| Wireless sensor network is a forward multidisciplinary research field formed with the rapid development of wireless communication, integrated circuit, sensors, micro-electronics, and other technologies. It has data-centered, self-organization, adaptive, distributed and other characteristics. In wireless sensor network, network nodes have weak communication capability and limited power due to their small size, which makes the performance of routing algorithm essential to wireless sensor network. Routing algorithm is not only the basis of realizing energy balance and congestion control, but also the key to keep network connective and improve the reliability of data transmission. Since wireless sensor networks and applications are closely related, with the application environment from the traditional outdoor environment gradually extended to the indoor environment, it will set a higher request to the wireless sensor network and the routing algorithm.This paper focuses on the new research domain of control discipline, theories and technologies involved in wireless sensor network. The classic wireless sensor network routing algorithms most took outdoor environment as research background, and difficult to adapt to multi-separation, multi-obstacle, dynamic environment and other characteristics in indoor application. Based on iterative optimization technique, optimization theory, swarm intelligence optimization, graph theory, and other computing methods, taking indoor environment as specific background, researches on indoor environment oriented routing algorithm and application technology are carried out. As an innovative work, the paper seeks breakthroughs on research methods and ideas. The important research results are as follows:①In indoor environment, there are walls, doors and other obstacles, propagation loss factors between nodes are different, and change with their surroundings, which make propagation loss factors difficult to estimate by the distance between nodes, and seriously influence the results of routing algorithms.Based on this, an Adaptive Indoor RSSI Estimation (AIRE) algorithm for wireless sensor networks based on Kalman filtering theory was proposed. According to the instability of RSSI, AIRE improved and simplified Kalman filtering algorithm to realize iterative optimization of RSSI observations, and thus, insured exactly RSSI estimation under constant environment. It also built judgment model of environmental parameter jump, and realized self-adaption to the change of environment. Simulation and experiment results show that when environment jumps, classic sliding window algorithm has adaptive ability but couldn't achieve convergence, Kalman-simple algorithm converges slowly, the convergence rate of AIRE is about 20 times than that of Kalman-simple, which is more suitable for wireless sensor network that applies to indoor environment.②Aiming at the variability of indoor environmental wireless Channel Parameters, limited adaptive ability and difficult to ensure network energy-balanced and Survival time in dynamic environments of classics wireless sensor network routing algorithms, a Weight coefficient Adaptive based Indoor Energy load-balanced Routing(WAIRE) was proposed based on AIRE algorithm.WAIRE used AIRE to adaptively and exactly estimate propagation loss between nodes, formulated the routing setup process as the typical multiple attribute decision making process, and established multi-attribute energy-balancing model, made a unitary processing for initial data to solve incommensurability among data, realized adaptive dynamic adjustment of weight with the change of data by using an integrated objective with subjective weight method and entropy weight coefficient method. The algorithm took many energy related parameters as decision-making attributes, chose an optimal node to relay the data, and ensured the dynamic balancing of network energy. Experiments show that WAIRE's average residual energy is 2.7% lower than MTE, 1.1% higher than WAIER-1, but the energy distribution of WAIRE is more even. WAIRE's relative entropy of residual energy is 0.07 times than MTE, 0.46 times than WAIER-1, the life time is 1.93 times than MTE, 1.15times than WAIER-1. It effectively balances network energy consumption and prolongs the network lifetime in indoor application.③When the wireless sensor network applications in the indoor environment, due to a closer link with the people, the quality of service routing algorithm (QOS) has put forward higher requirements, however,dynamic changes of wireless channel parameters in the indoor environment not only influences the network energy consumption, but also influences QOS related parameters.when applied to indoor environment, the existing QOS based routing algorithms couldn't give consideration to both adaptive ability and network energy consumption. Based on this problem, A Discrete particle swarm based Adaptive QOS Routing (DAQR) for wireless sensor networks that integrates AIRE results with WAIER results was proposed.DAQR defined different QOS parameters as different optimization objects, the QOS routing setup process was abstracted to multi-objective optimization process. It worked out routing setup oriented multi-objective discrete particle swarm optimization strategy based on classic particle swarm algorithm, genetic algorithm, niche algorithm and entropy method, realized local routing optimization. Simulation results show that DAQR's QOS parameters performs well when environmental parameters changes, whenηvalues 20%, the network lifetime of DAQR-1 is 28.1% higher than EQR, 131.8% higher than SAR. When network scale expands, DAQR shows better expandability, whenηvalues 5% and the number of nodes is 500, the network lifetime of DAQR-1 is 17.3% higher than EQR and 67.1% higher than SAR. It proves that DAQR's QOS parameters can effectively adapt to the change of environment, and shows better expandability when network scale expands.④Based on the research results of indoor-oriented routing algorithm, building energy-efficient monitoring prototype system that applies to indoor environment was built, that is, indoor environmental multi-modal monitoring prototype system based on wireless sensor network. Aiming at the new characteristics of Indoor dynamic network topology, the efficient routing algorithms based on AIRE and WAIRE were designed and realized, then, dynamic topology that has strong environmental adaptive ability was built, which reduces the system's energy consumption, and balances nodes'energy distribution.
|
PDF Full Text Request