Research on velocity estimation of narrow-band mobile communication channels

A new mobile station velocity estimator is presented based on the normalized auto-covariance of the Rician fading signal and that of the squared envelope respectively of the received signal. Focusing on how auto-covariance or squared envelope function of received signal and their derivatives vary as the corresponding parameters change, we find such a truth that estimation from auto-covariance function of received signal is more accurate than that from squared envelope when LOS component is not strong (K is small), and the conclusion is contrary if K is large. Based on the truth we propose a novel velocity estimation method. Compared to previous approaches, the proposed algorithm is shown to have better performance and is robust in a microcellular/macro cellular propagation environment, which can be characterized by no isotropic scattering and/or a specular component of unknown strength. Copyright © 2010 Binary Information Press May, 2010.