In delay tolerant WSNs mobile ferries can be used for collecting data from sensor nodes, especially in large-scale networks. Unlike data collection via multi-hop forwarding among the nodes, ferries travel across the sensing field and collect data from sensors. The advantage of using a ferry-based approach is that, it eliminates the need for multi-hop forwarding of data, and as a result energy consumption at the nodes is significantly reduced. However, this increases data delivery latency and as such might not be suitable for all applications. In this paper an efficient data collection algorithm using a ferry node is proposed while considering the overall ferry roundtrip travel time and the overall consumed energy in the network. To minimize the overall roundtrip travel time, we divided the sensing field area into virtual grids based on the assumed sensing range and assigned a checkpoint in each one. A Genetic Algorithm with weight metrics to solve the Travel Sales Man Problem (TSP) and decide on an optimum path for the ferry to collect data is then used. We utilized our previously published node ranking clustering algorithm (NRCA) in each virtual grid and in choosing the location for placing the ferry’s checkpoints. In NRCA the decision of selecting cluster heads is based on their residual energy and their distance from their associated checkpoint which acts as a temporary sink. We simulated the proposed algorithm in MATLAB and showed its performance in terms of the network lifetime, total energy consumption and the total travel time. Moreover, we showed through simulation that nonlinear trajectory achieves a better optimization in term of network lifetime, overall energy consumed and the roundtrip travel time of the ferry compared to linear predetermined trajectory. In additional to that, we compared the performance of your algorithm to other recent algorithms in terms of the network lifetime using same and different initial energy values.
展开▼
