The estimation of a random scalar parameter using sensors' noisy data has been studied in the context of wireless sensor networks. In energy-constrained networks, the sensors are restricted to low transmission rates to preserve battery power. Hence, in our distributed estimation method, only one bit of a B-bit quantization is sent from each sensor to the fusion center, where the unknown parameter is estimated. To optimize the estimation performance in an inhomogeneous sensor network, we find the best bit-sensor assignment based on the sensors' noise variance. In our method, we use the Hungarian algorithm for the assignment optimization problem.