Error correction mechanism in wireless networks protocol is complicated

which aggravate band burden of wireless communication. Blind source separation can reconstruct source signal without any prior knowledge

which becomes a hotspot in the field of wireless sensor networks（ WSNs）. However

it is difficult to determine the number of quantization bits and Modems in the network deployment. Aiming at above problems

propose a blind source separation algorithm based on square root volume Kalman filtering（ SRCKF）. Algorithm estimates separated vector with square root volume using Kalman filtering. The experimental results show that SRCKF algorithm can effectively calculate the best quantization bits and Modem ’s number. The efficiency of SRCKF algorithm is 1. 3578 times that of similar UKF algorithm.