Mobility prediction in long term evolution (LTE) femtocell network

Abstract

The Long Term Evolution (LTE) femtocell has promised to improve indoor coverage and enhance data rate capacity. Due to the special characteristic of the femtocell, it introduces several challenges in terms of mobility and interference management. This chapter focuses on mobility prediction in a wireless network in order to enhance handover performance. The mobility prediction technique via Markov Chains and a user's mobility history is proposed as a technique to predict user movement in the deployment of the LTE femtocell. Simulations have been developed to evaluate the relationship between prediction accuracy and the amount of non-random data, as well as the relationship between the prediction accuracy and the duration of the simulation. The result shows that the prediction is more accurate if the user moves in regular mode, which is directly proportional to the amount of non-random data. Moreover, the prediction accuracy is maintained at 0.7 when the number of weeks is larger than 50.