In wireless sensor networks (WSNs), each sensor node can estimate the global parameter from the local data in a distributed manner. This paper proposed a diffusion estimation algorithm based on sub-regions, which are referred to as the SR-DLMS algorithm. The proposed algorithm divides the sensor network into several sub-regions, which used to estimate sub-dimensions of the parameter. Compared with the diffusion LMS algorithm, communication cost of the proposed algorithm is greatly reduced without affecting the estimation performance. Numerical simulations are given to verify the effectiveness and advantages of the proposed algorithm.