Power Allocation Algorithm with the Consideration of Fading Channel Gain in Estimation Fusion

In most power allocation systems the fusion performance, such as the Mean Square Estimation error (MSE), often changes with the channel fading gain. So their optimal power allocation algorithms are usually not optimal. In this paper, we consider power allocation problem in wireless sensor networks with fading channel gain. The sensors amplify-and-forward their measurements (using uncoded analog transmission) to a remote fusion center over wireless fading channels. The global optimal power allocation algorithm is proposed, and the closed-form solutions for power amplification coefficient of the sensor is proposed. The expression of system power with MSE maximum and the expression of MSE with system power constraints are given, respectively. Channel gain does not appear in both expressions. Monte Carlo simulations are carried out to verify that the proposed scheme can have the fusion performance independent of channel gain.