Design and Optimization of a Hybrid Sensor Network for Traffic Information Acquisition

With the development of sensor and network technologies, a type of hybrid sensor network (HSN) has been developed in traffic information acquisition systems. Sensor types in HSNs are diversified, mostly by batteries or solar energy as their energy source. Energy determines their lifetime. If energy consumption of sensors can be reduced from the aspect of communication, their energy consumption can be saved and lifetime can be increased, so as to improve the benefit and efficiency of the whole network. To optimize the efficiency and consumption of HSNs, a graph theory-based model was proposed by considering nodes' connectivity power and the energy consumption of data transmission. Targeting different data volumes, two undigraph-based algorithms (PMST and DSPA) were introduced to solve the proposed optimization model. For a low data volume, the minimum spanning tree network based on PMST will be the optimal topology of an HSN. For a high data volume, the shortest path network based on DSPA will be the optimal topology. A numerical example of comprehensive traffic information acquisition at a signalized intersection was used to verify the proposed algorithms. The solution process and final optimal topologies based on the PMST and the DSPA were shown. Moreover, the energy parameters of different optimal topologies of HSNs were discussed, and it was proved that the optimal topology of an HSN is dependent on the data volume of each sensor node.