|DEVOTI FRANCESCO||Cycle: XXXII |
Tutor: CESANA MATTEO
Advisor: CAPONE ANTONIO Major Research topic
:Context-aware Resource Management and Network Design in 5G Millimeter Wave Access NetworksAbstract:
The exploitation of mm-wave technologies is one of the key-enabler for 5G mobile radio access networks, potentially providing several-GHz bandwidths. However, their introduction to cellular networks poses several challenges. The harsh propagation conditions limit the mm-wave access availability and make necessary the exploitation of high-gain antenna systems in order to overcome the high path loss and the limited power. Consequently, highly directional transmissions must be used with a significant increase in the system complexity. A further issue with mm-waves is related to their weak penetration ability, which makes every obstacle opaque to the mm-wave propagation, and thus a potential cause of link blockage.
Therefore, to fully unleash mm-wave great potential and meet the stringent 5G requirements, we need novel network design strategies as well as new ways of dealing with legacy network functionalities to provide a fast and reliable mm-wave access.
For guaranteeing a reliable signaling channel, legacy access technologies and a control and user plane (C-/U-plane) split need to be employed. This provides the opportunity to collect context information from the users that can support network operations management. We leverage the context information related to user positions to improve the directional cell discovery process, which is one of the most critical network operations in mm-wave access as it can cause non-negligible latency if not properly managed. We investigate the fundamental trade-offs of this process and the effects of the context information accuracy on the overall system performance. We also cope with obstacle obstructions in the cell area and propose an approach based on a geo-located context database where past system history is stored to guide future searches. Moreover, we investigate the coordination problem of multiple mm-wave base stations that jointly process user access requests. Analytic models and numerical results are provided to validate proposed strategies.
The availability of a rich and accurate context is fundamental to effectively drive network operations and overcome mm-wave propagation weaknesses. Nevertheless, the mm-wave sensibility to the propagation environment can be exploited to enrich the context. To prove this concept, we present a passive human detection system that passively monitors indoor environments leveraging beamforming alignment procedures on already deployed indoor mm-wave communication systems, detecting and locating persons with high accuracy. We implemented our system in commercial off-the-shelf devices and deployed in an office environment for validation purposes.
A widely adopted technique to guarantee mm-wave service reliability is to establish multiple connections from mobile to different base stations. Smart base-station selection must be made to minimize simultaneous blockage probability and maximize multi-connectivity effectiveness. However, the cell selection process is constrained by the network topology. The traditional approach to provide multi-connectivity is based on k-coverage planning. However, it is not guaranteeing reliable connection alternatives. Therefore, we propose a novel mm-wave access network planning framework specifically aimed at producing blockage robust network layout and ensuring the required QoS. Our framework provides the desired k-coverage while reducing the simultaneous blockage probability. The results show that our approach provides much better connection alternatives than traditional k-coverage against obstacle blockages.
To improve the coverage and throughput of the mm-wave access network, the Integrated Access and Backhauling (IAB) paradigm is under standardization. IAB allows to dynamically use the large bandwidth available at mm-wave by tightly scheduling access and backhaul links. Within this framework, we investigate the operational problems of mm-wave multi-hop backhaul networks and propose a MILP model to address the joint optimization of both traffic routing and transmissions scheduling, according to the IAB paradigm. The model captures several technological aspects proper of mm-wave hardware and includes power allocation strategies and rate adaptation. The model is followed by a thorough numerical evaluation where the performance of the multi-hop IAB network is compared against the one of a single base station, showing the potential of IAB architecture.