Resource Allocation Optimization Of Wireless Sensor Networks For Railway System Condition Monitoring

The construction and the operation of railway system have stepped into the phase of rapid development in China.Faster speed and longer operating mileage of the railway system,which brought more challenges for the operation safety and reliability.Hence,it is necessary to monitor the operating condition of the train,infrastructure,and the corresponding environment on-time,and then forecast their operation situations accurately.The operating environment of the railway in China is very complex,and the cost of the traditional wire monitoring system constructing and maintaining is always very high.Furthermore,it is very difficult to deploy the power supply and cable communication system in the remote area.Hence,the wireless monitoring network for the railway monitoring system is proposed in this dissertation,and the research focused on the energy and bandwidth allocation optimization which is the determining factor to maintain the life cycle and transmission ability of the whole monitoring network.The energy is the basis of the lifetime maintenance,and the bandwidth will determine the reliability of information transmission.However,the energy storage and bandwidth of the sensors are limited,which restrict the continuous operation of the wireless monitoring system and the reliable transmission of the inspected information.Based on the practical demands,the energy and bandwidth allocation strategies are proposed in this dissertation to promote the efficiency of the limited energy and bandwidth resources.The system lifetime and the reliability of the information transmission are enhanced,and the comprehensive utility of the wireless monitoring system is improved correspondingly.Currently,the application of the wireless sensor network is rarely focused on the field of railway monitoring,and the research solutions in some local applications verified that the energy and bandwidth management schemes are the determining factors for its feasibility and availability.Moreover,the researcher is just beginning to do some research in this field,and the systematic theoretical and technical results have not yet been developed.The dissertation aims to solve the issues in system lifetime and bandwidth efficiency maximization according to the research framework as "system construction-energy allocation optimization-bandwidth allocation optimization".The effectiveness of the proposed schemes is verified through the scenarios based simulation and data analysis.The main innovations are as following:(1)The rail wireless monitoring system with "subnetwork-backbone network"two-layered hierarchy is designed according to the characteristics of the railway infrastructure and operation environment,the structure of the existing wireless communication system and the deployment of wireless sensors network.The two-layeredly hierarchical system provided a basic information transmission frame in a high-efficient,energy saving,reliable and safe way.(2)The utilization of the limited energy is efficiently and dynamically allocated through communication protocols optimization and optimized data aggregation,so as to prolong the life cycle of the sensor network through improving the energy efficiency.?The optimized cluster-based communication protocol is proposed and verified to support the energy-efficient allocation in whole sensor network scale referred as the sub-network.The minimization and equalization of the energy consumption for all nodes in the sub-network are guaranteed according to the cluster heads(CHs)selection and clusters optimization,and the sub-network lifetime is maximized simultaneously.First,the times of the sensors selected as the(CH)in previous communication rounds,the residual energy,and the predicted energy consumption are all taken into consideration in the CHs selection.Second,the clusters are re-optimized,considering the clusters scale and the corresponding relationship between the CHs and non-CHs.?The data aggregation ratios optimization scheme is proposed and verified to support the energy-efficient allocation in the sub-network scale referred as the CHs.The total energy consumption of the CHs(energy consumption in information aggregation and communication)is minimized according to optimizing the data aggregation ratios,and the sub-networks lifetime is improved correspondingly.First,the effect of the aggregation ratios to the energy consumption in information aggregation and communication is studied.Second,the data aggregation ratios dynamic optimization model is designed based on the characteristics of the CHs' communication loads and distances.On this basis,the CHs will keep working with the lowest energy consumption,and the lifetime of the CHs and sub-network is extended.?The optimized multi-hop communication protocol is proposed and verified to support the energy-efficient allocation in the whole sensor network scale referred as the backbone network.The comprehensive effectiveness of the backbone network is maximized according to optimizing the multi-hop links and the data packets in each link,and correspondingly.First,lifetime and communication latency of the backbone network are maximized and minimized respectively through the comprehensive effectiveness optimization model.Second,the multi-hop links scheduling scheme is designed to optimize the sequences of the links,which could obtain the further reduction of the communication latency.(3)The utilization of the limited bandwidth is efficiently and dynamically allocated through system bandwidth utility optimization,so as to guarantee the reliable transmission of the monitoring information through improving the bandwidth efficiency.First,the real-time,priority,and importance of all the monitoring objects are analyzed in advance;second,the system utility model is designed based on the bandwidth utilization efficiency and the communication demands.Finally,the particle swarm optimization(PSO)algorithm is used to obtain the optimal allocation solutions for each monitoring objects.To summarize,this dissertation focuses on the optimization of the limited resources allocation in the railway wireless monitoring system.The communication protocols optimization model,data aggregation optimization strategy,and the bandwidth allocation optimization scheme are all proposed in this dissertation,so as to improve the utilization efficiency of the limited energy and bandwidth resources.The research may help to bridge the gap in the current literature on wireless monitoring network in the railway system.Besides,the conclusions of this dissertation may support the decision makings of practitioners,managers and researchers for the railway trains,infrastructure and environment wireless monitoring.