Mobile Elements Scheduling For Data Collection In Wireless Sensor Networks

Wireless sensor networks (WSNs) connect human beings with the physical world efficiently, which make people's life more convenient. As a fundamental issue in WSNs, data collection affects the network performance and quality of service (QoS) directly. As most of the collected data is time-related, it is necessary for real-time applications to collect sensor data as soon as possible, respond to the incident in time, and avoid data becoming invalid due to high delay. In these applications, this thesis mainly focuses on mobile elements scheduling in data collection on the premise of data freshness. The main work and contributions are as follows:Since existing data collection algorithms have the problem of excessive energy consumption of nodes, this thesis makes use of mobile elements (ME for short) to support data collection. It calculates the best positions for ME to receive collected data by analyzing the linear model, and proposes an efficient carrying routing algorithm. The algorithm, which combines with network data transportation, makes maximum use of ME and reduces the energy consumption of sensor nodes. The simulation results show that the proposed algorithm outperforms the existing data collection algorithms in terms of energy consumption under the condition that all data freshness requirements are satisfied.This thesis proposes two heuristic algorithms to solve the DFMES (Data Freshness Mobile Elements Scheduling) problem. The first algorithm, which performs better when data freshness constraints are relatively weak, starts from large tours solved by TSP (Travelling Salesman Problem), and proceeds to cut segments from these tours to form smaller ones until all time requirements are met. While in strict data freshness constraints networks, it plans the ME paths in a greedy manner:it starts by building short ME tours started from the sink and then expands them with the lowest cost nodes as much as possible, always maintaining the property that no tour violates the time constraints of the nodes it visits. Experiments indicate the superior performance of less ME and less data collection time of the proposed algorithms compared with other heuristics.The proposed routing and scheduling algorithms in this thesis can effectively increase energy efficiency, reduce mobile elements cost, and satisfy the data freshness requirements. It is valuable for the related research on data collection in WSNs.