Abstract:
The fifth-generation (5G) cellular network is expected to meet the demand for
ultra-high data rate, low latency and bulk connectivity due to the massive deployment of internet of things (IoT) sensors and devices. The existing orthogonal multiple access (OMA) techniques fall short to achieve these demands. Thus, 5G comes
with some unique exciting features to entertain this gigantic conn ectivity and low
latency network. Non-orthogonal multiple access (NOMA) has b een acknowledged
as a promising feature to significantly enhance the network capacity. Moreover, 5G
offers an extension to the current bandwidth (sub 6 GHz) by enabling millimetre-wave
(mmWave) communication. This unutilized mmWave high-frequency band s (30-300
GHz) can carry huge data with an increased bandwidth which paves the way for
massive multiple-input multiple-output (Massive MIMO) which facilitates high spectral efficiency (SE) and lowers path loss of mmWave transmission. But this gives
rise to the p ower-hungry rad io frequency chain which, in turn, increases the energy
consumption with an increased receiver complexity. An energy efficient b eamforming
technique is, therefore, vital for an efficient network with high SE. Interference mitigation is another prime concern for better performance in 5G network. The existing
interference mitigation techniques for OMA are not fully compatible for NOMA network. Thus, a proper interference mitigation technique is essential to increase the
sum rate of the network.
In this thesis, a mmWave-MIMO-NOMA system model is presented which provides better the sum rate than some existing counterp art. To deal with the energy
consumption, an energy-efficient massive MIMO network with a fuzzy logic switching-based hybrid beamforming (FeE-HBS) is presented. To mitigate the intra-cell
interference created by the non-orthogonal users of NOMA, successive interference
cancellation (SIC) is applied in the NOMA receivers. A modified fractional frequency
reuse (FFR) technique is proposed to avoid inter-cell interference (ICI) which provides lesser bandwidth splitting than the conventional techniques. This lower splitting
enables huge no des to accumulate p er sectors of a cell with reduced interference. Thus
it improves the network performance while ensuring minimu m ICI.
This work develops a mmWave-MIMO-N OMA network with better system capacity. The FeE-HBS system enables lower number of power-hungry phase shifters
when the requirement is less. This makes the system more energy efficient with almost
the same SE as the conventional digital beamforming design. Lastly, interference
mitigation is achieved by implementing SIC and a novel sectored FFR. This novel
FFR technique ensures lower interference at a better system throughput than other
existing techniques.
