Abstract: In order to develop the dynamic query method that suits environment, this paper proposes an optimized join query method that is applicable to limited bandwidth context. The dynamic join optimization problem in multi-hop wireless sensor network is resolved and a cost-based model is designed. Moreover, a self learning and adaptive heuristic method is put forward, which chooses the best algorithm and selective variable value according to the actual situation. This algorithm can predict the selectivity in advance, and can optimize the pairs of data streams. The effectiveness of the proposed method is verified by experiments on standard data sets.

Key words: join query, wireless sensor network, cost-based model, query optimization