The performance in mobility scenarios currently prevents mmWave from being used in cellular networks. Indeed, as described here, mmWave links suffer from blockage from obstacles such as the human body, vehicles, buildings, etc. Moreover, mmWave cells will cover a small area, and require a dense deployment. As the user moves, it may suddenly lose the connectivity to the current mmWave base station. It may even be in outage with respect to all of the cells. The mobility management under these hypothesis becomes challenging: the number of handovers and/or beam switch events increases, and the solutions adopted in traditional networks are not enough. Indeed, if the mobile terminal connects to a single Radio Access Network at any given time, an outage event would require a Radio Link Failure or a long and complex inter-RAT handover, for example to a legacy LTE network.
We propose a mobility management solution which relies on dual connectivity. The mobile user connects to an LTE and a mmWave base station at the same time. Moreover, a local mobility anchor coordinates each group of mmWave base station. Therefore, there is no need to interact with the core network for local mobility. The user plane is split at the PDCP layer.
This architecture makes it possible to:
- Collect channel measurements and track the optimal base station for each mobile user;
- Design faster network procedures, to quickly switch from LTE to mmWave and viceversa, and to perform a swift secondary cell handover without any interaction with the core network.
The dual connectivity architecture was analyzed with a first-of-its-kind simulation campaign. It combined a mmWave dynamic channel model and end-to-end TCP/IP and cellular network protocol stacks, using the tool described in the Simulation page. The channel model used experimental traces to model the transition between LOS and NLOS and vice versa. Our architecture is able to better adapt to the dynamic mmWave channel conditions than a traditional single connectivity solution. Therefore, it obtains a lower latency and smaller throughput variations. This increases the robustness of the network, enhancing the quality of experience of the mobile user.
LIST OF RELATED PUBLICATIONS
|M. Polese, M. Mezzavilla, S. Rangan, M. Zorzi, "Mobility Management for TCP in mmWave Networks", in Proceedings of the 1st ACM Workshop on Millimeter-Wave Networks and Sensing Systems 2017 (co-located with Mobicom 2017)||2017/10||Transport, Mobility|
|M. Polese, M. Giordani, M. Mezzavilla, S. Rangan, and M. Zorzi, “Improved Handover Through Dual Connectivity in 5G mmWave Mobile Networks”, to appear on IEEE Journal on Selected Areas in Communications (JSAC)||2017/09||Mobility, Transport, Simulation|
|M. Giordani, M. Mezzavilla, S. Rangan, and M. Zorzi, “An Efficient Uplink Multi-Connectivity Scheme for 5G mmWave Control Plane Applications", submitted to IEEE Transaction on Wireless Communications||2017/07||Tracking, Mobility|
|M. Polese, M. Mezzavilla, and M. Zorzi, “Performance Comparison of Dual Connectivity and Hard Handover for LTE-5G Tight Integration,” in SIMUTools 2016||2016/08||Simulation, Mobility|
|M. Giordani, M. Mezzavilla, S. Rangan, and M. Zorzi, “Multi-Connectivity in 5G mmWave cellular networks,” in IEEE 15th Annual Mediterranean Ad Hoc Networking Workshop||2016/06||Mobility, Tracking|
|Mezzavilla, M., Goyal, S., Panwar, S., Rangan, S., and Zorzi, M., An mdp model for optimal handover decisions in mmWave cellular networks, in Proceedings of European Conference on Networks and Communications (EuCNC), June 2016, Athens, Greece||2016/06||Mobility|