SVD And Graph Based Cooperative MIMO Transmission Scheme For WSN

Wireless sensor networks (WSN) is an autonomous system composed by a largenumber of micro-sensors, which integrated with data collection, data processing andcommunication capabilities. These micro-sensors are usually battery operated andwork in harsh environment. Charging or replacing the battery is very inconvenient.In addition, the more frequently the radio channel state changes, the greater the biterror rate is, which leading to a lot of energy waste in the data retransmission.Therefore, how to design a transmission strategy, which not only can guarantee acertain transmission quality but also energy-efficient, has become a key issue.Cooperative multiple input multiple output (MIMO) is a communicationtechnology in which the data collected by sensor nodes are transmitted by the virtualantenna array composed by a set of cooperative nodes. Cooperative MIMOtransmission can effectively reduce energy consumption and improve transmissionquality. It has become a hot topic in recent years.Based on existing research, we have done the following researches:(1) The clustering algorithm in WSN: we consider the spatial distribution ofclusters, the nodes’ residual energy, the number of cluster head nodes and the totalenergy consumption and expand the multi-objective evolutionary analysis. Then amulti-objective evolution based clustering method is proposed. Simulation resultsshow that the proposed method can give the optimal clustering solutions and therelevant parameters used in engineering practice.(2) Considering that the channel conditions is random changed, we propose SVDbased cooperative nodes selection algorithm. Simulation results show that theproposed method can improve the system channel capacity, and reduce bit error rate.(3) Taking into account that in large-scale WSN, multiple-hop transmission isusually employed to save energy, we propose a graph based multiple-hopcooperative MIMO transmission scheme. Simulation results show that the proposedscheme can significantly improve survival nodes and extend network lifetime.