In this paper, we investigate the impact of mobility on user performance in the context of dense LTE-A networks. To this end, we propose simple analytical models that capture mobility through the distribution of the mobile users sojourn time, i.e., the time a mobile user is physically present in a given cell. We use these models to derive the throughput of users who remain spatially static during their whole data transmission and the amount of handovers generated by moving users. We analyze the impact on performance of some key parameters such as the size of the small cell, the speed and proportion of mobile users and the distribution of their sojourn time. Numerical evaluation and simulation results are provided to assess the accuracy of the latter model and gain insight into the global system performance.

In this paper we compare two simulators: ns-3 and Vienna, in the context of LTE networks, on four basic scenarios for which well-known analytical results exist. These scenarios differentiate themselves by the nature of the traffic (data or voice) and by the number of sources (infinite or finite). Our goal is twofold. First, by confronting the results of the two simulators with exact results, we can assess the accuracy of both simulators and compare their efficiency. Second, and maybe more importantly, we want to compare the ease of handling and use of both simulators, and list the difficulties encountered in the context of the four basic scenarios, that will necessarily arise in more realistic simulated scenarios, and explain how we worked around the problems. We hope this comparison will help researchers who work on LTE networks to choose the simulator that best suits their needs.

In this paper, we investigate the influence of intraand inter-cell mobility of users on performance of 4G/5G cellular networks, such as LTE and LTE-A. To this end, we develop a multi-class PS queue model that captures mobility of users between zones of a cell and between cells, through a simple mobility model, that is decoupled from the cell model itself, enabling to directly apply the approach to more realistic mobility patterns. We first show that this model is consistent with known analytical bounds corresponding to a system with either static users or users having an infinite speed. We then compare our model to simulations for more realistic speeds, and show that it provides user and cell performance with a very good accuracy. The outcomes of our model confirm that mobility may improve both users and cells performance, and enable to quantify the gain as a function of users speed.

No abstract

No abstract