Investigation On Key Technologies Of Distributed Estimation And Cooperative Transmission In Wireless Sensor Networks

The third revolutionary wave in information technology, represented by Internet of Things (IOT), is arriving and deeply affecting the manners of our work and life. Wireless Sensor Network (WSN) technologies guarantee global interconnection and communication capacities of IOT and received extensive concerns in the areas of information terchnology. The WSN is an ad hoc network with limited resources, wireless multi-hop transmission and unreliable nodes, whose performance can be improved by multi-nodes cooperations.This dissertation investigates the multi-nodes cooperation problem of WSN, with an emphasis on the distributed estimation and cooperative transmission technologies. The main contributions of the dissertation are as follows:1. Based on linear minimum mean square error estimation (LMMSE) priniciple, the distributed parameter estimation model was established for the parallel and distributed WSN architecture. An efficient cooperative nodes selection and power allocation algorithm (OPA) was proposed using the mean square error of estimated parameter in LMMSE model as objective function. The proposed algorithm was analyzed both theoretically and numerically for two special cases:1) the WSN has power constraint; 2) the WSN as well as individual nodes have power constraints. Simulation results show that the performance of the proposed algorithm is better than the conventional equal power allocation schemes under the same outrage probality requirements.2. Based on maximum likelihood (ML) estimate priniciple, the distributed parameter estimation model was established for the clustered WSN architecture. An iterative parameter estimation method was proposed through message exchanges between the cluster head node and the ordinary nodes in the cluster. As a result, the information transmission between ordinary nodes and fusion center is avoided and the information processing of fusion center is reduced. The proposed iterative estimation algorithm is proved to converge to the ML estimation results. Besides, the proposed algorithm can be implemented in a parallel and distributed manner. Simulation results show that the proposed algorithm has fast convergence speed in the relatively high signal-to-noise ratio(SNR) regions.3. The cooperative transmission problem in clustered WSN using the amplify-and-forward (AF) technology was studied. The performance of dual-hop AF cooperative transmission system was analyzed. The closed-form probability density function (PDF) of instantaneous received SNR with maxium ratio combining (MRC) criteria was derived under both asymmetric and symmetry channels using the moment generating function (MGF) methods. An exact equation and the corresponding upper bound of the ergodic capacity were also given, which are more suitable to describe the general system.4. The outage probability performance of dual-hop AF cooperative transmission in WSNs was investigated. An approximate outage probability formula was derived with all the relay channels having different state parameters. The formula avoids the nonlinear function operation and overcomes high-SNR limitations of the previous outage probability approximated methods. The simulation results show the formula is more closed to the Monte Carlo performance.5. The power allocation problem of dual-hop AF cooperative transmission in WSNs was investigated. By using the priniciples of the proposed OPA, a relay nodes selection and optimal power allocation strategy was presented using the channel capacity as the objective function. The proposed strategy was analyzed both theoretically and numerically for two special cases:1) the relay nodes have total power constraint; 2) the relay nodes have total power constraint as well as individual relay nodes have power constraints. Simulation results show that the ergodic capacity of the proposed algorithm is higher than the conventional equal power allocation schemes. 6. The cooperative transmission problem in clustered WSN using in virtual MISO technologies was studied. The symbol-error-rate (SER) formulas were derived for the phases of intra-cluster broadcasting, inter-cluster cooperative transmission and end-to-end, respectively. Moreover, a low bound on the end-to-end SER was proposed. Simulation results show that system performance can be well described by the low bound. Besides, an optimal power allocation model using the end-to-end SER as constraints was proposed. An efficient method for solving the optimal solution of the model was proposed. Simulation results show that the model has energy-saving advantages under different SER requirements.