Research On The Optimal Energy Saving Algorithm Of Wireless Sensor Networks Based On Dual Therory

Wireless sensor networks(WSNs), which is composed of the convergence of sensor, wireless communications, automation, system optimization and other multi-disciplinary technology, is an effective tool for state-aware, information gathering and target tracking. There are great potential applications with WSN in many fields. Therefore, WSN has become a hot research topic in recent years. Although there are many invaluable applications for wireless sensor networks, there are also a lot of emerging problems and challenges that need to be solved, at the same time. The principal problem is how to efficiently use resources to prolong the overall system lifetime of such high energy-constrained wireless sensor networks.In this paper, based on the dual optimal theory, we mainly focus on the design of energy-efficient routing protocol and rate control in wireless sensor network.Firstly, considering the data redundancy when sensor nodes send information in wireless sensor network, we introduce the data aggregation technology to routing protocol design. The concept of data aggregation rate is proposed to denote the compressing capability of a sensor node in the sensor networks, and than a new distributed algorithms based on data aggregation is analyzed, to perform global optimal in traffic planning to prolong system lifetime as much as possible. The problem is formulated as a linear convex programming problem and dual algorithm is used to solve it. The proposed algorithm can effectively reduce data traffic, balance energy consumption, and prolong the network lifetime.The network lifetime and application performance are two fundamental, but conflicting, design objectives in wireless sensor networks. Secondly, we investigate the trade-off relationship between network lifetime maximization and application performance maximization. The paper introduce the network utility maximization(NUM) framework to describe the amount of data collected by sensor nodes, thus we get the criterion to evaluate the network application performance. To consider two sides of network performance synthetically, an optimal model is set up based on the trade-off factor. In this paper, the problem is formulated as a combinatorial optimization problem and primal-dual approach is used to achieve the optimal solution. The distributed algorithm for the optimal sending rates is given further. Simulations results indicate that a fundamental tradeoff between network lifetime and network utility is achieved by adjusting the parameters.