Non-homogeneous Environment For Wireless Multimedia Sensor Networks, Directional Rumor Routing Algorithm

Wireless multimedia Sensor Networks (WMSN) developed from traditional Wireless Sensor networks (WSN), which integrated sensor node has sensation functions of multimedia streaming (e.g. audio, video, image), perform a complete and accurate environment monitoring. The distinct features of multimedia sensor networks are various media category, strong network processing power, large volume of data and complex processing task and balance of energy consume. WMSN has wide applications in environmental monitoring, military inspection, medical nursing, disaster rescue and tracking. The study has great theoretical and practical value, which is one hotspot of Computer research filed.Routing protocol mainly resolve the problem data transmission from source node to sink node, which is one of the core technology in wireless multimedia sensor networks. The energy of multimedia sensor node is mostly provided by batteries, which is limited and hardly replaced. Therefore, how to save the energy of sensor node is very important in the routing design. Because WMSN introduce the video, audio, image and other multimedia information, make the routing protocol design not only to consider the network energy consumption and scalability, but also must consider the multimedia information transmission of real-time and reliability. Rumor routing is a classic routing algorithm based on the random walk of agents. Directed Rumor Routing algorithm integrate rumor routing and part geographical position information to find a near optimal paths to destination, which improve the latency and energy consumption of the traditional algorithm using propagation of query and event agents in straight lines, instead of using purely random walk paths.When a lot of events happen in the network, then the sending number of event agent increase, the agent hop numerical also increase, which lead to a lot of energy is consumed in the network. If the node density is large in the network, every time the algorithm select the shortest distance node in the direction to forwarding agent, then the number of forwarding node is increased and TTL quickly die., which decrease the hit ratio of the path of the query gents and path of the event agents and the reliability of routing discovery. Sending the agent more distance and faster in order to decrease the power consumption and increase the hit ratio of the path of the query gents and path of the event agents. This paper proposes an improved DRR algorithm which aims at deficiency of directional rumor routing (DRR). In contrast to DRR, there are three differences:(1)According to the defect of DRR algorithm about sending the largely number of event agent and query agent to cause TTL quickly die, then reduce the routing discovery probability. IDRR correspondingly introduce virtual sink node and virtual source node in the benchmark direction of event agent and query agent sending direction, aim to use the greedy strategy to make event agent and query agent transmit the far distance before TTL die. IDRR algorithm improves the probability of event agents hitting related query agent, increase reliability and real-time about routing found.(2)According to the largely cost of DRR avoid the routing hole question and do not well adapt the non-uniform environment, IDRR employ the virtual node to select backward area node, the next node is shortest from virtual, the effectively bypass routing cycles. The improve DRR algorithm simplify process of bypass routing cycles, save the cost of carry location of the source node and enhance the ability to adjust the non-uniform network environment.(3) According to the defect of DRR algorithm employ greedy strategy to select the next node in the second layer routing and optimization process, which is easy to cause node energy consumption inequality, then increase routing cycles and reduce the network life cycle. IDRR algorithm using the probability method to decided the next jump node in the second layer routing and optimization process. The probability method mainly contains the distance forward node from destination node and the residual energy of other nodes. The improve DRR algorithm balance the energy consumption , prolong the network lifetime.Through the simulation experiment compare DRR and IDRR algorithm with the probability of event agents hitting related query agent, node energy consumption, energy balance, network life and so on. The results show that IDRR improved and perfect the DRR defect. IDRR can increase the probability of event agents hitting related query agent, enhance the routing discovery real-time and reliability, prolong the network lifetime, and balance the energy consumption more effectively than traditional DRR algorithm.