On the performance of spectrum handoff framework for next-generation 5G networks
The lack of available radio spectrum and inefficiency in its usage necessitate a new communication paradigm requiring to exploit the existing spectrum opportunistically. One of the perspective spectrum sharing methods, which is currently under a heavy investigation by academia and industry as well across whole Europe, is called Licensed Shared Access (LSA). This novel technology allows for controlled sharing of spectrum between an original owner (primary user, incumbent) and a licensee (secondary user), such as the mobile network operators (MNOs), which coexist geographically. Despite certain benefits, there are still several issues to be solved before the LSA framework will be implemented in commercial infrastructure. One of them is the need to move secondary users (SUs) from the rented LSA band whenever the incumbent needs it. The potential solution for this problem is represented by spectrum handoff, which aims to help SUs to vacate the occupied licensed spectrum and find suitable network resources to resume the unfinished transmissions somewhere else in order to keep SUs satisfaction in terms of quality of experience (QoE) at negotiated quality of service (QoS) level. Inspired by this, we propose a decision making model considering several SUs attributes (RSSI, RSRP, RSRQ, SINR) in order to efficiently implement the handoff procedure and treat SUs to maximize total service time, spectrum utilization and SUs' satisfaction. As an input for our simulation model, we have used the set of measurements performed in real 3GPP LTE-A indoor cellular system located at Brno University of Technology, Czech Republic. Our achieved simulation results evaluate the spectrum utilization of three 3GPP LTE-A cells and provide the total service time for each active SU, while different values of primary user's activity ratio are considered for each cell. Authors would like to recall that this paper represents extended version of their previously published work at TSP 2017 conference.