A Markov-based Framework For Handover Process in Heteroegenous Cellular Networks

Abstract

Heterogeneous networks (HetNets) are a combination of small & large cells that are communicating to other networks just like a single cell of the network. This requires smooth transition of communication link when the user moves from one cell to another cell region of the HetNets. This process is called handover (HO). There are several issues in a handover process such as unnecessary HO and failure of HO which affect the performance of communication. Handover occurs when signal to noise ratio is less than a predefined threshold. In this paper, we analyze the performance of handover under different parameters, such as received signal strength (RSS), hysteresis margin (HM) and Cell Range Expansion (CRE). A Markov based framework is used to model the handover (HO) process for the user equipment (UE) by defining different positions of the UE at a possible trajectory as the Markov states. As the UE moves forward along its trajectory, the path loss of the signal changes and therefore SINR changes accordingly. This behavior is modelled in Markov chain by using transfer probabilities, where at a particular state the UE can stay in the same state, move to the next state or return to the first state. A few transitions of the Markov chain from a given initial state can specify the handover rate of the network. To analyze the effect of the above parameters on handover rate, we consider a single macro cell and femto cell with a UE following a specific trajectory towards the femto cell. Distances of UE along the trajectory from both macro and femto cell are computed that are used to find distance dependent path losses. This allows us to compute received signal strength which in turn identifies the transfer probabilities of each state. Finally the performance of handover rate is analyzed through transitions of the Markov chain. It is observed that the handover rate decreases as the values of CRE, HM and number of states increase.