Speed estimation techniques in cellular systems: unified performance analysis

Estimation of the mobile speed, or equivalently, the maximum Doppler frequency, is of importance in a variety of applications in wireless mobile communications. In this paper, a unified framework for the performance analysis of several major speed estimation techniques is presented, which allows a fair comparison between all the methods, analytically. Interestingly, it is proved that all these methods are equivalent, asymptotically, i.e., for large observation intervals. In addition, we have derived closed-from expressions for the bias and variance of a recently proposed covariance-based method. We have also introduced a new estimator which relies on the average number of maxima of the inphase component, and have calculated its variance, analytically. Our extensive performance analysis, supported by Monte Carlo simulations, have revealed that depending on the channel conditions, one needs to use a crossing- or a covariance-based technique, to achieve the desired estimation accuracy over a large range of mobile speeds.