Study On The Key Technologies To Energy Saving In Wireless Sensor Networks

As the key position in the Internet of Things nowadays, Wireless sensor networks (WSN) has broad prospects and attracts more and more attention from the academia and industry at home and abroad. The limited energy of nodes is a bottleneck problem among the factors which affect the network life of the wireless sensor networks, so energy saving has a great practical significance. There are important theoretic and realistic meanings into this subject.This dissertation investigates the key technologies to improve the energy efficiency of WSN from different aspects and many innovative achievements have been made.In order to solve the hot spot problem near the base station in wireless sensor network which adopts multi-hop communication, a ring-based routing protocol using particle swarm optimization(PSO) is proposed. Learning from the idea from uneven clustering, This protocol divides the monitoring area into clusters of varying sizes using PSO, which makes the clusters farther away from the base station have larger size and the cluster heads near base station have more energy to forward the data from other cluster heads, and thereby improves the imbalanced energy consumption of nodes in different regions. Simulations show that the protocol effectively prolongs the network lifetime. In addition, according to the impact of the number of cluster heads on the network lifetime of WSN, we give an expression about the optimal number of cluster heads for a specific scenario which is useful for the network design and verify it by simulations whose results are well consistent with theoretical values.Aiming at the information redundancy brought about by a large number of redundant nodes in WSN, a distributed source coding algorithm for clustering network is proposed. Based on the characteristics that all cluster members transmit their data to their cluster heads uniformly and the Slepian-Wolf coding theory, the key idea is to make each member transmit its data in accordance with a certain order and compress the data according with the correlation with other adjacent nodes in one cluster, and then cluster head decoded the data sequentially. This algorithm uses an encoding method with modulus value and a corresponding improved low-complexity decoding method in the realization. Simulation results show that the algorithm reduces the amount of data transmission effectively and thereby decreases the energy consumption to a large extent.In order to overcome the drawbacks of WSN which adopts multi-hop communication for long distance communication, an energy efficient beamforming mechanism is proposed. We first analyze the impact of the number of participating nodes, the phase difference of their signal at the target and their transmit coefficients on the energy consumption of the network, based on which a corresponding beamforming design criteria is given. Then we use this criterion combined with the residual energy of each node to select cooperative nodes as well as to determine their emission coefficients. Simulations results show that compared with the existing beamforming mechanisms, this mechanism increases the probability of successful data transmissions and make the energy consumption among the nodes more balanced, and thereby reduce the waste of energy due to the data transmission failures.Considering the long distance communication in the multichannel sensor network, a cooperative OFDM transmission mechanism based on frequency diversity is proposed to overcome the adverse effect of fading channel on the energy consumption of nodes. The mechanism takes advantage of the strengths of OFDM technology such as anti-multipath fading, flexible allocation of resources and high spectrum efficiency, etc. and achieves the frequency diversity by collaborating nodes transmitting the same information through different channels. The residual energy and power constraints of nodes are fully considered when choosing the collaboration nodes, and we use generalized selection combining (GSC) to combine each branch’s signal which minimizes the total energy consumption bottomed on meeting the system performance. This mechanism solves the drawbacks of some deeply fading channels which have little contribution on the diversity gain but lead to high energy cost. Simulations show that this mechanism can effectively extend the network lifetime.Finally, the content of the whole dissertation is summarized, and several valuable research directions of wireless sensor networks are discussed.