Video data have been showed to dominate a significant portion of mobile data traffic and have a strong influence on a backhaul congestion issue in cellular networks. To tackle the problem, proactive caching is considered as a prominent candidate in terms of cost efficiency. In this chapter, we study a novel popularity-predicting-based caching procedure that takes raw video data as input to determine an optimal cache placement policy, which deals with both published and unpublished videos. For dealing with unpublished videos whose statistical information is unknown, features from the video content are extracted and condensed into a high-dimensional vector. This type of vector is then mapped to a lower-dimensional space. This process not only alleviates the computational burden but also creates a new vector that is more meaningful and comprehensive. At this stage, different types of prediction models can be trained to anticipate the popularity, for which information from published videos is used as training data.

In this chapter, we discuss the application of edge caching to enhance the physical layer security of cellular networks with limited backhaul capacity. By proactively sharing the same content across a subset of base stations (BSs) through both caching and backhaul loading, secure cooperative multiple-input multiple-output (MIMO) transmission of several BSs can be dynamically enabled in accordance with the cache status, the channel conditions, and the backhaul capacity. We formulate a two-stage nonconvex optimization problem for minimizing the total transmit power while providing quality of service (QoS) and guaranteeing communication secrecy during content delivery, where the caching and the cooperative MIMO transmission policy are optimized in an offline caching stage and an online delivery stage, respectively. Caching is shown to be beneficial as it reduces the data sharing overhead imposed on the capacity-constrained backhaul links, introduces additional secure degrees of freedom, and enables a power-efficient communication system design.

Driven by the inherent spatiotemporal correlation in wireless data demand, cellular network design is becoming increasingly content-centric. An integral component of this new paradigm is the network's ability to cache popular content at its edge, which includes base stations, access points, and handheld devices. This additionally reduces latency, which is one of the key challenges facing the next generation of cellular networks. As discussed in the earlier chapters, the huge size of a typical library of popular files and the relatively smaller storage capacities of edge devices, especially small cell base stations (SCBSs) and handheld devices, make it necessary to carefully determine the set of files (cache) that should be placed on each device. Compared to a wireless network for which caching mechanisms are fairly well understood, a distinctive feature of content-centric wireless networks is the mobility of the end users, which needs to be included in the system design. Inspired by this, we investigate the impact of mobility on edge caching in this chapter.

In this chapter, we present the concept of stochastic caching in large wireless networks with randomly distributed nodes. Specifically, we consider a random network where user devices can directly communicate and exchange information through device-to-device (D2D) communication. The distribution of D2D-enabled devices follows a Poisson point process (PPP), and each user stores proactively the popular files based on some probabilistic caching policy. The optimal caching probabilities depend on the specific objective functions to be optimized. We investigate three different caching schemes – namely maximizing the cache-hit probability, maximizing the density of successfully served requests by local caches, and minimizing the delay to receive the requested content. By comparing the performance achieved with these schemes, we show that the success probability of physical layer (PHY) transmission plays a critical role in the throughput and delay performance of large wireless networks with stochastic caching methods.

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.