Chapter 4 analyzes in detail – from a theoretical perspective – the first practical caveat towards such linear growth of capacity in the ultra-dense regime, i.e. that of the impact of the transition of a large number of interfering links from non-line-of-sight to line-of-sight. Importantly, this chapter shows that the theoretical tools used until then to analyze traditional sparse or dense small cell networks, such as that presented in the previous chapter, do not directly apply to ultra-dense ones, and neither do their conclusions. In this chapter, we detail the path loss modelling upgrades necessary for a more realistic and accurate modelling of ultra-dense networks, present the subsequent and new theoretical derivations, and analyze the obtained results for the better understanding of the readers.

We overview the main characteristics of the power-line channel, such as noise, attenuation, and its broadcast nature. We identify the key factors that affect end-to-end performance of single links. We discuss the PHY layer functions and the evolution of PLC technologies. We present a typical PLC transceiver, its signal modulation and coding techniques, and their parameters. We discuss the new features of HomePlug AV2 compared to IEEE 1901 and the differences between Wi-Fi and PLC PHY layers.

We discusse PLC efficiency when multiple users contend for the medium. To resolve contention conflicts, PLC uses carrier sense multi- ple access with collision avoidance (CSMA/CA) on the MAC layer. The stations have to sense the medium before they transmit, and to wait for a random interval of idle-medium time slots before they transmit. The PLC CSMA/CA protocol is similar but more complex than that of Wi-Fi. We present the IEEE 1901 CSMA/CA protocol and certain MAC-layer processes, such as the priority resolution for QoS classes, inter-frame spaces, and frame aggregation. We discuss the new features of HomePlug AV2 compared to IEEE 1901 and the differences between Wi-Fi and PLC MAC layers.

We introduce an experimental framework for PLC.We explain how to configure PLC devices and how to measure certain statistics, such as the capacity of the links, packet errors, modulation information, and collision statistics. We rely on PLC management messages and on open-source tools. We give examples of these messages and guidelines on employing the tools. We also explain how to develop new custom PLC tools.

We explain the fundamentals of PLC network management and security. We describe how privacy and security are guaranteed in PLC. We introduce guidelines for configuring security keys in commercial PLC devices. Finally, we discuss potential issues with PLC security andcompare PLC with Wi-Fi from a physical-layer and security-protocols perspective.

We present an experimental study on the end-to-end capacity of PLC links. We discuss the variability of PLC capacity over space, frequency, and time by using a testbed on a realistic enterprise environment. These measurements are used for characterizing end-user performance, for establishing practical link-metric guidelines, and for facilitating future deployment of hybrid networks that comprise PLC technologies. We present link-metric guidelines that use metrics proposed by IEEE 1905.1 standard.

We evaluate the performance of the PLC CSMA/CA protocol by analysis, simulations and testbed measurements. We explain the impact of each parameter on performance and the motivation behind the design choices for PLC. Both throughput and short-term fairness are discussed and modelled. Short-term fairness is achieved when all users share equally the medium over a short time horizon. If short-term fairness is not guaranteed, there exists high delay-deviation that harms certain delay-sensitive ap- plications. We provide fundamental evidence about performance trade- offs of the PLC CSMA/CA. Our study suggests that PLC parameters should be judiciously selected in order to exploit the tradeoff between throughput and fairness and to achieve high QoS for best-effort and delay-sensitive applications. This chapter also presents a cross-layer PHY/MAC evaluation of contending links and investigates the challenges of stations with different rates and hidden terminals.

In this chapter, we summarize the content of our book and we discuss current limitations of PLC technology. Buidling on these limitations, we highlight new research areas for residential and enterprise PLC networks.

In this, we first explore the evolution of heterogeneous networks and the candidate technologies for augmenting network reliability, including PLC, MoCA, Wi-Fi and Ethernet. We then turn our attention to power-line communications. We discuss high-bandwidth and IoT PLC applications, standardizations, and specifications. Finally, we present the book organization.

The IEEE 1905.1 standard aims at making interoperability between different technologies (Wi-Fi, PLC, MoCA) easier and more flexible. By introducing an abstraction layer between the MAC and IP layers (layer 2.5) and a common interface, it enables a device to seamlessly use or alternate between two or more technologies, depending on the network conditions. Its goal is also to make the management of heterogeneous networks simpler, by providing end-to-end QoS, autoconfiguration, common secured authentication and setup methods.

In this chapter, we present different standardized solutions for heterogeneous net- works. We describe in particular the IEEE 1905.1 standard. We also present algorithms that can be used in heterogeneous networks to improve performance.