In mobile crowd sensing networks data forwarding through opportunistic contacts between participants. Data is replicated to encountered participants. For optimizing data delivery ratio and reducing redundant data a lot of data forwarding schemes, which selectively replicate data to encountered participants through nodes data forwarding metric are proposed. However most of them neglect a kind of redundant data whose Time-To-Live is expired. For reducing this kind of redundant data we proposed a new method to evaluate nodes data forwarding metric, which is used to measure the nodes probability of forwarding data to destination within datas constraint time. The method is divided into two parts. The first is evaluating nodes whether have possibility to contact destination within time constraint based on transient cluster. We propose a method to detect nodes transient cluster, which is based on nodes contact rate. Only node, which has possibility to contact destination, has chances to the second step. It effectively reduces the computational complexity. The second is calculating data forwarding probability of node to destination according to individual ICT (inter contact time) distribution. Evaluation results show that our proposed transient cluster detection method is more simple and quick. And from two aspects of data delivery ratio and network overhead our approach outperforms other existing data forwarding approach.
Download