Machine Learning in Wireless Networks
With an ever increasing density of mobile broadband users, next generation wireless networks (5G) need to support a higher density of users compared to today’s networks. One approach for meeting this need is to more effectively share network resources through femtocells. However, lack of guidelines for providing fairness to users and significant interference caused by unplanned deployment of femtocells are important issues that have to be resolved to make heterogeneous networks (HetNets) viable. However, the interference caused by femtocells to neighboring cells is a limiting performance factor in dense HetNets. This interference is being managed via distributed resource allocation methods. However, as the density of the network increases so does the complexity of such resource allocation methods. Yet, unplanned deployment of femtocells requires an adaptable and self-organizing algorithm to make HetNets viable. As such, in this project a machine learning approach based on Q-learning will be developed to solve the resource allocation problem in such complex networks. By defining each base station as an agent, a cellular network will be modeled as a multi-agent network. Subsequently, cooperative Q-learning can be applied as an efficient approach to manage the resources of a multi-agent network. Furthermore, we can consider the quality of service (QoS) for each user and fairness in the network.
Massive IoT Networks and Internet of Energy Networks
The rapid advancement in wireless communication and network technologies have enabled the deployment of IoT networks a reality. However, a variety of challenges remain to be addressed in 5G as predicted there will be more than 20 billion IoT devices by 2020. Such challenges are associated with large-scale IoT device access, resource management, limited power, bursty demand, extended coverage range and reduced device cost. Recently, we have developed some promising solutions for devices-to-device (D2D) communications and low powered communications to deal with massive access challenges in IoT networks. And also, we have modelled an IoT application: a smart internet of energy network that consists of IoT elements such as sensors from household appliances, electric vehicles, energy generator sources and renewable energy sources, and the wireless as a transmission medium to reduce the power fluctuation level. Our work will continue to tackle the challenges related to massive access on IoT networks for 5G and make the smart internet of energy networks as a reality of an IoT application.
UAV Enabled Communication Networks
UAV assisted communication networks, where UAVs act as roving access points or relays, have attracted great attention from academia and industry for their flexible and rapid deployment features. Leading internet companies, such as Google, Facebook and major network operators, such as Verizon and AT&T, all have researched and developed prototypes for providing internet access via UAVs under various scenarios, i.e. cellular network coverage extension, wireless access provision for unserved or underserved remote areas, emergency management and disaster recovery, and temporary fast wireless network deployment for events etc. Our work focuses on addressing research challenges in coordinated multi-UAV networks, such as high mobility of UAVs, time-variant network topologies, seamless coverage to the user demand, and limited onboard power for communications etc. In particular, the research utilises Artificial Intelligence (AI) for optimising resource allocation, enhancing self-organising UAV deployment and autonomous flying control to address the unique challenges in UAV assisted communication networks.
Non-orthogonal multiple access for 5G and IoT networks
Non-orthogonal multiple access (NOMA), which has been recently proposed for the 3rd generation partnership projects long-term evolution advanced (3GPP-LTE-A), constitutes a promising technology of addressing the above-mentioned challenges in 5G networks by accommodating several users within the same orthogonal resource block. By doing so, significant bandwidth efficiency enhancement can be attained over conventional orthogonal multiple access (OMA) techniques. This motivated numerous researchers to dedicate substantial research contributions to this field. We advocate on power-domain multiplexing aided NOMA, with a focus on the theoretical NOMA principles, multiple antenna aided NOMA design, on the interplay between NOMA and cooperative transmission, on the resource control of NOMA, on the co-existence of NOMA with other emerging potential 5G techniques and on the comparison with other NOMA variants. We highlight the main advantages of power-domain multiplexing NOMA compared to other existing NOMA techniques. We will continue to work on NOMA with tackling some implementation issues as well as identifying promising research opportunities for the future.
Ultra-dense networks (UDN) constitute one of the most promising techniques of supporting the 5G mobile system. By deploying more small cells in a fixed area, the average distance between users and access points can be significantly reduced, hence a dense spatial frequency reuse can be exploited. However, severe interference is the major obstacle in UDN. Most of the contributions deal with the interference by relying on cooperative game theory. However, this method requires heavy information exchange overhead, which is not feasible for UDN. We advocate the application of C-RAN philosophy to UDN, thanks to the recent development of cloud computing techniques. This network architecture is the so-called ultra-dense C-RAN. Under ultra-dense C-RAN, centralized signal processing can be invoked for supporting CoMP transmission. However, there are several challenges associated with ultra-dense C-RAN, such as the acquisition of the global CSI, high computational complexity, limited fronthaul capacity, high hardware cost, etc. Recently, we have developed some promising solutions to deal with these challenges. We will continue to tackle these challenges to make this network architecture a reality.
Molecular Communications: Unleashing the Internet of Nano Things (IoNT)
Communication and information theory underpins coordination across the fabric of modern civilization. As enabling technologies allow device dimensions to shrink, a new frontier of connected embedded devices has emerged. We now live in an age where nano-devices have the potential to sense and coordinate microscopic operations ranging from targeted in vivo drug delivery to precision material self-healing. The Internet-of-Nano-Things (IoNT) paradigm has the potential to dramatically transform society and is recognised as one of the top 10 emerging technologies by the World Economic Forum. Looking 10-20 years ahead, swarms of nano-devices will need to communicate and coordinate, leading to unparalleled transformations in healthcare and other industrial sectors. The aim of this project is to address practical issues concerning the design and implementation of a new generation of nano-scale networked systems, especially in electromagnetically denied biological environments and in complex industrial settings. More specifically, the objectives of this project are: (1) obtain fundamental understanding to how information can propagate in a variety of in-body diffusion channels by developing multi-scale biophysical models, (2) develop capacity achieving communication protocols with low-complexity down-scaling potential, and (3) design and build functional systems and demonstrate their communication capability in complex macro- and micro-scale diffusion-advection environments.