To Drop or Not to Drop: Receiver Design Principles for Estimation Over Wireless Links
In this paper, the authors consider estimation of a multiple-input multiple-output dynamical system over a wireless fading communication channel using a Kalman filter. They are interested in finding the optimum receiver design in terms of handing noisy samples. They reformulate the estimation problem to include the impact of stochastic communication noise in the noisy packets. They will show how the eigenvalues of the state transition matrix A affect the optimum receiver design. They prove that, in the absence of a cross-layer information path, packet drop should be designed to balance information loss and communication noise in order to optimize the performance.