Non-orthogonal multiple access techniques for 6G and beyond radio access networks

Abstract

Radio Access Networks (RANs) have experienced an enormous development in a relatively short time. Since their appearance at the beginning of the 1980s, they have become a key tool in human beings' progress. They have completely shifted the way humankind communicates and performs daily activities in a relatively short period of time. This huge success also brings important challenges when designing future access networks. Everyday, there are more devices requiring to be connected to mobile networks, and the services they request have become extremely demanding in terms of data rate, latency and reliability. RANs use multi-user techniques to allow several users to be connected simultaneously. Traditionally, these techniques have been developed utilising orthogonality as an important design factor. It has facilitated the process to keep users' data easily separable and without interfering other users' data. Orthogonal Multiple Access (OMA) techniques have been prosperous in matching the service requirements during the first generations of wireless cellular technology. Some well-know examples are: Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Code Division Multiple Access (CDMA) and Orthogonal Frequency Division Multiple Access (OFDMA). These techniques utilise time, frequency, code, or a combination of some of them, to create orthogonal resources that are allocated to users. It is precisely the available amount of this resources which set the capacity limit on any of these multiple-access techniques. Nevertheless, this never-ending quest for improvement, has obligated multi-user techniques to look beyond the orthogonality threshold they have held for more than 40 years. Non-orthogonal Multiple Access (NOMA) techniques appear as an efficient way to take advantage of the limited electromagnetic spectrum resource, making it possible to handle modern service needs. Therefore, NOMA is able to accommodate more users than OMA utilizing the same radio resources, making a more efficient use of them. This is crucial when dealing with massive amounts of devices, which is precisely the scenario envisioned by the Internet of Things (IoT) and Massive Machine-Type Communications (mMTC). Additionally, it improves the scheduling process, reducing the time that users need to wait to access the service. It also opens the door to achieve higher data rates, which will be required by new applications and rich multi-media content applications. In the present research work, we contribute with the complex task of enhancing NOMA techniques by conducting performance analysis, and through the development of suitable new schemes and algorithms. We also studied the application of NOMA in other important telecommunication technologies. In the first part of our work, we investigate the fairness behaviour of NOMA and delve into NOMA configurable properties and inherent design features that rule it. For this purpose we design a framework to specifically understand fairness in NOMA, both at system and at pair level. This framework includes various reference levels to make it versatile and applicable to diverse scenarios. We propose a power allocation scheme based on fairness, and a pairing scheme prioritizing fairness at individual and pair level. We validated our schemes through simulations and demonstrate the high adaptability that NOMA is able to provide under specific fairness requirements. In the second part of our research, we explore the application of NOMA in Cognitive Radio (CR). By conducting a theoretical study of successive interference cancellation (SIC) hybrid algorithms, we developed closed-form expressions for the probability of success of the primary and secondary users. We analysed the differences between two different SIC hybrid algorithms and highlight the advantages that NOMA is able to deliver to CR. We verify our theoretical analysis utilizing Monte Carlo simulations. In the third part of this research, we study the Physical Layer Security (PLS) performance of NOMA, which is a transcendent point to consider in any emerging technology. Specifically, we analyse the Secrecy Performance of the system under a coordinate eavesdropping scenario, to explain and propose methods to enhance security when designing NOMA. We show how SIC decoding order has an important impact on the system secrecy rate. Two solutions are proposed to the SIC decoding order selection. The first one based on interference, and a second one utilizing deep-learning to enhance the selection process. Both solutions are compared with the optimal one, and they are demonstrated to reach high performance at lower computational complexity level. In the final part of this work, we have a look at satellite communications, where we believe NOMA can be applied to address current challenges. We analyse the effects of intra-beam and inter-beam interference in a NOMA multi-beam satellite system. An optimisation problem to enhance system capacity is proposed considering interference. We also consider secrecy capacity maximisation as an alternative to enhance the system bit rate and improve security at the same time. The solutions to the aforementioned problems are given in terms of NOMA power allocation, which were verified using numerical simulations. In this way, this research work contributes towards the development and enhancement of NOMA, and its application in other modern communication technologies, for Next-Generation Radio Access Network.