The enormous proliferation of wireless data traffic has caused many challenges to the current telecommunications system. For both users and operators, one of the main obstacles is the balance between the deployment expenditure and scarcity of available spectrum. Per recent works, a possible solution for next generation mobile telecommunications is the millimetre-Wave (mmWave). In this thesis, we proposed a solution: 5G on Unlicensed band (5G-U), which aggregates the 28GHz licensed or lightly licensed mmWave bands, and the 60GHz unlicensed mmWave bands, namely, Wireless Gigabit (WiGig). Followed by giving the background information of LTE on Unlicensed band (LTEU) and other spectrum sharing technologies, the 5G-U system and other related concepts were analysed, the frontier of research and potential use cases were discussed in this thesis. The framework of 5G-U was thoroughly studied from four points of view: the integration of licensed and unlicensed bands of 5G-U; the coexistence in the 5G-U and WiGig multi-Radio Access Network (multi-RAN); the optimisation of the association strategy in the proposed system; finally, the enhancement of the resource allocation strategy for 5G-U dual-band transmitters. A few tools and methods were used in this thesis. In short we introduced 5G-U ON/OFF Duty Cycle to aggregate licensed and unlicensed bands in the simple network; Stochastic Geometry, Laplace transform, Poisson distribution, etc., were used in the coexistence of a complex 5G-U multi-RAN; Game Theory was used for optimising the association strategy of 5G-U and WiGig users in the multi-RAN; and finally, we used Queueing Theory and Markov Chain to study the resource allocation strategy for licensed and unlicensed resource pools for the dual-band transmitters in the multi-RAN. Consequently, we found that the capacity of the 5G-U receiver is significantly higher than that of the WiGig and baseline 5G receivers in a simple multi-RAN; moreover, we proved that the 5G-U network is a good neighbour to the WiGig network in a complex multi-RAN, and the capacity of the 5G-U receivers is higher than that of the WiGig receivers under the same development density; furthermore, we developed an optimised association strategy that the coverage probability of the 5G-U receivers is largely improved compared to that of the WiGig receivers when the threshold Signal to Interference and Noise Ratio is 10 dB and 20 dB; last, we introduced an efficient resource allocation strategy to increase the system resource utilisation coefficient by 150% under all cases.
|Date of Award||1 Nov 2019|
|Supervisor||Mischa Dohler (Supervisor) & Toktam Mahmoodi (Supervisor)|