Integration of LiFi, BPL, and Fiber Optic Technologies in Smart Grid Backbone Networks: A Proposal for Exploiting the LiFi LED Street Lighting Networks of Power Utilities and Smart Cities

Abstract

This paper presents a proposal for extending an existing terabit-class backbone network architecture to enable the use of LiFi technology by power utilities and smart cities. The proposed architecture provides a practical means of integrating three smart grid communication technologies—fiber optics, BPL networks and LiFi LED street lighting networks—across the transmission and distribution power grids of smart cities. In addition to expanding the backbone communications network architecture, the paper provides a comprehensive overview of LiFi technology and analyzes the concept of LiFi LED street lighting networks in a smart city. The analytical investigation of the operation and performance of LiFi LED street lighting networks focuses on the following aspects: (i) typical LED street lighting configurations and default configuration parameter values encountered in smart sustainable cities; (ii) the applied LiFi channel model and corresponding default model parameters; (iii) SNR computations and LiFi channel classifications for a variety of scenarios; and (iv) available LiFi LED street lighting network architectures for integrating LiFi LED street lighting networks with the backbone network. The paper also discusses the potential benefits of LiFi LED street lighting networks for power utilities, smart cities and individuals.

1. Introduction

The smart grid is the modernized version of the vintage electric power grid that exploits fiber optics and other communications technologies to enable the real-time monitoring and control of power generation, transmission, distribution and consumption [1,2,3,4,5,6,7,8]. Through the adoption of the smart grid, the reliability, efficiency, security and sustainability of the electric power system can significantly improve by integrating renewable energy sources, demand response programs, distributed generation and energy storage systems [9,10,11]. Also, smart grid can enhance customer participation and satisfaction by offering more information and choices regarding customers’ energy usage and cost issues [12,13].

Among the communications technologies of the smart grid, fiber optic technology can support the operation of a terabit-class backbone communications network across the existing transmission and distribution power grids [1]. Fiber optics present many advantages for smart grid applications, such as high bandwidth, low attenuation, high data rates, immunity to EMI and long transmission distances [4,5,6]. In [1], a thorough investigation of the fiber optic network deployment across overhead power grids has been given where the available architectures of the fiber optic backbone communication network across the overhead transmission and distribution power grids that could be adopted for use have been presented. In addition, the network scalability and network expansion capabilities of the fiber optic network across transmission and distribution power grids with other available broadband communications technologies of the smart grid, such as WSNs, DSL, fiber optics, WPAN, WiFi, WiMAX, 4G, 5G, GSM and BPL, have been outlined. The aforementioned communications system architecture can be considered to be an open platform that can integrate other communications technologies, such as LiFi technology.

LiFi is a wireless VLC technology that exploits visible light, ultraviolet and infrared spectra for high-speed data communications. LiFi technology is a paradigm-shifting 5G technology that extends the VLC concept so that high speed, secure, bi-directional and fully networked wireless communications may be achieved while it has already been integrated with other broadband communications technologies existing in the smart grid, such as WiFi and BPL [7,14,15,16,17,18,19,20,21]. In fact, LiFi can be treated either as a complementary technology to WiFi or as a force multiplier since LiFi internet speeds could be significantly higher than the conventional WiFi ones [22]. At the same time, the evolving IoT creates a large and diverse customer base for wireless applications that are characterized by heterogeneous performance requirements, such as high data rates, reliable time-sensitive networking, support for high mobility and high user density thus rendering outdoor VLC technologies promising for IoT applications under the aegis of power utilities and smart cities [23]. More specifically, the utilization of LED street light lamps around a smart city acting as LiFi access points could offer free broadband communications services for customers and residents/visitors as part of the benefit programs that could be supported by power utilities and smart cities, respectively [24]. In this paper, the essentials of LiFi technology are given as well as typical LiFi LED street lighting configurations of a smart city and the appropriate channel modeling techniques. As the performance of the fiber optic backbone communications network and LiFi networks is evaluated across the smart grid, SNR of the typical LiFi LED street lighting configurations are computed and the proposed LiFi LED street lighting network architecture schemes are shown. In accordance with [1], BPL networks, which are installed across the transmission and distribution power grids, are expected to act as the intermediary technology between the fiber optic backbone communications network and LiFi systems. Then, the combined field installation and operation of the fiber optic backbone communications network and the LiFi LED street lighting networks are discussed on the basis of a real overhead MV power grid. Issues regarding: (i) benefits for the power utilities and smart cities; (ii) the power utilities’ and smart cities’ benefit programs to their customers and residents/tourists, respectively; (iii) smart tourism; and (iv) sustainable and green power utilities and smart cities; are also outlined.

The rest of this paper is organized as follows: In Section 2, a brief overview of LiFi technology (i.e., LiFi system architecture, advantages, disadvantages and applications) is given. Section 3 analyzes the concept of LiFi LED street lighting networks regarding: (i) the typical LiFi LED street lighting configurations encountered in smart cities; (ii) the applied communications channel model and corresponding SNR computations of the examined scenarios; and (iii) available LiFi LED street lighting network architectures based on the BPL networks for the further technology integration. Section 4 highlights the combined operation of the fiber optic backbone communications network across the overhead power grids of [1] and the LiFi LED street lighting networks. Also, incentive ideas for power utilities and smart cities to motivate and promote the use of LiFi LED street lighting concept are presented.

2. LiFi Technology Essentials

Historically, the earliest known VLC technology application comes from A. G. Bell, who demonstrated in 1880 the photophone that transmitted voice data over 200 m using sunlight beams [25,26]. Recently, H. Haas has proposed a modern concept of the LiFi system while IEEE 802.15.7 has standardized its PHY and MAC layers [15,27,28,29]. Being able to deliver high data rates for various broadband services via IEEE 802.15.7, LiFi systems can cope with optical transmission mobility, compatibility with artificial lighting and the interference which may be generated by ambient lighting.

In Figure 1, the block diagram of a typical LiFi system is presented. A typical LiFi system consists of three operation areas, namely [30,31,32]:

(i)

LiFi Transmitter Side: At the LiFi transmitter side, the data input is first converted to binary information through an ADC and then fed into a LED driver circuit that modulates the signal. Remaining unnoticeable by the human eye, the high speed light intensity changes of LED lamps, which occur between 400 THz and 790 THz, allow the information to be transmitted as optical pulses through the wireless channel; i.e., the LiFi communications channel;

(ii)

LiFi Communications Channel: Corresponding to a non-licensed spectrum band, LiFi communications channels may offer THz bandwidths and up to 100 Gbps capacities against the respective GHz bandwidths and multi-hundreds Mbps capacities of WiFi communications channels. The optical carriers that are used during the LiFi information transfer allow the LED lamps to illuminate at the same time. In accordance with Table 1 of [30] and verified in Section 3.2, maximum transmission distances obtained have been at 10 m for high data rate transmission (i.e., LiFi communications channel suffers from high attenuation) while the clear LOS propagation type is important to high-speed communications. Finally, the LiFi communications channel may be affected by the non-linearity of the voltage and the interference from other light sources and obstacles; and

(iii)

LiFi Receiver Side: At the LiFi receiver side, the optical carriers of the received signal are first interpreted by a photodetector or an imaging sensor into an electrical signal and then the double stage inverting amplifier, which consists of a transimpedance amplifier and a comparator, delivers the binary information to the original message as the output signal of the LiFi system. The LiFi receiver consists of a concentrator and an optical filter for effectively collecting the LiFi signal.

It is obvious that LiFi systems may simultaneously deliver illumination and broadband communications by using the same existing lighting infrastructure [15]. Thus, LiFi can be examined as a 5G/6G technology that can support IoT applications, drive Industry 4.0 and further enhance the LAAS exploitation [14]. With reference to Figure 1, LiFi technology advantages, disadvantages and applications are further detailed in the following subsections.

2.1. LiFi Technology Advantages

The integration of LiFi technology in IoT devices and communications networks can bring new opportunities for future applications. In this subsection, the features provided by LiFi technology are detailed thus explaining its important alternative role to today’s communication technologies. Some of the noted advantages of LiFi systems over RF, microwave and IR ones are:

(i)

THz Frequency Range Spectrum [25,30]: As the demand for wireless data transmission is constantly growing in the IoT era, the RF and microwave spectra are becoming increasingly congested while their spectral management becomes a challenging issue. It should be remembered that the usage of RF and microwave spectra is regulated so as to prevent EMI and ensure efficient spectral usage. On the other hand, only the visible light spectrum, which corresponds to the frequency band of 400 THz to 790 THz, has 10,000 times greater frequency range than both the RF and microwave ones, that correspond to the frequency band of 3 kHz to 300 GHz. Therefore, the higher LiFi bandwidth due its wider and unregulated spectrum may satisfy the growing IoT demands. However, while the visible light spectrum is unregulated in the sense that specific frequencies are not licensed, there are regulatory considerations, safety standards and guidelines related to the overall usage of light for communication purposes so that Li-Fi technology does not interfere with other light-based systems or be harmful to human health;

(ii)

Capacity Boost [30,33]: Unlike in RF, microwave and IR communications systems, no restrictions on transmitted optical power may be imposed on the use of LiFi technology. Combined with the ample LiFi bandwidth, very high capacities are expected that can satisfy the growing IoT data demand. In addition, large transmitted optical power entails higher maximum transmission distances;

(iii)

Not Harmful for the Human Body [25,34,35,36]: In illumination conditions, there are no restrictions concerning the transmitted optical power of the visible light. Unlike IR, the high transmitted optical powers of visible light satisfy eye and skin safety regulations and are not harmful for the human life;

(iv)

EMI Free [25,30]: LiFi technology is intrinsically safe without causing EMI to other RFs, microwave and IR based systems. Therefore, LiFi systems become suitable for EMI sensible devices in aviation, healthcare, industry and military applications;

(v)

Security [33,37]: Due to its relatively shorter maximum transmission distances, LiFi technology can transmit data more securely than other communications technologies whose signals can be easily detected. In fact, LED signal rays are confined by opaque boundaries and do not pass through walls as RF waves do. The well-defined coverage zones of LiFi signal prevent snooping thus enhancing LiFi security in comparison with the RF technology one;

(vi)

Spatial Reuse: The confinement of LED signal rays by opaque boundaries does not permit LiFi signals of adjacent areas to interfere with each other. In contrast with WiFi technology where EMI among adjacent access points of the same carrier frequencies can cause a degraded performance, different users of LiFi technology can coexist in proximity exploiting the same optical carrier frequencies;

(vii)

Easy Implementation [25,30,38]: LiFi technology can exploit the already existing lighting infrastructure with the addition of a few relatively simple and cheap front-end components. Anyway, the LiFi transmitters of Figure 1 can be considered to be an updated version of existing LED lamps intended for only illumination purposes. Since the implementation of LiFi systems can be carried out in the existing electrical infrastructures and there can be no need for additional wiring excess to the existing wiring, the LiFi installation is simple with dynamic structure and of low cost;

(viii)

Energy Efficiency [25,35,39,40,41]: LED lamps used in LiFi technology consume at least 75% less energy and last many times longer than the vintage lighting (i.e., about 60 times more than the filament lamps and about 6 times more than the fluorescent lamps). Adding minimum extra power consumption, LiFi technology can exploit the transmitted optical power of LED lamps, that is mainly used for illumination. Therefore, LiFi technology can be extremely energy efficient and be part of green communications and green technology initiatives, in general.

(ix)

Integration of LiFi Technology and BPL Networks [18,19,20,39,42,43,44,45,46]: On the first hand, BPL networks can enable the use of power lines across the transmission and distribution power grids for communications. In fact, a system architecture that integrates the operation of the fiber optic backbone communications networks across the power grids and the BPL networks has been analytically presented in [1]. On the other hand, the existence of BPL networks alleviates the need for installing new communication cables with the already existing lighting infrastructure so that LiFi systems can operate. BPL technology specifications have been consolidated into IEEE 1901 and ITU-T G.9960/61 while IEEE 802.15.7 has standardized PHY and MAC layers of LiFi technology. Since the standardization for both BPL and LiFi technologies had been completed, the integration of LiFi and BPL networks has intensively been investigated. Therefore, the integration of BPL networks as a backbone network for LiFi networks aims at efficiently and economically delivering broadband access to LiFi access points thus significantly reducing the implementation cost of LiFi technology; and

(x)

Low Installation, Operation and Maintenance Costs [25,39,47,48,49]: As has already been mentioned, the implementation cost of LiFi technology remains relatively low since only few upgrades of the existing LED lighting infrastructure are required rather the installation cost of an entire communications network from scratch. Numerically speaking, some popular RF systems (e.g., Bluetooth module) operating over approximately 10 m provide data rates of up to 1 Mbps and cost around $5 whereas LiFi systems can operate at the same distance providing 50 Mbps data rates and costing $1.7 per module. Note that LiFi systems are in a symbiotic relationship with LED lamps thus the cost per LiFi module can be further reduced due to the LiFi LED lamp scale economy.

Knowing the strengths and the weaknesses of LiFi technology also helps us to create more targeted and effective applications. Prior to presenting possible LiFi applications, the disadvantages of the LiFi technology are reported in the following subsection.

2.2. LiFi Technology Disadvantages

Apart from the great number of advantages, LiFi technology suffers from certain drawbacks that are:

(i)

LOS Operation [18,19,20,35]: It is obvious that the broadband access via LiFi technology can be used where light of LED is available. Moreover, the optical power that reflects from the surrounding opaque surfaces such as walls is less than the optical power that directly reaches the LiFi receiver; the LOS transmission path. But the LOS operation of LiFi technology cannot always be mentioned as a disadvantage because this operation can be very useful for secure communications and geolocation applications, as described in Section 2.3;

(ii)

Shorter Maximum Transmission Distances [30,31,32]: LiFi technology has a short maximum transmission distance because the light is used as the transmission medium of radio waves. As has already been mentioned, the maximum transmission distances obtained have been 10 m for high data rate transmission from the literature; and

(iii)

Interference from Light Sources and Sunlight: Since light is used as the transmission medium for LiFi technology, other optical sources that are present nearby and sunlight may affect LiFi performance. In general, LiFi technology is sensitive to environmental factors such as obstructions and ambient lighting conditions.

By considering both the advantages and disadvantages of the LiFi technology, a better design of the LiFi technology applications can be made in the following subsection.

2.3. LiFi Technology Applications

LiFi can be treated either as a complementary technology to WiFi or as a force multiplier since LiFi access point internet speeds are significantly higher than the ones of the conventional WiFi. There is a plethora of applications where LiFi can provide better performance than the traditional WiFi, namely:

(i)

Aviation [50]: On average, there are 200 to 300 LEDs in a commercial airplane that can be used as a network switch by the passengers for enjoying wireless internet connectivity during their flights. Apart from the many times faster LiFi internet connectivity of the passengers, LiFi does not create EMI to aircraft equipment, e.g., the radar, navigational system, other avionics;

(ii)

Healthcare [22,51]: In many areas of hospitals, e.g., operating rooms, WiFi is not allowed as a precautionary measure against the anticipated EMI either to medical equipment or to the patient’s health. The real time monitoring of patient’s health (body temperature, heart rate, blood pressure and breathing rate) and the continuously evolving robotics based medical treatment methods require a continuous and uninterrupted internet connection that can be safely ensured by LiFi;

(iii)

Educational Institutions [52]: To take advantage of digitalization, educational institutions require reliable, high-performance broadband connectivity campus wide so that all students can have access to broadband services. LiFi may provide broadband connectivity where the lighting infrastructure exists;

(iv)

Industry [52]: Deploying WiFi in plants, e.g., power plants, chemical plants, is not recommended due to its interfering and inflammable nature. Li-Fi technology is a safe and economical option that can be used instead;

(v)

Military [53]: Electronic warfare-based jammers can block RF based military devices thus crippling critical pieces of telecommunications equipment. LiFi can ensure the continuous localized operation of military devices and the military secrecy of critical communications;

(vi)

Underwater Applications [50,54,55]: Typical underwater communications become feasible by deploying wired fiber optic cables, wired shielded ethernet cables and wireless VLF communications but with proneness to faults. As a limited role capability, LiFi can be applied so as to allow point-to-point underwater communications;

(vii)

Natural Disasters [53]: Climate change has been linked to an increase in natural disasters such as fires, earthquakes and tsunamis. Natural disasters can harm existing telecommunications infrastructure, disrupting communications. Mounted LiFi-enabled LEDs can ensure the continuity of communications in such types of emergencies;

(viii)

Charging Smart Devices and Electric Vehicles [50,56]: Apart from the communications aspect of LiFi, the integration of a thin crystal layer that may act as a solar photovoltaic cell can allow the charging of smart devices and electric vehicles. Therefore, LiFi can provide the communications channel, the illumination and the wireless power transfer for smart devices and electric vehicles, e.g., ad hoc vehicular networks, short-range LiFi systems between smart devices, LiFi drones;

(ix)

Smart Home [25,57,58]: LiFi technology may keep smartphones, smart televisions and other home devices online without competing for the indoor WiFi bandwidth. Indeed, the home ecosystem and indoor IoT activities may drive the adoption of indoor LiFi where indoor LiFi can use invisible infrared light, so it works even in the dark;

(x)

Smarter Grid [18,19,20,25,42]: Apart from illumination, power utilities can become telecommunications providers by exploiting energy-efficient LiFi LED illumination systems for broadband access. The ubiquity of LED lighting in urban environments and roads with the power lines could be a strong complementary wireless technology to WiFi/Ethernet coexistence;

(xi)

Sustainable Green Smart Cities [50,53,59]: One of the key reasons for the popularity of LED technology is its energy efficiency; LED lamps consume less than 25% of the energy consumed by vintage fluorescent tubes while they offer higher lifespans. Apart from the lower consumption and improved security of LiFi networks, minimal effects on the environment can be observed when LiFi networks are deployed; as LiFi exploits the light spectrum and common household/office equipment, it is suitable for future smart cities and green cities. Due to its environmentally friendly behavior, LiFi technology can contribute to decreasing the carbon footprint of smart cities as it is about 24 times better in the performance-to-energy-consumption ratio with respect to the one of public WiFi access technology that is widely used by smart cities today. At the same time, VLC is estimated to be ten times cheaper than Wi-Fi thus creating fiscal space in smart cities for the upcoming smart city’s investments; and

(xii)

Street Lights, Traffic Lights and Car Headlights [25,30,52]: With the aim of improving the QoL of its population and tourists, a smart city can be fully connected through its lights (e.g., road signs, public transportation systems, ambulance services, public safety, traffic management, geolocation). In [60], the integration of LiFi technology in streetlights, which is of interest in this paper, is proposed in order to transmit the internet by their light across a smart city. Taking the concept of broadband access through smart city’s lights even further, a smart city’s vehicles can become smart too; if vehicles are equipped with LED headlights and LED backlights, vehicles will communicate with each other while traffic signal changes will depend on the number of vehicles waiting.

From the previous plethora of LiFi technology applications, applications (x)–(xii) attract further interest in this paper. It is evident that outdoor IoT activities, smart cities, smart grid and LiFi technology are some of the emerging trends in technology that have the potential to transform our lives. In particular, the study of this paper is going to focus on the concept of LiFi LED street lighting networks across a smart city under the aegis of the power utilities and smart cities. Prior to the investigation of LiFi LED street lighting networks, several useful issues concerning the advantages, disadvantages and applications of the LiFi technology presented in this section are further discussed with reference to the LiFi LED street lighting networks:

• All the advantages of the LiFi technology reported in Section 2.1 occur in LiFi LED street lighting networks of this paper. The synthesis of the advantage (ix) with the application (x) deserves special mention since the integration of LiFi technology and BPL networks is further promoted in this paper by expanding the existing communications system architecture of [1] where its network system architecture now integrates the operation of fiber optics and BPL networks across the overhead transmission and distribution power grids. Note that fiber optics and BPL networks are among the most discussed communications technologies of the smart grid.
• As the LiFi disadvantages (i) and (ii) of Section 2.2 concern LiFi LED street lighting networks, a LiFi receiver easily obtains the LOS signal as the LiFi transmitters are on the suspended LED lamps of street lighting configurations. Due to their placement across the urban environment of a smart city, the surrounded opaque surfaces of LED street lighting configurations are expected to have a low impact on the performance of the LiFi LOS communications channel while the maximum transmission distance is physically defined by the surrounded urban environment of each LED street lighting configuration (e.g., park, road, square) as well as the user position with respect to the position of the LED street lighting configuration.
• As far the LiFi disadvantage (iii) of Section 2.2 is concerned, LED street lighting configurations, which are intended to improve visibility at night, provide safety and security for pedestrians and motorists, facilitate traffic and enhance the aesthetic appeal of the city at night, are located on the edge of roads or paths in open space without any shading installations. This means that sunlight can significantly deteriorate LiFi performance during daily hours. For the previous reason, LiFi LED street lighting networks are treated as a complementary technology and as a force multiplier to the WiFi networks of the smart cities, while their use is expected to be maximized except for daytime hours.
• With reference to application (viii), achieving long battery longevity for the devices of the IoT era is going to require an accumulation of divergent energy harvesting strategies [56,61,62,63]. Among the major wireless charging techniques (i.e., sunlight, inductive coupling, magnetic resonance, coupling, microwave radiation, distributed laser charging), LiFi LED street lighting configurations may be capable of safely providing power to mobile devices [50,56,61,64]. Therefore, LiFi LED street lighting networks under the aegis of power utilities and smart cities can also provide free broadband connectivity and battery charging.

From this section, it is obvious that LiFi LED street lighting systems may simultaneously deliver illumination, broadband communications and power charging by using the same existing lighting infrastructure. In the following section, a more detailed view of LiFi LED street lighting networks is given by focusing on the configurations, suitable LiFi channel model, SNR computations and network architectures for their further integration.

3. LiFi LED Street Lighting Networks

LiFi LED street lighting networks depend on street lighting configurations and network architectures. In this section, issues such as the typical LiFi LED street lighting configurations encountered in smart cities, LiFi channel model, SNR computations of LiFi LED street lighting systems and LiFi LED street lighting network architectures that are based on the BPL networks are investigated.

3.1. Typical LiFi LED Street Lighting Configurations

In Figure 2a–c, three typical street lighting configurations that may be encountered in cities are presented while their lamps that are going to be replaced with the LiFi LED lamps are shown inside the red squares. LED lamps of the street lighting configurations are mounted almost horizontally in various lighting designs (either single- or double-sided lighting design in this paper) but with fixed vertical aiming. The applied LED street lighting configuration and the height of LED lamps from the road differ and depend on: (i) the road type (e.g., local, urban, suburban, rural, ring, radial); (ii) the traffic intensity; (iii) the location inside the city; and (iv) the application for pedestrian purpose. As far as the wattage of LED street light lamps is concerned, this can be determined by the height of LED lamps and the aforementioned road factors. In general, LED street light lamp heights from 6 m to 8 m require from 30 W to 100 W, respectively [65]. As far as the luminous efficiency (lumens) of the LED street light lamps is regarded, it should be noted that the luminous efficiency depends on many different factors, such as LED efficiency, power efficiency, lens efficiency and heat dissipation conditions and may range from 60 LM/W to 120 LM/W [65]. As far as the spacing between two LED street lighting configurations is regarded, this should be roughly from 2.5 to 3 times the height of LED lamps; i.e., in projects where a 5 m-LED street lighting configuration is deployed, 12.5 m to 15 m spacing can be left between the LED street lighting configurations [66,67].

In Figure 3, a generic schematic representation is shown for the study of the typical urban street lighting configurations of Figure 2a–c. The LiFi LED street lighting configuration lies beside the road while its LiFi LED lamp, which acts as the LiFi transmitter of Figure 1, is assumed to stand at height H above the road with a fixed vertical aiming as shown in black dashed line. The photodetector or the imaging sensor, which acts as the LiFi receiver of Figure 1, stands at height h above the road thus establishing the LOS transmission path of the LiFi signal, which is highlighted as the blue solid line in Figure 3, from the LiFi transmitter to the LiFi receiver. The LOS transmission path length and the horizontal distance between the LiFi transmitter and the LiFi receiver are equal to r and d, respectively, while θ and ψ denote the angles of irradiance and incidence, respectively. Since the LOS transmission path length r, the irradiance angle θ and the incidence angle ψ are the LiFi LED street lighting configuration parameters of Figure 3 that are of interest for the LiFi channel model of the following subsection, the LiFi LED street lighting configuration support type -either the metallic pole of Figure 2a or the wooden pole of Figure 2b or the horizontal wire rope of Figure 2c-, the LiFi transmitter relative position -either above the road as in Figure 2((a, high) and b,c) or beside the road as in Figure 2(a, low)- and the LiFi receiver relative position -either on the road as in Figure 2b,c or beside the road as in Figure 2a- can be treated as different geometrical case studies of the generic schematic representation of Figure 3.

With reference to Figure 2a–c, the LiFi LED street lighting configuration parameters of Figure 3 can be numerically approached, namely:

(i)

LiFi Transmitter Height H: This height is equal to 4.77 m (low)/6.82 m (high), 4.54 m and 6.41 m for the street lighting configurations of Figure 2a–c, respectively. Typically, the height of LiFi LED lamps above the ground ranges from 2.74 m to 8 m [65,68]. By considering the previous reports, the default LiFi transmitter height is assumed to be equal to 5 m for the computations of the next subsection;

(ii)

LiFi Receiver Height h: Theoretically, this height can range from 0 m (e.g., LiFi soil sensor) to the height of the LiFi LED lamp H (e.g., LiFi drone). Practically, the suggested desk top height for a sitting and a standing person, which depends on the person’s height, may range from 0.57 m to 0.76 m and from 0.93 m to 1.26 m, respectively [69]. With reference to Figure 2a, the height of the arm horizontally stretching from the road, which can approximate the maximum height of holding a smartphone, is equal to 1.46 m. The default LiFi receiver height is assumed to be equal to 1.25 m for the computations of the next subsection;

(iii)

Road Width: The road width depends on the number of vehicle lanes when the width of a vehicle lane typically ranges from 2.70 m to 4.60 m [70]. Although the road width is not directly involved with the LiFi LED street lighting configuration parameters of Figure 3, it can indirectly affect the horizontal distance between the LiFi transmitter and LiFi receiver d;

(iv)

Horizontal Distance between LiFi Transmitter and LiFi Receiver d: Where the light of the LiFi LED street lighting configuration exists, an LOS transmission path theoretically exists. A LiFi receiver can stand: (a) directly under the LiFi LED lamp (d = 0 m); (b) on the sidewalk of the LiFi LED street lighting configuration -d ≤ 2.30 m –i.e., the maximum shortest distance is assumed to be equal to the half of the vehicle lane of Figure 2b,c on the road (when it is possible due to the road traffic); and (d) on the sidewalk across from the LiFi LED street lighting configuration (here, it should be taken into account the road width with its vehicle lanes). Practically, the first two cases are of interest for this paper and for this reason the distance ranges from 0 m to 2.5 m while the default horizontal distance between LiFi transmitter and LiFi receiver is assumed to be equal to 1 m for the computations of the next subsection; and

(v)

Incidence Angle ψ: Although the irradiance angle θ is well formed by the arms of the LiFi LED lamp vertical aiming and the LOS transmission path, the incidence angle also depends on the relative direction of the LiFi receiver, which further depends on the users’ behavior pattern, and would significantly affect LiFi communications performance [31]. In most published research, the LiFi receiver is assumed to be oriented perpendicularly upwards and the incidence angle ψ set equal to the irradiance angle θ as alternate interior angles [71]. But the best case for the incidence angle, that is the logical user’s behavior reaction to the fact that the LiFi signal originates from the LED lamp light, is when the LOS transmission path is perpendicular to the active surface of the LiFi receiver (normal vector); i.e., the incidence angle is equal to 0^o. Intuitively, the incidence angle ranges from 0^o to θ while the default incidence angle is assumed to be equal to θ for the computations of the next subsection.

The LiFi LED street lighting configuration parameters of this subsection with their values define a plethora of LiFi LED street lighting usage scenarios. To further assess the performance of LiFi LED street lighting systems, a suitable channel model is presented in the following subsection.

3.2. Channel Model and SNR Computations for the LiFi LED Street Lighting Systems

Similarly to other communications channels, LiFi channel modeling remains a challenge. In the literature, a wide variety of channel models has already been proposed especially for indoor VLC systems [18,31,71,72,73,74,75,76,77,78,79,80,81,82,83,84]. Conversely to the available indoor LiFi LED channels of the literature, LiFi LED street lighting channel model can safely focus on the LOS component and ignore the NLOS one as has already been analyzed in Section 2.3.

To compute the SNR of LiFi LED street lighting channels and further classify their SNR performance, the channel model needs to determine the frequency response of the LiFi communications channel. As far as the frequency response of the LiFi communications channel H(f) is concerned, it can be assumed to be relatively flat and near to DC; i.e., H(f) = H(0) [31,76,80]. Hence, the DC channel gain H(0) is applied to relate the average received optical power P_r with the average transmitted one P_t through

$$P_r=H\left(0\right)·P_t$$

(1)

Since it can be reasonably assumed for the LiFi LED street lighting systems of this paper that: (i) the LiFi transmitter can be treated as an LOS and generalized Lambertian transmitter; and (ii) the LiFi receiver can be treated as an LOS and generalized Lambertian receiver that consists of a photo-detector, a concentrator and an optical filter for effectively collecting the LiFi signal, the DC channel gain of the LiFi communications channel of Equation (1) can further be approximated by [31,76,80]:

$$H\left(0\right)=\frac{\left(p+1\right)·A}{2·\pi {·r}^2}{·cos}^{\rho }\left(\theta \right)·T_s\left(\psi \right)·g\left(\psi \right)·cos\left(\psi \right) , 0\le \psi \le {\psi }_c ,{\psi }_c<\psi$$

(2)

$$p=-\frac{ln\left(2\right)}{ln\left[cos\left({\theta }_{\frac{1}{2}}\right)\right]}$$

(3)

$$g\left(\psi \right)=\{\frac{n^2}{{sin}^2\left({\psi }_c\right)} , 0\le \psi \le {\psi }_c ,{\psi }_c<\psi$$

(4)

where p is the order of the Lambertian emission, ${\theta }_{\frac{1}{2}}$ is the half-power semi-angle of the LiFi transmitter, ψ_c is the LiFi receiver photo-detector FOV, A refers to the physical area of the LiFi receiver photo-detector, T_s(ψ) is the transmission efficiency of the LiFi receiver optical filter that can be assumed to have a constant value T_s in the cases of the high ψ_c values of this paper, g(ψ) is the gain of the LiFi receiver concentrator and n is the internal refractive index of the LiFi receiver photo-detector.

Given the frequency response of the LiFi communications channel of Equations (1) and (2), the SNR of the LiFi communications channel can be computed as follows [31,81,85,86]:

$$SNR=\frac{{\gamma }^2·P_r^2}{{\sigma }_{total}^2}=\frac{{\gamma }^2·{\left[H\left(0\right)\right]}^2·P_t^2}{{\sigma }_{total}^2}$$

(5)

$${\sigma }_{total}^2={\sigma }_{shot}^2+{\sigma }_{thermal}^2$$

(6)

$${\sigma }_{shot}^2=2·q·I_{bg}·I_2·B+2·q·\gamma ·{\rm B}·H\left(0\right){·P}_t$$

(7)

$${\sigma }_{thermal}^2=\frac{8\pi ·k·T_k·C_{pd}·A·I_2·B^2}{G}+\frac{16{\pi }^2·k·T_k·\Gamma ·A^2·C_{pd}^2·B^3·I_3}{g_m}.$$

(8)

where γ denotes the responsivity of the LiFi receiver photo-detector, ${\sigma }_{total}^2$ represents the standard deviation of the total noise that contains the shot noise variance ${\sigma }_{shot}^2$ and the thermal noise variance ${\sigma }_{thermal}^2$, q is the electronic charge, I_bg is the background current, I₂ is the noise bandwidth factor for the rectangular pulse shape, B is the noise bandwidth, k denotes the Boltzmann constant, T_k is the environmental temperature, C_pd is the fixed capacitance of the LiFi receiver photo-detector per unit area, G is the open-loop voltage gain, Γ is the FET channel noise factor, I₃ is the noise bandwidth factor for a full raised-cosine equalized pulse shape and g_m is the FET transconductance. With reference to Equations (5)–(8), the following remarks should be taken into consideration while considering the SNR computation of the LiFi communications channels:

• In contrast to conventional communications channels, the SNR of LiFi communications channels depends on the square of the received optical average power and not on the first power [31,76,79,86]. This implies that high transmitted optical powers and limited LOS transmission path losses are affordable in LiFi LED street lighting channels.
• As already mentioned in Section 2.2 and Section 2.3, LiFi LED street lighting networks are treated as a complementary technology to the WiFi networks of the smart cities and their use is expected when the LED street light lamps are turned on by the smart cities and power utilities. To examine the worst case scenario of LiFi LED street lighting system performance, ambient light due to sunlight is assumed. In such a case of bright skylight conditions, the preamplifier noise can be neglected, as already performed in the total noise computation of Equation (6) while the dominant noise component becomes the shot noise [31,76]. Here, it should be remembered that shot and thermal noises arise in the photodetector when it detects light signals, the shot noise is due to the number of photons collected by the photodetector and the thermal noise occurs as an energy equilibrium fluctuation phenomenon [81]. As already shown in Equation (6), the shot and thermal noises are also dependent on some environmental parameters, such as the ambient temperature, ambient light, etc. [87].
• To compute the SNR of the different LiFi LED street lighting usage scenarios of Section 4.1, apart from the default values of the LiFi LED street lighting configuration parameters of Section 3.1, the default values of the parameters of Equations (2)–(8) are reported in Table 1. Note that the default average transmitted optical power is assumed to be equal to 80 W with reference to Section 3.1 while the silicon-based photodetector OSD5-15T type of [76] is used at the LiFi receiver side.
• With reference to [76,79,88,89], to benchmark the SNR performance of the different LiFi LED street lighting usage scenarios of Section 3.1, the following SNR classification of channel areas is made: (i) excellent channel areas where SNR values are greater than 65 dB; (ii) good channel areas where SNR values range from 60 dB to 65 dB; (iii) acceptable channel areas where SNR values range from 40 dB to 60 dB; and (iv) unacceptable channel areas where SNR values are lower than 40 dB. Therefore, the SNR threshold of 40 dB is going to define the lowest bound of acceptable SNR value for the LiFi LED street lighting channels.

Table 1. Default parameter values for SNR calculations [76] (Reproduced with permission from [Azizan, L.A., Ab-Rahman, M.S., Hassan, M.R., Bakar, A.A.A.A. and Nordin, R.], [SPIE/Optical Engineering]; published by [SPIE], [2014]).

Parameter	Symbol	Value [Unit]	Parameter	Symbol	Value [Unit]
Average transmitted optical power	$P_t$	80 W	Noise bandwidth factor for the rectangular pulse shape	$I_2$	0.562
Half-power semi-angle of the LiFi transmitter	${\theta }_{\frac{1}{2}}$	60°	Noise bandwidth	B	50 × 106 Hz
LiFi receiver photodetector FOV	${\psi }_c$	70°	Boltzmann constant	k	1.38 × 10−23 J/K
Physical area of the LiFi receiver photodetector	A	10−4 m2	Environmental temperature	Tk	300 K
Transmission efficiency of the LiFi receiver optical filter	$T_s$	1	Fixed capacitance of the LiFi receiver photodetector per unit area	$C_{pd}$	1.12 × 10−6 F∕m2
Internal refractive index of the LiFi receiver photodetector	$n$	1.5	Open-loop voltage gain	G	10
Responsivity of the LiFi receiver photodetector	γ	0.21 A∕W	FET channel noise factor	Γ	1.5
Electronic charge	q	1.6 × 10−19C	Noise bandwidth factor for a full raised cosine equalized pulse shape	I3	0.0868
Background current	$I_{bg}$	5100 × 10−6A	FET transconductance	gm	30 × 10−3 S

Table 1. Default parameter values for SNR calculations [76] (Reproduced with permission from [Azizan, L.A., Ab-Rahman, M.S., Hassan, M.R., Bakar, A.A.A.A. and Nordin, R.], [SPIE/Optical Engineering]; published by [SPIE], [2014]).

Parameter	Symbol	Value [Unit]	Parameter	Symbol	Value [Unit]
Average transmitted optical power	$P_t$	80 W	Noise bandwidth factor for the rectangular pulse shape	$I_2$	0.562
Half-power semi-angle of the LiFi transmitter	${\theta }_{\frac{1}{2}}$	60°	Noise bandwidth	B	50 × 106 Hz
LiFi receiver photodetector FOV	${\psi }_c$	70°	Boltzmann constant	k	1.38 × 10−23 J/K
Physical area of the LiFi receiver photodetector	A	10−4 m2	Environmental temperature	Tk	300 K
Transmission efficiency of the LiFi receiver optical filter	$T_s$	1	Fixed capacitance of the LiFi receiver photodetector per unit area	$C_{pd}$	1.12 × 10−6 F∕m2
Internal refractive index of the LiFi receiver photodetector	$n$	1.5	Open-loop voltage gain	G	10
Responsivity of the LiFi receiver photodetector	γ	0.21 A∕W	FET channel noise factor	Γ	1.5
Electronic charge	q	1.6 × 10−19C	Noise bandwidth factor for a full raised cosine equalized pulse shape	I3	0.0868
Background current	$I_{bg}$	5100 × 10−6A	FET transconductance	gm	30 × 10−3 S

With reference to Table 1 and Section 4.1, different LiFi LED street lighting usage scenarios can be assessed with respect to their SNR behavior. More specifically, in Figure 4, the SNR contour plots are curved with respect to the x- and y-coordinates of the generic LiFi LED street lighting configuration of Figure 3 when the default parameter values are applied. From Figure 4, it is evident that the SNR contour plots present a center symmetry with a maximum SNR value of 52.44 dB just below the LiFi transmitter at the axis origin while an acceptable channel area, where SNR values remain above 40 dB, is established for horizontal distances from the axis origin that are less than 3.86 m. The previous horizontal distance threshold of 3.86 m is greater than the horizontal distance of 2.5 m of interest that was defined in Section 3.1. Note that the horizontal distance threshold, which is going to be used in the following analysis of this subsection, is defined as the distance beyond which the SNR value becomes less than 40 dB and the corresponding channel areas are unacceptable.

In Figure 5a, the SNR is plotted with respect to the horizontal distance from the axis origin for seven different LiFi transmitter height scenarios that are derived from the analysis presented in the parameter (i) of Section 3.1 when the default values are applied for the remaining parameters. In Figure 5b, the horizontal distance threshold is shown with respect to the seven different LiFi transmitter height scenarios of Figure 5a. It is clear that lower LiFi transmitter heights imply higher maximum SNR values while SNR values decrease more rapidly until they reach the LiFi receiver photodetector FOV of Equations (2) and (4). In contrast, greater LiFi transmitter heights imply lower maximum SNR values due to high LOS transmission path losses while the SNR values decrease slowly with respect to the horizontal distance since the irradiance angle changes remain low. In fact, there is a trade-off relationship between the horizontal distance threshold and the LiFi transmitter height; with reference to Figure 5b, the maximum horizontal distance threshold is achieved near to the default value of the LiFi transmitter height where the irradiance angle change and the LOS transmission path losses are average. The same figures as Figure 5a,b are given in Figure 6a,b, respectively, but for seven different LiFi receiver height scenarios that are derived from the analysis presented in the parameter (ii) of Section 3.1 when the default values are applied for the remaining parameters. Also, a trade-off relationship between the horizontal distance threshold and the LiFi receiver height occurs. Anyway, with reference to Figure 3, the results of Figure 5b and Figure 6b can be considered to be complementary since the difference between the LiFi transmitter height H and the LiFi receiver height h affects the LOS transmission path length r.

In Figure 7a, the SNR is plotted with respect to the horizontal distance from the axis origin for seven different incidence angle scenarios that are derived from the analysis presented in the parameter (v) of Section 3.1 when the default values are applied for the remaining parameters. In Figure 7b, the horizontal distance threshold is shown with respect to the seven different incidence angle scenarios of Figure 7a. As it was expected in the analysis of the parameter (v) of Section 3.1, it is verified that the best case for the incidence angle occurs when the incidence angle is equal to 0°, that is anyway the logical reaction towards the source of light and signal. Here, maximum SNR values and the horizontal distance threshold both deteriorate as the incidence angle increases till the incidence angle becomes equal to the LiFi receiver photodetector FOV. Note that the horizontal distance threshold difference may reach up to 2.75 m for the extreme incidence angles; say incidence angles of 0° and FOV.

In Figure 8a, the SNR is plotted with respect to the horizontal distance from the axis origin for seven average transmitted optical power scenarios that are derived from Table 1 when the default values are applied for the remaining parameters. In Figure 8b, the horizontal distance threshold is shown with respect to the seven different average transmitted optical power scenarios of Figure 8a. As already mentioned in advantage (iii) of Section 2.1, there are no restrictions concerning the transmitted optical power. In agreement with Equation (5), the maximum SNR values and the horizontal distance threshold both increase as the average transmitted optical power increases. Practically, the power consumption and the traffic/pedestrian purposes of the LED street lighting configurations of smart cities set the upper limit of the applied transmitted optical power. Note that the horizontal distance threshold can exceed 5.5 m and good channel areas where SNR values exceed 60 dB can occur when average transmitted optical powers that are greater than 200 W are applied. However, it is not necessary to always use the high average transmitted optical powers in all the LED street lighting configurations of a smart city, but they can be applied to specific LiFi LED street lighting configurations of interest.

From Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8, it is evident that each LiFi LED street lighting configuration can support its LiFi attocell that may be a part of the heterogeneous VLC–WiFi network across the smart city where various macro-, pico- and femto-cell environments coexist [15,90]. In fact, in the vast majority of the LiFi LED street lighting usage scenarios examined, acceptable SNR channel areas occur when the horizontal distance from the axis origin of Figure 3 is less than 2.5 m even though ambient light due to sunlight has been assumed during the computations of this subsection. Depending on the road type, the traffic intensity, the location inside the city and the application for pedestrian purposes, each LiFi LED street lighting configuration can adopt its own system parameters thus being characterized by its unique horizontal distance threshold and data rates. However, regardless of their system parameters, the LiFi LED street lighting configurations should access the backbone information network; i.e., the fiber optic backbone network that is installed across the transmission and distribution power grids and has been presented in [1]. Also, BPL networks may act as an intermediary technology since they can utilize the power line network to power the LiFi LED street lighting systems and bridge the information transfer between the LiFi LED street lighting systems and the fiber optic backbone network [82]. In the following subsection, LiFi LED street lighting network architectures that exploit the BPL networks to access the fiber optic backbone network of the transmission and distribution power grids are presented.

3.3. BPL Networks and LiFi LED Street Lighting Network Architectures

In the communications system architecture of [1], BPL networks can allow the required two-way information flow of the smart grid since they may support the operation of an advanced IP-based communications network enhanced with a plethora of broadband applications for power utilities, consumers and third parties upon the already installed wired power grid infrastructure and, at the same time, be interconnected with the other already installed communications solutions of the smart grid through their wired/wireless interfaces. With respect to [1], BPL networks, which are installed across the transmission and distribution power grids and are part of the communications system architecture, are expected to act as the intermediary technology between the fiber optic backbone communications network and LiFi systems of this section.

To study the BPL networks of the communications system architecture of [1], each network is divided into concatenated simpler BPL topologies while a simple BPL topology is shown in Figure 9. In fact, each BPL topology is bounded by its transmitting end (position A) and receiving end (position B) where BPL units are installed. These BPL units can be either BPL signal injectors/extractors or BPL signal repeaters, depending on the location and the role of the BPL topology across the overall BPL network [2,3,8,20,91,92,93,94,95,96,97,98,99]. To address the EMI concerns of the potential interference caused by the BPL systems, regulations and standards have been developed; with reference to [8,92], regulated EMI for BPL networks typically involves adherence to specific emission limits, signal power masks, technical requirements and compliance testing to ensure that BPL systems do not interfere with other already licensed communications services. To deliver traffic from/into the BPL network into/from the fiber optic communications network of [1], BPL gateways may be additively deployed onto the BPL units and interconnect the BPL network with the fiber optic communications network through the BPL unit wired/wireless interfaces. As far as the circuital properties of the simple BPL topology are concerned, except for the main power line of length ${\sum }_{i=1}^{N+1}{L}_i$, which anyway connects the transmitting and receiving ends, N branches may be encountered across the main power line. Each k branch, k = 1, N is located at distance ${\sum }_{i=1}^k{L}_i$ from the transmitting end at the position A_k while it is characterized by its length L_bk and its branch termination at the position ${A\prime }_k$. Branch terminations may be a BPL network piece of equipment or a power grid one; i.e., BPL units, matched terminations, open-circuit terminations, transformers, circuit breakers, isolators, various loads, street lighting configurations, etc.

As far as the integration of the LiFi and BPL systems is discussed, the first integration prototype was presented in [100] while significant progress has been so far achieved towards the integration application in indoor environments [18,42,82,101,102,103,104]. The indoor integration of LiFi and BPL systems comes from the fact that the LED lamps require their connection to power lines for their power supply while power lines can naturally act as a communications medium for LiFi systems given their activation by indoor BPL networks. It is evident that the advantages of indoor LiFi LED network architectures that integrate LiFi with BPL systems, such as these of [82,101,102], against the ones that suggest the exploitation of different wired communications media are the savings in wiring and the ease of installation. Hence, this paper focuses on the extension of the indoor LiFi LED network architectures to the outdoor environment of LiFi LED street lighting systems.

With reference to [82] and Figure 9, the extension of the indoor LiFi LED network architecture to the LiFi LED street lighting network architecture of this paper can be a simple matching task. Indeed, the two network architectures from the perspective of the deployed systems show the following similarities: (i) for either the indoor LiFi LED network or the LiFi LED street lighting one, the BPL network is treated as a “black box” backbone communications network thus performing its intermediary technology role well; (ii) the PLC modem of [82], which is added on each LED lamp and consists of the DAF unit, presents the same functionality with the BPL unit of Figure 9; i.e., both devices receive the required data from the data bus (i.e., power lines). In fact, their same operation is graphically synopsized by the light green block of data input in Figure 1; (iii) the “PLC to VLC” module of [82] modulates the signal to light. Its operation is graphically synopsized by the cyan blocks of the ADC and LED driver in Figure 1. The corresponding LiFi-BPL module of this paper is going to perform the same functionality with the “PLC to VLC” module between the BPL unit and the LiFi LED lamp for the LiFi LED street lighting systems; and (iv) LiFi LED lamps present the same functionality in both network architectures with differences only in their lamp technical characteristics. To highlight the integration of the LiFi and BPL systems, the typical simple BPL topology of Figure 9 is enriched with the aforementioned network components in Figure 10a where two LiFi LED street lighting systems are connected at the positions ${A\prime }_2$ and ${A\prime }_k$.

Conversely to indoor LiFi LED network architectures that are based on the PLC technology, the full exploitation of the BPL unit capabilities allows for the definition and application of more adaptive LiFi LED street lighting network architecture schemes, which may serve needs more efficiently, as shown in Figure 10a,b. Comparing the two schemes of LiFi LED street lighting network architectures of Figure 10a,b, their main differences have to do with the deployment position of BPL gateways as well as the number of the supported LiFi LED street lighting systems across the BPL topology. In both cases, BPL gateways present the same functionality; i.e., they exploit their wired/wireless interfaces so as to establish the connection with the corresponding switches of the fiber optic communications network of [1] (see Section 4). However, the deployment position of the BPL gateways depends on the number of the installed LiFi LED street lighting systems and, hence, the density of the access points of the LiFi LED street lighting network across the smart city. More specifically, in Figure 10a, the BPL gateway is deployed onto the BPL unit of the transmitting end while two LiFi LED street lighting systems are shown across the BPL topology. Scheme A intends to serve the traditional network concept where more than one LiFi LED street lighting system creates a relatively dense network of LiFi access points around the BPL topology area. BPL units of the LiFi LED street lighting systems are treated as traditional connected devices across the typical BPL topology of Figure 9. In contrast, in Figure 10b, the BPL gateway is deployed onto the BPL unit of the only LiFi LED street lighting system of the BPL topology. Scheme B is intended to exclusively serve individual LiFi access points so that the LiFi LED street lighting system simply appears as a matched termination to the rest of the BPL topology of Figure 9. Essentially, Scheme B describes a hybrid BPL network architecture where the BPL unit interfaces are only exploited through the BPL gateway while leaving open the prospect of further LiFi LED street lighting system exploitation by simply modifying the BPL topology parameters and BPL gateway location.

With reference to Figure 9 and Figure 10a,b, LiFi LED network architectures allow the integration of the LiFi and BPL systems so that the BPL networks can be connected through their gateways with the fiber optic backbone communications network the communications system architecture of [1]. In accordance with [1], the fiber optic backbone communications network across the HV, MV and LV power grids may act as a set of pathways to other communications networks, such as DSL, fiber optic, WPAN, WiFi, WiMAX, GSM and satellite through the appropriate connection of their respective gateways and the switches to the fiber optic backbone communications network. In the following section, the technology coexistence across the backbone communications network and the combined operation of fiber optics with LiFi LED street lighting networks are demonstrated while ideas regarding the LiFi project promotion are reported.

4. Communications System Architecture Expansion for LiFi LED Street Lighting Networks (Theoretical Field Installation) and LiFi Project Motivation

In this section, the architecture expansion of the fiber optic backbone communications network across the overhead transmission and distribution power grids of [1] is presented by taking into account the typical LiFi LED configurations with their corresponding SNR computations and LiFi LED street lighting network architectures based on the BPL technology of Section 3. A theoretical field installation of a fiber optic backbone communications network across a real overhead MV power grid with LiFi LED street lighting networks that exploit the capabilities of overhead LV BPL networks is discussed. Also, incentives for motivating the installation and use of LiFi LED street lighting networks are reported.

4.1. Theoretical Field Installation of LiFi LED Street Lighting Networks

With reference to [1,105,106], the spatial distribution of a real-world suburban power grid is shown in Figure 11a. In accordance with Table 2 of [105], this suburban 10 kV Chinese MV power grid covers an approximate area of 66 km² with an approximate population of 31,908 residents while presenting a radial structure with a main grid supply point. With reference to [1], it is assumed that the power grid of Figure 11a is an overhead MV power grid, which consists of overhead MV distribution lines and branches. Two fiber optic subnetworks are assumed to be supported by the main grid supply point (i.e., fiber optic subnetworks A and B) while an overhead MV BPL network has been already installed and operates across the power grid. The adopted distributed fiber optic backbone architecture, which is suitable for the large-scale overhead MV power grid of Figure 11a, supports the operation of three switches, three BPL gateways and twelve BPL units. For the purposes of the communications system architecture expansion, two LiFi LED field test areas, which are shown in pink color in Figure 11a, are reserved on the map to carry out the integration of two LiFi LED street lighting networks with the fiber optic backbone architecture through BPL networks.

In Figure 11b, the LiFi LED street lighting network architecture of the LiFi LED field test area 1 of Figure 11a is analytically demonstrated. MV/LV transformer 1 is connected at MV node 1 so that an LV distribution line (either overhead or underground) with its LV branches can carry the electrical power from the MV/LV transformer to the end customers of the area. At the same time, an LV BPL network with eight BPL units (say, BPL units A.L.1–A.L.8) is deployed across the LV distribution line and its branches. Here, the deployed LV BPL network helps towards the better operation and management of the LV power grid of the area through its real-time monitoring, metering and controlling of the surrounded power grid equipment and wired infrastructure (e.g., MV/LV transformer 1) while its BPL units remain the same with the ones of the MV BPL network [98,99,107]. As far as the BPL unit location and type are concerned, there are five BPL units across.

The LV distribution line (i.e., BPL units A.L.1, A.L.2, A.L.3, A.L.7 and A.L.8) while the remaining three BPL units act as the branch terminations and future connecting points (i.e., BPL units A.L.4, A.L.5 and A.L.6). All the BPL units of the area are BPL signal injectors/extractors except for the BPL units A.L.2, A.L.3 and A.L.7 that are solely BPL signal repeaters. Actually, the three BPL units that act as branch terminations may support three corresponding LiFi LED street lighting systems (i.e., LiFi LED 1, 2 and 3). Literally, each LiFi LED street lighting system consists of its BPL unit, its LiFi – BPL module and its LiFi LED lamp; for example, the first LiFi LED street lighting system consists of the BPL unit A.L.4, the LiFi – BPL module 1 and LiFi LED 1. With reference to Section 3.3 and Figure 10a, a LiFi LED street lighting network architecture scheme A is adopted for the integration of the three LiFi LED street lighting systems while the BPL unit A.L.8 with its wireless/wired interface is connected to the BPL gateway A.L.1 that is further integrated with the switch A.1 of the fiber optic subnetwork A, as detailed in [1]. Synoptically, the presented LiFi LED street lighting network architecture scheme A does not differ from the traditional BPL network concept that is enriched with LiFi access points as branch terminations. Note that the segmentation of the LV BPL network to LV BPL topologies of shorter lengths may allow the support of a denser LiFi LED street lighting network across a smart city.

In Figure 11c, the LiFi LED street lighting network architecture of the LiFi LED field test area 2 of Figure 11a is illustrated. Similarly to the LiFi LED field test area 1, the MV/LV transformer 2 is connected at the MV node 2 while it feeds an LV distribution line (either overhead or underground) with its LV branches. In contrast with the LiFi LED field test area 1, there is no installed BPL network across the LV distribution line in the LiFi LED field test area 2 but only one BPL unit that aims at exclusively serving an individual LiFi access point. In fact, only one LiFi LED street lighting system that consists of the BPL unit A.L.9, the LiFi–BPL module 4 and LiFi LED 4 is installed in the area. Exploiting its wireless/wired interfaces, BPL unit A.L.9 is connected to the BPL gateway A.L.2 that is further integrated with the switch A.2 of the fiber optic subnetwork A. Synoptically, the presented LiFi LED street lighting network architecture scheme B cannot be considered as a BPL-based network architecture but a rapid communications solution integration so that individual LiFi LED street lighting systems can create ad hoc connections and get connected to the fiber optic backbone communications network.

The main advantage of the applied communications system architecture across a smart city is its scalability and its network expansion capabilities. In fact, the deployment of gateways may allow other available communications technologies, such as DSL, fiber optic, WPAN, WiFi, WiMAX, 5G, GSM, satellite and LiFi networks to be added to the existing communications system architecture. In fact, the complementary use of the aforementioned communications technologies under the aegis of the terabit-class fiber optic backbone communications network of the smart city’s transmission and distribution power grids offers the opportunity to improve resource management of many assets related to city life and urban QoL, including smart tourism, intelligent transportation systems, smart buildings, space/occupancy management (indoors and outdoors), resource monitoring and sensing, immersive IoT services (including wearables and crowdsensing), sustainability and the greening of the environment (energy consumption, pollution monitoring) [108]. Anyway, the easy access to very high-speed communications services through the gateways of various communications technologies is a promise for urban economic growth and development since these services can be delivered from either power utilities or smart cities to either the power utilities themselves or to smart cities themselves or to third parties by leasing/selling or to consumers [1,99,109].

This paper addresses the first step of a wider investigation, which consists of proposing guidelines for integrating fiber optics, BPL networks and LiFi LED street lighting networks in smart cities, with sustainable criteria while the research methodology scheme that is followed is shown in Figure 12. With reference to [110] and Figure 12, the research methodology relies on a case study strategy according to three major phases, namely:

• (1) data collection; (2) evaluation (proof of concept); and (3) development of proposal. The first phase of data collection, which is the review part of the paper, has entailed both theoretical work (Phase 1a) and field work (Phase 1b): it started with a comprehensive literature review of relevant and up-to-date citations for LiFi technology and systems in Section 2 while quantitative field work obtained a compilation of data for further analysis, characterization and comprehensive knowledge of LiFi LED street lighting configurations in Section 3.1. The second phase of the evaluation (proof of concept), which consists of numerical results concerning case studies of LiFi LED street lighting systems (Phase 2a) and LiFi LED street lighting network architectures (Phase 2b) rather than simulations, was aimed at obtaining the findings of Section 3.2 and Section 3.3 concerning the operational performance of LiFi LED street lighting systems and networks, respectively. After diagnosing the consistency of Phases 2a and 2b, an integration approach and general theoretical principles establishment of Section 4.1, as well as specific ones regarding the communications system architecture expansion, have been determined. The third phase is going to be divided into three work stages: project motivation (Phase 3a), economic assessment (Phase 3b) and conclusions (Phase 3c, see Section 5). In the following subsection, Phases 3a and 3b are further analyzed: incentive ideas for power utilities and smart cities so that the LiFi LED street lighting network services become known are highlighted while the integration of LiFi LED street lighting networks with the fiber optic backbone architecture through BPL networks can be more efficiently and profitably supported by power utilities and smart cities.

4.2. Future Directions for the Exploitation of LiFi LED Street Lighting Networks in Smart Sustainable Cities

This subsection focuses on the future directions for the exploitation of LiFi LED street lighting networks in smart sustainable cities; i.e., how power utilities and smart cities can leverage their existing infrastructure and services to boost the acceptance of LiFi LED street lighting networks as a broadband communications technology alternative and their provided smart-living and QoL services. At the same time, the LiFi LED street lighting networks can become a profitable marketing tool for both power utilities and smart cities; i.e., a win-win business venture for companies, municipalities and individuals.

As far as the side of the power utilities is concerned in the business venture, new market conditions in power industry, such as the possibility of free choice among electricity’s suppliers given to the individual customers, force power utilities to innovate, experiment with new business strategies and consider promoting new marketing tools [111]. Among the available marketing tools, benefit programs are offered by companies to stimulate continued patronage among consumers through discounts, coupons, free products or special services [112]. In fact, the decision to establish a benefit program comes from a commonly held apocryphal belief that it is six to seven times more expensive to acquire a new customer than it is to retain a current one and to encourage them to increase their consumption [113,114]. Free high-speed internet access and the free charging of smart devices under the activated LiFi LED street lighting configurations can be considered special services of a power utility’s benefit program focusing on customer loyalty, collecting customer and shopping habit data, retaining customers and selling more rewarding frequent shoppers, and promoting customized offers [115]. Finally, as has already been mentioned, LiFi is a sustainable and eco-friendly communications technology thus power utilities that invest in LiFi technology may benefit from the green marketing that further entails increased customer loyalty and a positive brand reputation, apart from the environment protection and the contribution towards a more sustainable future.

As far as the side of the smart cities is concerned in the business venture, the smart city’s applications, storage and processing capabilities are concentrated in cloud data centers. Heterogeneous communications technologies are interconnected across the entire smart city’s region and, of course, to the smart city’s communications infrastructure so that edge networks, such as WSNs, DSL, fiber optics, WPAN, WiFi, WiMAX, 5G, GSM, LiFi and other public networks can provide the broadband internet connectivity and the access to the smart city’s capabilities to the smart city’s residents and tourists [116]. The voracious appetite for broadband speed from a heterogeneous smart city’s IoT devices (e.g., traffic cameras, drones, sensors) and the explosive increase of personal wireless devices (e.g., smartphones, tablets, wearables) both represent an ever-rising bar for smart city communications infrastructure and edge networks. As has already been mentioned, the ubiquity of LED street lighting in the smart city’s environments could be a strong complementary wireless technology to WiFi access points. Apart from free internet access, the smart city’s benefit program can include the free charging of personal wireless devices under the LiFi LED street lighting configurations so that smart city’s residents and tourists can easily gather in common areas and engage in face-to-face and virtual communications, encourage integrated real-time promotions/incentives/coupons from local stakeholders and provide real-time information about the smart city’s activities [117]. Especially from the perspective of smart tourism, the smart tourism experience can be treated as the benefit program pinnacle of an extensive technological pyramid consisting of services derived from smart city’s applications, access to communications networks/the cloud and the access to power charging (e.g., Barcelona smart tourism ecosystem) [118]. All the aforementioned smart tourism requirements can be fulfilled by terabit-class backbone communications network architecture across the smart city’s existing transmission and distribution power grids that has been proposed in Section 4.1 where WiFi and LiFi networks can coexist. Finally, as the terms smart cities and sustainability come intertwined, smart city planning involves implementing eco-friendly projects, such as LiFi LED street lighting systems, that may improve the QoL of residents and tourists while promoting environmental awareness [119,120,121].

5. Discussion and Conclusions

In this paper, a holistic approach to the exploitation of the LiFi LED street lighting networks across a smart city has been presented. First, an overview of the LiFi technology has been given where the advantages, disadvantages and possible applications of this VLC technology have been recognized. Special attention has been given to the advantages, disadvantages and applications concerning the concept of LiFi LED street lighting networks across a smart city’s infrastructure under the aegis of power utilities and smart cities. Apart from the extremely high data rates and their easy implementation across the smart city’s existing transmission and distribution power grids, it has been underlined that the LiFi LED street lighting systems may simultaneously deliver illumination, broadband communications and power charging by using the same existing lighting infrastructure. The complementary use of LiFi and WiFi networks during the daytime hours due to the interference from sunlight has been noted. Second, LiFi LED street lighting configurations and their network architectures have been investigated. Typical LiFi LED street lighting configurations that are encountered in smart cities have been presented while a suitable channel model for the LiFi LED street lighting configurations with its default model parameters has been analyzed. Based on the applied LiFi channel model, SNR computations have been performed and have revealed that acceptable channel areas where SNR values exceed 40 dB may occur for horizontal distances from the typical LiFi LED street lighting configurations that may reach up to 5 m. A wide variety of parameter scenarios have been examined that have validated the previous SNR findings concerning acceptable channel areas. As far as the available LiFi LED street lighting network architectures are concerned, two schemes have been proposed that exploit the BPL network capabilities of the communications network of the smart city’s HV, MV and LV power grids. Third, the architecture expansion of the communications system of [1] has been demonstrated across a real overhead MV power grid so that the fiber optic backbone communications network, the BPL networks and the LiFi LED street lighting networks can coexist for satisfying the smart city’s needs for ubiquitous broadband internet access. In fact, two LiFi LED field test areas have been reserved in the smart city map where a theoretical field installation of the two proposed LiFi LED street lighting network architecture schemes has been demonstrated. Fourth, the first research step towards the integration of broadband communications technologies with the smart grid backbone network architecture of [1] has been proposed in this paper via the LiFi LED street lighting networks. The integration of other smart grid communications solutions, such as WPAN, WiFi, WiMAX, 5-6G, GSM and satellite, with the smart grid backbone network architecture of [1], their performance assessments and their performance comparison are among the future research steps. Finally, incentive ideas for promoting the win-win business venture of LiFi LED street lighting networks among the power utilities, smart cities and individuals have been reported.