Research On Optimal Energy-saving Technology For Wireless Sensor Network Based On Cross-layer Design

The rising of wireless sensor network technology has brought a revolutionizedchange to information perception, acquisition and processing technology. Nodes inwireless sensor network are so small; typically supply itself by its own battery power.Consequently, there are limited of energy, communication and computing capacity,storage because of the hardware design of nodes in wireless sensor network.Meanwhile, the shared channel between adjacent nodes easily lead to conflicts whentransmitting and receiving data, which can cause packet loss, thus, increasing thedelay of retransmission and consumption of energy.Since the independence between layered design in the traditional hierarchicalarchitecture of the system, layers keep the transparent between each other, whichpartly simplify the complexity of design in the network protocol, while the difficultyof the realization of the system is reduced. However, there are no interactionsbetween the layers in layered approach, network system can not well integrated toachieve energy efficiency by the information integration between layers, so that theoverall performance of the network has not been effectively improved.Relative to the other layers, the MAC layer and the network layer are the mostimportant, and they cost the more energy consumption in sensor nodes at the sametime. The design of these two layers can be very significantly affect the energyconsumption, thus related to the overall system performance. Consequently, we cananalysis the system by combining the MAC layer and the network layer, usingcross-layer protocol to combined optimization.On this basis, we analyzed the power control strategy and clustering algorithmwhich can effectively improve both the performance of the wireless sensor networkin detail. Considering the existing clustering algorithms typically need moretheoretical conditions, can not adapt to dynamic network changes. Furthermore, theapplication of the algorithms are not so strong, and not easy to achieve in a realenvironment. In view of this situation, we design a link clustering algorithm whichcan dynamically adapt to network change, and is more energy-efficient, also morestable. Meanwhile, a power control mechanism with optimum transmitting powerand the ability of reversing effect to interfering node is applied to the clusteringalgorithm. On the basis of clustering algorithms, by using optimizing the powercontrol technology, we designed a cross-layer optimization protocols--power controland stable efficient clustering protocol(RPC-SEC) of wireless sensor networks, it achieves a good reduce of the energy consumption, extends the throughput at thesame time. Simulation results show that, compared with traditional typical protocol,the cross-layer protocol in this paper has better energy efficiency in wireless sensornetwork system, and also expands the throughput of the network system, on the basisof ensuring good system stability.