumber of populations dynamically, choosing algorithm conversing time and update strategy of local pheromone dynamically.