Research on vertical handover in LTE two-tier Macrocell/Femtocell Systems based on fuzzy neural network

Abstract

This paper proposed a smart handover algorithm based on fuzzy neural network. In LTE Two-tier Macrocell/Femtocell Systems, the fuzzy neural network is used to select the appropriate network which to access to taking several factors into consideration. As radius of Femtocell coverage is small, it is essential to introduce UE speed pre-judgement module in order to reduce unnecessary handovers. In the offline phase, information of UE speed, Femtocell and Macrocell RSS(Received Signal Strength), network load are collected to establish an initial database. Fuzzy neural network is trained by means of BP algorithm with training data which has been pre-processed. After establishing fuzzy neural network, information of UE speed, Femtocell and Macrocell RSS, network load is mapped with network accessing decipline. In the online phase, collecting user real-time measurement of the three-dimensional information above as the fuzzy neural network input, result of handover decision will be given through fuzzy neural system. Experimental results indicate that this algorithm can achieve a lower Ping-pong handover rate than other handover algorithms, such as RSS based algorithm and Femtocell first algorithm.