Open Access
Add to BibSonomy
Abstract
Monitoring climate change can be accomplished by deploying Internet of Things (IoT) sensor devices to collect data on various climate variables. Providing continuous power or replacing batteries for these devices is not always available, particularly in difficult-access locations and harsh environments. Here, we propose a design for a self-powered weather station that can harvest energy, decode information using solar cells, and is controlled by a programmable system-on-chip. A series of experimental demonstrations have shown the versatility of the proposed design to operate autonomously.
© 2022 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Simultaneous lightwave information and power transfer (SLIPT) has received considerable attention as a green technology suitable for next-generation internet of things (IoT) and the internet of underwater things (IoUT) devices. SLIPT is the optical counterpart of the simultaneous wireless information and power transfer (SWIPT) in the radio frequency (RF) technology [1,2]. Although there has been considerable progress with electromagnetic power transfer, SWIPT suffers from low energy conversion efficiency and receiver circuit design challenges [3]. SWIPT is equally subjected to possible health issues caused by electromagnetic radiation. However, SLIPT benefits from the broad unlicensed light spectrum and the continuous advances in solar cell (SC) technologies. Many proof-of-concept studies and experiments have been conducted, showing huge potential for SLIPT-based systems.
A seminal experiment was performed by Kim and Won, who reported the use of a large area SC to receive visible light communication (VLC) signals at a low data rate of 3 Kbps while harvesting solar energy [4]. Their motivation was to use SCs instead of state-of-the-art photodiodes that require external electric power supplies to decode signals. Using a commercially available solar panel, Wang et al. designed an optical receiver that was used to establish a short communication link of tens of cm with a data rate of 11.84 Mbps while generating 2 mW of power [5]. An organic SC with an active area of 8 mm$^2$ (similar to those that are widely integrated into portable devices) was used to decode information signals transmitted at a 34.2 Mbps rate while harvesting energy [6]. The transfer of 7.2 W of optical power over a 30 m distance indoor at nighttime was reported in [7]. A silicon (Si)-based SC with 5-cm$^2$ area was used in [8] to detect optical signals at a wavelength of 405 nm propagating through a 7-meter long underwater channel. Fakidis et al. established a 500 Mbps free space connection over a propagation length of 2 m using a gallium arsenide (GaAs) photovoltaic cell with a small active area of 0.8 mm$^2$, illuminated by an 847 nm vertical-cavity surface-emitting laser (VCSEL) [9]. The high-speed experimental demonstration involved an optical orthogonal frequency division multiplexing (OFDM) digital modulation technique with adaptive bit and power loading that required offline processing [9]. Follow-up studies showed that the achievable rates could touch the 1 Gbps boundary with GaAs cells illuminated by infrared (IR) light VCSELs [10,11]. A hybrid SC system, consisting of a mono-crystalline silicon solar panel for energy harvesting and a thin-film amorphous silicon solar panel for signal detection, is demonstrated in [12] and reported the decoding of 1.2 Mbps and 1.6 Mbps signals. A data rate of 53 Mbps was reported in [13] using a custom-made perovskite SC with an active area of 6.5 mm$^2$. The detected signals were emitted by a 40 cm far 660 nm laser diode and modulated using OFDM modulation with adaptive bit and power loading [13]. The optimization between the communication and energy harvesting functions has been equally the subject of several theoretical studies on VLC systems for indoor and outdoor applications [2,14].
We previously presented the different SLIPT approaches in various domains, namely time, power, and space [15]. We also highlighted the potential of SLIPT to charge next-generation IoUT devices. We reported two underwater time switching SLIPT demonstrations. The first demonstration involved charging a module equipped with temperature and turbidity sensors. The temperature sensor was then used to monitor the temperature variation of a water tank. The second demonstration consisted of charging the super-capacitor of a submerged IoUT device using an LED source. The IoUT device was then used to stream a video captured in real-time by an integrated analog camera using a red laser diode. We also provided the main open problems and future research directions related to SLIPT in a point-to-point configurations and network settings.
In addition to VLC and underwater communication systems, the concept of simultaneous information decoding and energy harvesting has been further suggested as a key solution to tackle the connectivity divide for rural areas and scattered communities by using off-the-shelf solar panels as optical receivers while also harvesting solar energy [16,17]. Indeed, a real-world demonstration reported the detection of signals emitted from a distance of 30 m at a data rate of 8 Mbps using a commercially available solar panel that could also harvest 5 W of power from sunlight [17].
With the progress in optical wireless communication (OWC) systems, SLIPT can make niche applications to provide continuous powering to remote devices while communicating, including those used for environment control and global climate change monitoring. Indeed, connected, self-powered IoT devices are the key to fulfilling these missions [18]. In this context, we demonstrate in the present work a self-powered weather station capable of receiving information and operation commands using large-area solar panels that are equally used to power the designed platform. A dual solar panel configuration is considered to allow multiple users to communicate simultaneously with the same station, fasten the platform charging process, and fulfill the simultaneous information decoding and energy harvesting. A series of demonstrations are conducted to show the potentials of the proposed system, which can be installed in difficult access and remote locations (as illustrated in Fig. 1). In the first demonstration, we show the ability of the system to decode information with one SC and operate by harvesting energy with another SC without the use of a battery. The second demonstration involves decoding two independent signals from different laser sources using the system’s two SCs and the simultaneous charging and information transfer using the same laser source. The final demonstration consists of weather data collection relying on solar harvested energy stored on a battery for over one week. The designed platform is another step toward autonomous IoT devices for climate change monitoring.
Fig. 1. An illustration of the self-powered weather station for climate change monitoring applications and which can be installed in remote areas and difficult access locations.
2. System design
The designed weather station is equipped with multiple sensors to measure temperature, relative humidity, pressure, dust, and wind speed. The system also contains a set of sensors to monitor various gas concentrations in the atmosphere, such as carbon dioxide (CO$_{2}$), carbon monoxide (CO), ammonia (NH$_{3}$), hydrogen (H$_{2}$), and methane (CH$_{4}$). Visible and infrared (IR) light sensors are equally installed. The different sensors are placed inside a solar radiation shield to protect them from sunlight, rain, and other weather conditions (such as dust storms), except for the wind, dust, and light sensors, which need to be kept directly exposed to external conditions.
Two silicon solar panels are fixed on top of the platform and connected to the system circuitry. The control circuit is a programmable system-on-chip (PSoC) that is powered by a battery. The PSoC microcontroller (MUC) is chosen to allow seamless integration with the SLIPT circuit design and reading the sensors, logging the data, receiving, decoding, and transmitting the information. The PSoC can be, in particular, programmed to control the charging process of a system battery by a SC. For example, in the case of a low battery, it is set to a photovoltaic mode to feed the circuit. Once the battery is fully charged or above a pre-defined threshold, the PSoC inverts the solar panel’s polarity (the solar panel becomes reverse-biased), which is then switched to a photoconductive mode.
The main circuit and the system diagram are depicted in Fig. 2. The MUC, shown in Fig. 2 (a), comes with a transimpedance amplifier, operational amplifiers, a universal digital block (UDB), look-up tables, and high-speed and low-power comparators. Seemly as a field-programmable gate array, the PSoC can run in parallel different synchronous state machines using digital and analog blocks. The circuit box is covered by PMMA (Polymethyl methacrylate) glass to protect it from dust and humidity.
The system battery can be charged by the two $20\times 13$ cm$^2$ solar panels, denoted as SP1 and SP2, and connected to a switch circuit, as seen in Fig. 2(b). The switch circuit is connected to a maximum power point tracking (MPPT) controller to optimize the match between solar power and the harvested energy in the battery as the amount of received power from light varies in the atmosphere [19]. The MPPT of the solar panel in our circuit works without affecting the decoded signals, as both SC modes are independent. Both solar panels are also used to decode information received from one or two transceivers in case that they are reverse-biased and operating in a photoconductive mode. The switching mechanism between the two SC modes is based on a low-power relay that consumes 1.47 mJ of energy during the switch. Importantly, the PSoC independently controls both solar panels; thus, it is possible to set both SCs to operate in different modes. The experimentally measured maximum harvested power levels by each of the solar panels per light source are given in Table 1. All the sources were fixed at 20 cm from the solar panel in a dark measurement site.
The platform is equipped with a 638 nm red laser to communicate the collected information by the different installed sensors to a data collection station or forward information to a nearby device. Five sensor units are connected to the main circuit, as shown in Fig. 2(b). Sensor 1 is a BME680 sensor that measures temperature, in the -40$\sim$85 [$^\circ$C] range, humidity, in a relative range from 0 to 100%, and pressure in the 300-1100 [hPa] range (hPa: hectoPascal). Sensor 2 is an MH-Z16 CO$_{2}$ sensor that measures in the 1-5000 [ppm] range (ppm: parts per million). Sensor unit 3 comprises two sensors; a SiLabs SI1145 light sensor and a Seeed Studio PPD42NS dust sensor. The light sensor measures the visible and IR light and approximately calculates the ultraviolet (UV) index using a sensing algorithm based on the visible and IR light from the sun [20]. The IR sensor spectrum is 550-1000 nm with a central wavelength of 800 nm, and the visible light sensor spectrum is 400-800 nm with a central wavelength of 530 nm. The dust sensor measures the concentration of particles with sizes of more than 1 $\mathrm {\mu }$m and returns a value in [pcs/l] (particles/liter). Sensor 4 is an Adafruit MiCS5524 mixed gas sensor that measures CO (in the 1 to 1000 [ppm] range), ammonia (in the 1 to 500 [ppm] range), ethanol (in the 10 to 500 [ppm] range), hydrogen (in the 1 - 1000 [ppm] range), Methane / Propane / Iso-Butane (> 1000 [ppm]). Note that Sensor 4 does not tell which gas is measured in particular but returns a mixed concentration of all the gases that the sensor unit can measure. Sensor 5 is an Adafruit anemometer that measures wind speed in the range from 0.2 to 32.4 [m/s]. An SD card is installed on the system to save the recorded data of the sensors. The power consumption of the sensors (calculated using the manufacturers’ specifications of the sensors) and the circuit board are summarized in Table 2.
Fig. 2. (a) A photograph of the circuit board. (b) An illustration of the circuit diagram and the various connected sensor units, powered by two solar panels (SP1 and SP2) that can also be used to decode information. UART: Universal Asynchronous Receiver/Transmitter; DAC: Digital-to-Analog Converter; I2C: Inter-Integrated Circuit (protocol); MMPT: Maximum Power Point Tracking RTC: Real-Time Clock; SD: Secure Device Memory Card; SP: Solar Panel.
Table 1. Experimental maximum harvested power by source.
Table 2. Power consumption by component.
3. Experimental methodology
Three experimental demonstrations were conducted using the designed sensor platform. The main objectives of the demonstrations, testing the capability of the designed platform, are summarized in Table 3.
Table 3. Summary of the objectives of the conducted demonstrations.
3.1 Demonstration 1: a battery-free sensing platform outdoor testing
In the first experimental demonstration, we propose powering the platform with the solar panel SP1 using a light source and decoding information using solar panel SP2 in an outdoor environment at nighttime. This scenario can be viewed as a SLIPT demonstration in the space domain, where the two SLIPT approaches are fulfilled simultaneously. The platform does not have any connected battery in this demonstration. Only SP1 is used to power the system in real-time, while panel SP2 is used solely for information decoding. A 100-W off-the-shelf incandescent (office) lamp is installed in proximity to SP1 (28 cm). A 430 nm blue laser diode, placed at 52 m from the weather station, is used to transmit data. The experimental setup for this demonstration is depicted in Fig. 3.
Fig. 3. An illustration of the experimental setup of the simultaneous charging and communication with the designed system that has not any battery. SP1 is used to power the system by harvesting energy from a nearby incandescent lamp. SP2 is used to decode the information carried by the blue laser beam sent by a transmitter located at a distance of 52 m.
At 52 m, a 115 Kbps on-off keying (OOK)-based signal transmission rate is achieved with a bit error rate (BER) of $5\times 10^{-4}$. The system operation is solely insured using the 360 mW harvested power using SP1. This amount of power is enough to perform the decoding of received optical signals using SP2 but cannot power the sensors at nighttime. Therefore a battery is needed for the sensing operation.
3.2 Demonstration 2: independent control of both solar panels
The second scenario consists of testing the system’s ability to decode independent information signals received by the two solar panels and also charging and communicating with the platform simultaneously using the same light source. To test the ability of the system to handle signals received simultaneously from two different transmitters, each by a solar panel, two directly modulated blue laser diodes are aligned with each of the solar panels with a separation distance of 5.5 m, as can be seen in Fig. 4 (a). The system was able to decode 115 Kbps OOK signals simultaneously with a BER of $6\times 10^{-6}$ per channel. A directly-modulated light beam from a 430 nm laser diode is then divided into two using a 50:50 beam splitter (see Fig. 4 (b)). The first light beam is aligned with SP1, and the second is directed to SP2 using a mirror. SP1 is set to operate in the photovoltaic mode, while SP2 operates in the photoconductive mode.
The distance between the laser and the weather station is kept to 5.5 m. Error-free transmission of an OOK 115 Kbps signal between the laser and the sensing platform is obtained. The open-circuit voltage and the short circuit current of SP1 are 3.672 V and 0.02 mA, respectively.
Fig. 4. Schematic illustration of simultaneously (a) communicating with the weather station from two sources, (b) charging and communicating using the same source. BS: Beam Splitter.
3.3 Demonstration 3: outdoor data collection
The third demonstration consists of fixing the system on an apartment balcony to collect weather variables, light, dust concentration, and gas emission data using the various installed sensors over one week ($\sim$ 210 operation hours). A 12 V 7 Ah (ampere-hour) battery is used to power the system. Both solar panels are set to power the system battery continuously. A photograph of the setup is shown in Fig. 5. The sensors are programmed to measure every 30 minutes and enter a sleep mode after each measurement to maintain the battery’s lifetime. A warm-up time is needed for the CO$_{2}$ and the mixed gas sensors. For such a reason, we programmed the system to start measuring after one minute of operation to provide correct data. The measurement data is saved on the SD card fixed on the control circuit.
The system operated smoothly over the whole measurement period. The collected data from the various sensor units for approximately 9 days were saved on the SD card installed on the system. At the end of the measurement period, the data was collected from the weather station using the system’s red-light laser. The obtained measurements for temperature, ambient pressure, relative humidity, and wind speed are shown in Fig. 6. The collected temperature measurements, shown in Fig. 6(a), were found to be slightly higher, in the daytime, than the temperature data for the region of Thuwal collected in the same period. The main reason for the temperature difference is that the used solar radiation shield was fully exposed to the sun and placed near a metal frame on the balcony where the measurements were conducted, which results in a rise in the recorded temperature. The fluctuations of the relative humidity seen in Fig. 6(b) are also caused by the high daytime temperature inside the solar radiation shield. The particle concentration, CO$_{2}$, and mixed gas concentration are shown in Figs. 7(a)-(c). A strong particle concentration in Fig. 7(a) can be seen and the two spikes correspond to two dust storms. The recorded data for the visible light illumination, IR light, and the UV index is depicted in Fig. 7(d)-(f) with a clear daytime/nighttime pattern. The minimum visible light illumination is around $\sim$120 lux, caused by the outdoor lighting system. The IR illumination readings are in [ADC counts] (analog to digital converter counts), and the minimum IR illumination value is around $\sim$253 ADC counts. The average daily maximum UV index is 8.5658.
After more than one week of continuous operation, we noticed that the nominal power that can be harvested by each of the solar panels dropped by $\sim$30$\%$ due to the accumulated dust, in particular, that both solar panels were covered with thin plastic layers that can get sticky when heated with sunlight.
Fig. 5. A photograph of the weather station installed on an apartment balcony to collect data over a period of one week. The positions of the solar panels were kept fixed over the entire testing period.
Fig. 6. Recorded (a) temperature in [$^\circ$C], (b) ambient pressure in [hPa], (c) relative humidity, and (d) wind speed in [m/s] over a period of $\sim$ 9 days.
Fig. 7. (a) Dust particles (with sizes more than 1$\mu$m) concentration in [pcs/l] over a 30 second measurement time window. The two spikes correspond to two quick dust storms that are common in the university region in October. Recorded (b) CO$_{2}$ concentration in [ppm] (c) mixed gas concentration data in [ppm] over a period of $\sim$ 9 day. Recorded (d) visible light illumination data in [lux] (e) IR light in [ADC counts] and (f) UV Index data over a period of $\sim$ 9 days. Note that, as we pointed out in section 2., the UV index is not directly measured by the sensor with an actual UV light sensor but rather approximated based on the visible and IR light from the sun.
4. Discussion
The main novelty of the presented work is the ability of the designed system to operate autonomously over several days using two solar panels, each with an active area of 260 cm$^{2}$ without incurring any bulky cabling to be connected to the network. Yet, further improvements in sensor calibration and outdoor sensor protection are required.
The use of large-area SCs to decode information eases the pointing, acquisition, and tracking requirements imposed for the state-of-the-art optical wireless communication systems using photodetectors or SCs with active areas in the order of sub-mm$^2$ [9–11]. Indeed, the large active area comes at the cost of the detection bandwidth, i.e., the larger the solar panel, the slower the communication speed. However, for some IoT applications, a data rate in the order of a few hundred Kbps is enough to transfer operational commands from an unmanned autonomous vehicle (UAV) or a distant control station to an IoT device. SCs also perform better in outdoor lighting conditions and solar exposure compared to various photodetector types, as demonstrated in [21]. Note that an avalanche photodetector and P-type intrinsic N-type (PIN) detector were considered for the comparison with SCs in [21]. In harsh outdoor environments, solar panels should also be covered with protective layers to maintain operational performance over several months.
Using real-valued advanced modulation formats, such as OFDM with adaptive bit or power loading algorithms, can increase the transmission capacity [13]. But usually, OFDM modulation imposes additional hardware complexity and requires offline digital signal processing operations [12]. Performing such operations online can derive extra cost to the system circuit. Using an array of small-size solar panels in a “Massive MIMO” (multiple-input multiple-output ) fashion can increase the transmission capacity by decoding independent signals from the same or different users using multiple solar panels. A multi-solar panel configuration can also improve the charging performance if the solar panels are pointed in different directions to ensure solar exposure over longer daylight hours without the need for active daily sunlight tracking. A future idea to discover is to use MIMO in a spatial diversity scheme to decode the same data signal using multiple SCs. Spatial diversity could be an effective solution to pointing errors or beam wandering induced by atmospheric turbulence in outdoor applications [22]. Another idea to explore is integrating a low-power and small-footprint RF module to transmit and receive data in non-line-of-sight (NLOS) situations. A futuristic concept to further consider is to use UV light sources for NLOS communication scenarios. The enhanced scattering of UV light can guarantee NLOS transmissions ([23], and references within). The provided data rates by NLOS UV communication can be enough to transfer weather data collected by a remote station.
Another crucial idea to discuss is building self-powered systems that could decode signals with wavelengths within the low atmospheric attenuation windows (wavelength bands where the effects of scattering and absorption are low) and yet fulfilling the eye safety requirement to ensure communication reliability. Three bands fulfill these criteria: (850 nm, 1230-1270 nm, and 1470-1670 nm). There are commercially available laser sources that could operate in the 850 and 1470-1670 nm bands. Laser sources around the 850 nm band are expensive and unsuitable for IoT applications, yet not entirely eye-safe. For the 1470-1670 nm, there are plenty of laser sources. Silicon SCs are the commonly used receivers for SLIPT demonstrations, as we discussed in Section 1., but such SCs exhibit low conversion efficiency at these bands. GaAs SCs are efficient for those low-absorption IR bands. Nevertheless, maintaining strict alignment due to the small areas of those solar panels is required [9]. In this case, a possible idea to expand the field of view for these SCs would be to integrate it with fused fiber optic tapers [24]. Designing a solar panel that matches the wavelength of the transmitting laser can further increase conversion efficiency and drive it nearly to 90%.
5. Conclusion
In summary, we demonstrated a stand-alone weather station that can collect weather data and monitor particles and gas concentrations in the atmosphere. Two large area solar panels are used to charge the designed system’s battery and decode the received optical signals. The proposed design contains a laser that can transmit the collected data to a central location or a nearby operator. With three demonstrations covering various scenarios, we demonstrated the versatility of the proposed system to operate using harvested energy and also sunlight over a long period. Our designed weather station paves the way toward smart and self-powered autonomous IoT devices. Future demonstrations will involve high efficiency and specially designed SCs that could enhance the system performance in terms of harvested energy and the supported data rates.
Funding
King Abdullah University of Science and Technology.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. T. D. Ponnimbaduge Perera, D. N. K. Jayakody, S. K. Sharma, S. Chatzinotas, and J. Li, “Simultaneous wireless information and power transfer (SWIPT): Recent advances and future challenges,” IEEE Commun. Surv. Tutorials 20(1), 264–302 (2018). [CrossRef]
2. P. D. Diamantoulakis, G. K. Karagiannidis, and Z. Ding, “Simultaneous lightwave information and power transfer (SLIPT),” IEEE Trans. on Green Commun. Netw. 2(3), 764–773 (2018). [CrossRef]
3. X. Lu, P. Wang, D. Niyato, D. I. Kim, and Z. Han, “Wireless charging technologies: Fundamentals, standards, and network applications,” IEEE Commun. Surv. Tutorials 18(2), 1413–1452 (2016). [CrossRef]
4. S.-M. Kim and J.-S. Won, “Simultaneous reception of visible light communication and optical energy using a solar cell receiver,” in 2013 International Conference on ICT Convergence (ICTC), (2013), pp. 896–897.
5. Z. Wang, D. Tsonev, S. Videv, and H. Haas, “On the design of a solar-panel receiver for optical wireless communications with simultaneous energy harvesting,” IEEE J. Select. Areas Commun. 33(8), 1612–1623 (2015). [CrossRef]
6. S. Zhang, D. Tsonev, S. Videv, S. Ghosh, G. A. Turnbull, I. D. W. Samuel, and H. Haas, “Organic solar cells as high-speed data detectors for visible light communication,” Optica 2(7), 607–610 (2015). [CrossRef]
7. J. Fakidis, S. Videv, S. Kucera, H. Claussen, and H. Haas, “Indoor optical wireless power transfer to small cells at nighttime,” J. Lightwave Technol. 34(13), 3236–3258 (2016). [CrossRef]
8. M. Kong, B. Sun, R. Sarwar, J. Shen, Y. Chen, F. Qu, J. Han, J. Chen, H. Qin, and J. Xu, “Underwater wireless optical communication using a lens-free solar panel receiver,” Opt. Commun. 426, 94–98 (2018). [CrossRef]
9. J. Fakidis, S. Videv, H. Helmers, and H. Haas, “0.5-Gb/s OFDM-based laser data and power transfer using a GaAs photovoltaic cell,” IEEE Photonics Technol. Lett. 30(9), 841–844 (2018). [CrossRef]
10. J. Fakidis, H. Helmers, and H. Haas, “Trade-off between energy harvesting and data communication towards a 1 Gb/s laser and photovoltaic link,” in Optical Wireless Fiber Power Transmission Conference, (2020), pp. 1–3.
11. J. Fakidis, H. Helmers, and H. Haas, “Simultaneous wireless data and power transfer for a 1-Gb/s GaAs VCSEL and photovoltaic link,” IEEE Photonics Technol. Lett. 32(19), 1277–1280 (2020). [CrossRef]
12. M. Kong, J. Lin, Y. Guo, X. Sun, M. Sait, O. Alkhazragi, C. H. Kang, J. A. Holguin-Lerma, M. Kheireddine, M. Ouhssain, B. H. Jones, T. K. Ng, and B. S. Ooi, “AquaE-lite hybrid-solar-cell receiver-modality for energy-autonomous terrestrial and underwater internet-of-things,” IEEE Photonics J. 12(1), 1–13 (2020). [CrossRef]
13. N. A. Mica, R. Bian, P. Manousiadis, L. K. Jagadamma, I. Tavakkolnia, H. Haas, G. A. Turnbull, and I. D. W. Samuel, “Triple-cation perovskite solar cells for visible light communications,” Photonics Res. 8(8), A16–A24 (2020). [CrossRef]
14. A. M. Abdelhady, O. Amin, B. Shihada, and M.-S. Alouini, “Spectral efficiency and energy harvesting in multi-cell SLIPT systems,” IEEE Trans. Wireless Commun. 19(5), 3304–3318 (2020). [CrossRef]
15. J. I. d. O. Filho, A. Trichili, B. S. Ooi, M.-S. Alouini, and K. N. Salama, “Toward self-powered internet of underwater things devices,” IEEE Commun. Mag. 58(1), 68–73 (2020). [CrossRef]
16. H. Haas, S. Videv, S. Das, J. Fakidis, and H. Stewart, “Solar cell receiver free-space optical for 5G backhaul,” in 2019 Optical Fiber Communications Conference and Exhibition (OFC), (2019), pp. 1–3.
17. S. Das, E. Poves, J. Fakidis, A. Sparks, S. Videv, and H. Haas, “Towards energy neutral wireless communications: Photovoltaic cells to connect remote areas,” Energies 12(19), 3772 (2019). [CrossRef]
18. N. Saeed, T. Y. Al-Naffouri, and M.-S. Alouini, “Around the world of IoT/climate monitoring using internet of X-things,” IEEE Internet Things M. 3(2), 82–83 (2020). [CrossRef]
19. “What is maximum power point tracking MPPT,” Last accessed 18 April 2021.
20. “Proximity/UV/ambient light sensor IC with I2C interface,” https://www.silabs.com/documents/public/data-sheets/Si1145-46-47.pdf. Last accessed: 7-12-2021.
21. N. Lorriére, N. Bétrancourt, M. Pasquinelli, G. Chabriel, J. Barrére, L. Escoubas, J. Wu, V. Bermudez, C. M. Ruiz, and J. Simon, “Photovoltaic solar cells for outdoor LiFi communications,” J. Lightwave Technol. 38(15), 3822–3831 (2020). [CrossRef]
22. A. Trichili, M. A. Cox, B. S. Ooi, and M.-S. Alouini, “Roadmap to free space optics,” J. Opt. Soc. Am. B 37(11), A184–A201 (2020). [CrossRef]
23. A. Vavoulas, H. G. Sandalidis, N. D. Chatzidiamantis, Z. Xu, and G. K. Karagiannidis, “A survey on ultraviolet C-band (UV-C) communications,” IEEE Commun. Surv. Tutorials 21(3), 2111–2133 (2019). [CrossRef]
24. O. Alkhazragi, A. Trichili, I. Ashry, T. K. Ng, M.-S. Alouini, and B. S. Ooi, “Wide-field-of-view optical detectors using fused fiber-optic tapers,” Opt. Lett. 46(8), 1916–1919 (2021). [CrossRef]
