A Field Experimental Evaluation of Mobile Terminal Velocity Estimation Based on Doppler Spread Detection

Abstract

To deal with the recent explosion of mobile data traffic, heterogeneous cellular network (HetNet), in which a large number of small cells are deployed in a macro-cell coverage area, is considered to be a promising approach. However, when a mobile terminal (MT) traveling at a high velocity moves through plural small cells in a short period of time, the frequent handovers (HOs) lead to a deterioration of user quality of experience. To avoid such HO problems, while improving the network capacity in the cellular it is effective to introduce an inter-layer HO control policy where MTs travelling at high velocities are connected to the macro-cell layer to reduce the number of HOs and MTs travelling at low velocities or which are stationary are connected to the small-cell layer for offloading traffic from the macro- cells to the small-cells. However, to realize such inter-layer HO control policy in the layered cellular network, it is crucial to estimate the velocity of each MT. Due to the technological constraints of MT velocity estimation based on the Global Positioning Systems (GPS), we focus on MT velocity estimation algorithms which do not require information provided by GPS. First, we discuss the issues of the existing MT velocity estimation algorithms and then introduce a MT velocity estimation algorithm based on Doppler spread detection. Meanwhile, there have been few studies on evaluating Doppler spread based MT velocity estimation techniques in actual radio propagation environment. Therefore we evaluate the introduced MT velocity estimation algorithm in actual environments and show that good MT velocity estimation accuracy is achieved in both line-of-sight (LOS) and non-line-of-sight (NLOS) outdoor propagation environment.