Study On Energy Balance And Delay Optimization Of Wireless Sensor Networks

Wireless sensor networks(WSNs), the network composed of a large number ofself-organizing sensor nodes, has showed its great value in various fields such asmilitary, industry and agriculture, and is considered as one of the most importanttechniques in21st century. However, because of the limitation of technique, economyand some other factors, the problems like poor energy and long packet delay stillhamper the development of WSNs.To solve these problems, according to the characters of practical applications ofWSNs, this paper aims to design a novel scheme based on Directed Diffusion: basedon the stored data, the system can detect and predict the change of object, anddetermine the data sending rate based on the prediction. At the same time,energy-aware routing is introduced in our strategy in order to balance the loading ofthe all the sensor nodes and enhance the energy utilization of the whole system, andfurther prolong the lifetime of the WSN. Based on this thinking, this paper proposestwo improved strategies: Directed Diffusion based on Dynamic Data-rated (DD) andDirected Diffusion based on Dynamic Data-rated and Energy-aware (DDEA). Theresults of the experiment proved that DD and DDEA are superior to the traditionalscheme in the aspects of both energy balancing and the life time of system.In order to solve various problems in traditional “Data Mules” policy, such as theissue that the transmission delay of packets is too large as well as the fact that thefairness is difficult to guaranteed in such kind of strategy, this paper proposes toimprove this method by adjusting the working region of Mules and design thetrajectory of the movement of the mule. Through mathematical analysis, we find thatshrinking the working region and the overlapped area can efficiently decrease thetransmission delay for the packets, while improving the fairness of packet transmission.Based on this conclusion, this paper proposes an improved “data mule” scheme, anddesigns several optimized movement trajectories which can be taken as references. Thesimulation results show that the improved method is significantly superior to theprevious one in both aspects of the fairness and the transmission delay for all packets.