Time Synchronization Algorithm For TD-LTE Home ENnodeBs Based On UE Measurement And Assist

With the continuous evolution of communication technology, the performance of the mobile network coverage, capacity and mobility is required to improve constantly. Home eNodeB in mobile communication system is becoming increasingly prominent. The salient feature of the network of home eNodeBs is intensive and unplanned deployment, so interference among home eNodeBs is inevitable. In the TDD system, the sub-frame of adjacent cells is asynchronous or the time slot allocation mode is different, which can cause serious slot interference. In order to solve the problem of interference caused by non-synchronization in the TDD system, this thesis proposes a time synchronization algorithm for TD-LTE home eNodeBs based on UE measurement and assist.Synchronization algorithm based on network listening for home eNodeBs can achieve well synchronization. Synchronization based on air interface listening has been often discussed. Home eNodeBs sniff the signals from neighboring cells through the air interface, and adjust self timing aligned with neighboring cells to achieve synchronization by imitating the procedure of uplink synchronization. However the behavior that home eNodeB analogs UE will affect the normal service. Aiming at this problem, we consider using UE instead of base station to measure the timing difference. An UE can be selected as an assistant UE by its serving home eNodeB. The timing differences among the serving home eNodeB and neighboring home eNodeBs are measured and reported to the serving home eNodeB by the assistant UE. The serving home eNodeB uses time differences to calculate and update its own timing. Serving home eNodeB uses weighted average algorithm to calculate the timing differences and updates its own timing. The whole network of home eNodeBs can achieve synchronization through multiple iterations. Synchronous updating scheme and asynchronous updating scheme are both designed according to the different modes of time updating. In order to solve the problem of slow convergence speed in the asynchronous updating scheme, we introduce the time information interaction technology based on X2 interface. In every iteration, the home eNodeB obtaining timing differences from assistant UE interacts the information with a plurality of neighboring home eNodeBs so that multiple home eNodeBs can update their timing in per iteration. This will speed up the convergence. In reality, the situation where home eNodeBs have no assistant UE may exist. In this thesis, the problem is discussed in two aspects of the synchronous updating scheme and the asynchronous updating scheme. The base station dormancy mechanism has been introduced and an improved algorithm based on the combination of UE measurement and air interface listening has been proposed to solve the problem.The qualitative and quantitative analysis of the proposed algorithm is carried out by simulation experiments. The simulation results show that the time synchronization algorithm based on UE measurement and assist can achieve well synchronization in the network of home eNodeBs. The convergence performance of the synchronous updating scheme is better than that of the asynchronous updating scheme. In depth analysis of the synchronous updating scheme using base station dormancy mechanism, we find that when there is few UEs, separate subsets will appear in the network of home eNodeBs. the convergence of time can be achieved in each subset. The simulation of this particular situation shows that the convergence of time can be achieved in each subset.