With the increase of access point (AP) density and the exponential growth of mobile devices supported by ultra dense networks (UDNs), overlapped user-centric (UC) clustering is becoming a promising design principle for guaranteeing the quality of service (QoS) required by each UE. However, the overlapped UC clustering has to be jointly designed with resource allocation in UDNs. In this context, both the traffic-load balancing and the limited availability of orthogonal resource blocks (RBs) are carefully considered in UDNs. To tackle these challenges, we formulate a joint overlapped UC clustering and resource allocation problem with the goal of maximizing the system’s spectral efficiency (SE). With the aid of the graph-theoretical framework, the problem is decoupled into two independent subproblems, and a distributed overlapped UC clustering solution as well as a graph-based resource allocation scheme were proposed. Our numerical results quantify the superior performance of the proposed framework in terms of both its per area aggregated user rate (PAAR) and user rate.

Ultra-dense cloud radio access network (UDCRAN) architecture, which integrates the capability of cloud computing and edge computing with the massively deployed radio access points, is a promising solution for the fifth-generation and beyond mobile communications. In order to accommodate the anticipated explosive growth of data traffic, fronthauling technology becomes a challenge technical issue in the fifth-generation and beyond UDCRANs. Moreover, the schemes related to interference management and resource management need to be reconsidered. In this chapter, we will provide a comprehensive review of the current research progress on fronthauling technology. Moreover, we will compare the advantages of various candidate fronthaul schemes.

The wireless edge caching is considered as a promising technique to cope with rapid increase in mobile traffic demand. The fundamental idea of edge caching is to offload the data traffic to local cache memories by dealing with content requests with the pre-fetched contents on network edge nodes. The wireless edge caching consists of two main phases: content placement and content delivery. Since the strategies for these two phases are highly dependent on which devices are capable of caching in the network, the characteristics and types of achievable caching gains appear to vary with the location of cached data. The cached data at the transmitter side can be utilized to reduce the traffic load on backhaul and the latency, while the cached data at the receiver side can be utilized to improve the network resource efficiency and the quality of experience (QoE) of the end-users. This chapter introduces the state-of-the-art wireless edge caching techniques for transmitters and receivers of ultra dense networks and offers a design guideline on reaping the promising gain of wireless edge caching.

Ultra dense networks with directional antennas, like millimetre wave (mmWave) networks, have some promising features about secure communications. This chapter explores the potential of physical layer security in mmWave ultra dense networks. Specifically, we mainly introduced the impact of mmWave channel characteristics, random blockages, and antenna gains on the secrecy performance. Our results reveal that mmWave frequency to high mmWave frequency is demanded to obtain a higher secrecy rate. In addition, new antenna pattern models are needed to well characterize the effective antenna gain for a random interferer seen by the typical receiver when the number of mmWave antennas grows large.

To take full advantage of the ultra-dense architecture and efficiently serve the traffic with spatiotemporal fluctuation, the transmission mechanisms should be redesigned under the constraints of backhaul and energy consumption. In this chapter, we summarize and classify the spatiotemporal arrival properties of different traffic in ultra-dense networks, and optimize several promising technologies to match the traffic. A new approach based on combining stochastic geometry and queueing theory is proposed to provide a useful guidance for the design of ultra-dense networks.

Full duplex ultra-dense network (FDUDN) is envisioned as a promising network paradigm for spectrum efficiency enhancement. This chapter presents a power management scheme, which maximizes the total capacity of FDUDN, under given Quality-of-Service (QoS) and cross-tier interference constraints. The inter-cell power control is formulated as a non-convex optimization problem and the variable substitution is used to transform it into a convex one. Furthermore, the problem is solved through a low-complexity heuristic scheme, which utilizes the water-filling theorem in inter-cell power allocation. Simulation demonstrates the enhancement effect of the proposed scheme in terms of the capacity and the power efficiency.

This chapter investigates the application non-orthogonal multiple access (NOMA) in heterogeneous ultra-dense networks (HUDNs).Particularly, we propose a unified NOMA framework first. Then the applications of the proposed unified NOMA framework in HUDNs will be discussed. With the fact that small cells are densely deployed and the non-orthogonality of resource sharing, the system suffers severe interference. In this chapter, we identify the key challenges in the unified NOMA enabled HUDNs, especially for user association and resource allocation. In addition, we carry out the related case studies for the proposed unified NOMA enabled HUDNs including the user association based on matching theory and resource allocation based on optimization techniques. Furthermore, some critical insights will be provided for the design of NOMA enabled HUDNs, which can promote network access capacity in the next generation of communication systems.

The network densification is one of the prominent solutions for fifth-generation (5G) networks to utilize spectrum resources through intensive deployment of small cells. However, the traffic management in dense networks become a serious challenge for underlying infrastructure supporting the virtual core network. Moreover, 5G will employ different types of communication frameworks: ultra-reliable low latency communication (URLLC), enhanced Mobile Broadband (eMBB), and massive Internet of Things (mIoT). Each identify standard slice type (STT) that have different performance requirements and enabling technologies. The current network developers do not provide any concise identification on how those logic networks would be administrated on top of physical network. This chapter investigates the 5G sliced networks and study virtual networking options to meet the performance requirements of service-based architecture.

Recently, with the development and popularization of unmanned aerial vehicles (UAVs), researches on UAVs have also been attracted increasing attention in wireless communications. By fully exploiting their potentials, leveraging UAVs to assist UDNs can greatly improve the system performance. The probability of having line-of-sight transmissions is higher than the probability in the terrestrial transmissions. Moreover, their attitudes can be freely adjusted. This chapter presents a vision of UAV based UDNs to exploit the potential merits of UAVs in UDNs. The channel characteristics are first discussed. Followed that, the representative scenarios where aerial UAVs are introduced to enhance terrestrial UDNs are investigated in detail, i.e., UAV-enabled BSs, UAV-enabled relays, and UAV-enabled energy transfer. UAV based UDNs also face many challenges, such as the limited spectrum resource and on-board energy. From the spectrum sharing perspective, this chapter also discusses the robust spectrum sharing optimization between UAV communications and terrestrial communications, where the scenarios with abnormal behaviors and uncertain/incomplete information are considered.

Due to the proliferation of smart devices, internet-of-things (IoT) devices, and device-to-device (D2D) communications, the amount of mobile traffic is ever-growing. To satisfy this skyrocketing wireless data traffic demand of mobile users, the densification of the network is unaviodable. However, the denser network not only improves the transmission rate, but also increases the impact of interferences. Therefore, careful deployment of base stations (BSs) is needed to guarantee the communication quality. This chapter provides insights into the deployment of BSs in the multi-layer ultra dense network (UDN). Throughout this chapter, we will model the channel between the BSs and a typical user equipment (UE) by considering the antenna height of the BSs and derive the expressions for the interference, the coverage probability, and the area spectral efficiency (ASE) of our considered multi-layer UDN using stochastic geometry. Through numerical results, we will show how the network performance can be maximized by selecting the proper antenna heights and the densities of BSs in the multi-layer UDN.

As mobile data traffic keeps growing and mobile applications pose increasingly stringent and diverse requirements, wireless networks are facing unprecedented pressures. Network infrastructure densification presents promises to further evolve wireless networks and maintain their competitiveness. Deploying more radio access points equipped with storage and computation capabilities can increase network capacity, improve network energy efficiency, provide low-latency services and access for massive devices. The benefits of network densification can be exploited using the emerging fog radio access network (Fog-RAN) architecture by pushing computation and storage resources to network edges. However, it comes with formidable technical challenges. Innovative methodologies are needed to operate such networks with various resources. This chapter develops a generalized low-rank optimization model for performance enhancements in ultra-dense Fog-RANs, supported by various motivating design objectives including mobile edge caching and topological interference alignment. A special attention is paid on algorithmic approaches for nonconvex low-rank optimization problems via Riemannian optimization.