The moniker's similarity to its rf equivalent, Wi-Fi, is no accident: proponents of Li-Fi are suggesting it could become as ubiquitous as IEEE802.11. But, to be fair, even the most ardent evangelists accept that it will take quite a while for the technology to mature.
Professor Harald Haas, chair of mobile communications at the University of Edinburgh, told New Electronics: "We should consider Li-Fi as complementary to today's wireless networking technology, though with a whole range of advantages, not least in terms of energy efficiency, security and the fact that it unlocks a vast range of unused and, importantly, unregulated electromagnetic spectrum.
"The aim is fully fledged networks, where a Wi-Fi network would offload data to Li-Fi when applicable, easing the spectrum crunch we are facing, much like an LTE based cellular network offloads networking to Wi-Fi."
Along the way, the technology is more likely to be exploited in sector specific applications, notably where rf links are not desirable or possible. Amongst those being touted include deploying specially equipped LED lighting in aircraft cabins that would allow passengers to connect to laptops, mobile phones and tablets in flight; in hospitals, where rf signals are often prohibited; data exchange between smartphones (Casio has already demonstrated an application for this using prototype mobiles where signals are transmitted using by varying the light intensity of the screens ); and street lighting communications (French company Oledcomm has partnered with Thorn Lighting to demonstrate this).
Prof Haas said he is talking to the oil and gas industry about using point to point Li-Fi to monitor conditions inside wells. At the moment, companies have to shut down their wells intermittently to install wire probes at huge expense every time.
For now, Prof Haas and his team, who have been working on the underlying technologies involved for nearly a decade and described the first 'proof of concept' in 2006, are located in the Alexander Graham Bell Building, named after the Scottish inventor and Edinburgh alumnus who devised the Photophone, the first device to transmit voice using modulated light. Bell directed sunlight at a parabolic mirror that captured and projected his vocal vibrations. Unfortunately, the unpredictability of sunlight scuppered the idea.
Prof Haas first demonstrated – and coined the term – Li-Fi at a widely acclaimed talk organised by TEDGlobal in July 2011. Posted a week later on YouTube, the clip has to date achieved 1.5million hits. "After the talk, the technology attained a much higher profile and was soon followed by the establishment of the Li-Fi Consortium."
This focuses on commercial applications of the technology. Prof Haas himself cofounded and acts as chief scientific officer of Edinburgh based pureLiFi, a spin out of work at the University which is planning to partner with light bulb and light fixture manufacturers to develop and commercialise Li-Fi applications.
Li-Fi works as a signal transmitter with high bandwidth white LEDS, typically 200MHz, and as a signal receiver with either a PIN or an avalanche photodiode. This means Li-Fi systems can not only illuminate a room, but also provide wireless data connectivity. This is accomplished by modulating the incoherent light generated by LEDs using a modified version of orthogonal frequency division multiplexing. Prof Haas has dubbed this Spatial Modulation OFDM. "With this technique, developed at Edinburgh, we can exploit all four dimensions: colour, time, frequency and space," he noted.
The group has demonstrated data transfer at up to 1.6Gbit/s on a single colour LED and expects to achieve 2Gbit/s on each of the RGB channels this year, adding up to an impressive 6Gbit/s.
Prof Haas maintains the first part – tinkering with the electronics of LEDs so they generate the flickering signals for data transmission – is the easy bit, since they are already semiconductor devices. But while streaming real time video from a white LED at fairly high data rates over a single channel point to point link, as demonstrated by Prof Haas in his TEDGlobal talk, is one thing, the networking aspect of the technology – including multiuser access, interference coordination and overcoming the line of sight conundrum – is another.
In fact, Prof Haas suggests the latter, often seen as Li-Fi's Achilles Heel, has been overcome. Researchers within his group and pureLiFi, working in collaboration with one of the group's beta project partners, has shown that Li-Fi can operate by using incident light (which includes reflections) and does not necessarily require line of sight connection between receiver and transmitter. Completed late in 2013, the project demonstrated high speed Li-Fi from reflections, streaming four videos in parallel.
Prof Haas said the group has concluded similar beta project partnerships 'with global players in the healthcare, aircraft, data and industrial communications sectors, from which we gain valuable feedback on all aspects of Li-Fi development'.
For full duplex communication, an uplink will clearly be required and Prof Haas suggests the most suitable technique is wavelength division multiplexing, where the two communication channels are established over different wavelengths. IR transmission is seen as perhaps the most viable option to link mobile terminals to the optical access point (AP).
Here, VLC borrows from the small cell concept widely used in rf networks and these APs are dubbed attocells, considered to be analogous to femtocells. They not only improve indoor coverage and bandwidth reuse, but, importantly, will not interfere with macrocellular networks.
In the long term, the goal is to enable seamless interoperation between optical attocells and rf femtocell/macrocells to ensure maximum spectrum relief for rf systems.
Prof Haas said the commercial proposition for pureLiFi is a series of OEM products, including miniaturised modems and dongles. The first product will be available on limited release as from this month. Offering full duplex communication with a capacity of 5Mbit/s in both the uplink and downlink path over a range of up to 3m, the device supports data rate densities of 1Mbit/s/m2 while providing ample desk space illumination.
This will be followed by the Li-Fire platform that will allow pureLiFi partners to develop applications. On the horizon is the Ceiling Unit that will connect to the data network via standard Ethernet RJ45 port. The units receive and decode the uplink signal using infrared detectors and optics. The protocol stack on the ceiling units enables seamless handover between APs, along with multiple access at each individual AP, creating an indoor atto-cellular network.
The connection is important as the two methods of bringing data into the LEDs and fittings are either power line communication, as a retrofit, or standardised Power over Ethernet. Once connected to the IP network, the lighting infrastructure can communicate with any other IP device or VLC enabled device, indoors or outdoors.
Further down the road is the Li-Fire Desktop Unit. This incorporates a processing unit with a visible light decoder to capture the continuous sequence of light intensity changes, decode the binary stream and transmit it to the client device via a USB connection. The desktop unit receives data from the client device, encodes it and transmits it to the ceiling unit using an IR emitter. The protocol stack on the desktop unit allows the user to move from one AP to the next without losing data connection.
The Edinburgh group is also involved in a related four year EPSRC funded project. Called UP-VLC – Ultra-Parallel Visible Light Communications – the project will develop and implement a Li-Fi network using micron sized GaN based LEDs that are said to be able to 'flicker' some 1000 times faster than current devices. The collaborators also suggest each micron sized LED could act as a tiny pixel and, when combined into an array, could display information whilst simultaneously providing a Li-Fi link.
The project, led by scientists at the University of Strathclyde, includes groups at the universities of Cambridge, Oxford and St Andrews, as well as industrial partners such as STMicroelectronics, Thorn Lighting, Osram, Avago Technologies, EV Group, Compound Semiconductor Technology and BAE Systems. In recent tests, each RGB channel sent data – admittedly over very short distances – at 3.5Gbit/s, a total of 10.5Gbit/s.
Commenting on the 'energy saving parallelism' of the breakthrough, project leader Professor Martin Dawson of Strathclyde said: "Imagine an LED array beside a motorway, helping to light the road, displaying the latest traffic updates and transmitting internet information wirelessly to passengers' laptops, netbooks and smartphones."
While stressing the UK's significant expertise in this field, Prof Haas acknowledges research groups in the US, China, Japan and, notably, Germany, as well as many potential competitors to the kind of designs and platforms coming out of pureLiFi.
He is particularly impressed by work being done at the Fraunhofer Heinrich Hertz Institute (HHI) for Photonics Systems in Berlin and suggests there is healthy competition between the groups. "Even though they are, perhaps, approaching the opportunity from the LED and optics angle, while our focus has been more on the data modulation and networking aspect."
Anagnostis Paraskevopoulos, a senior researcher at HHI, concurs partially and suggests the biggest difference between the two groups is that HHI focuses on developing products and devices that can be integrated speedily. "For instance, plug-and-play modules, and on Ethernet LAN type devices that can be easily combined with existing networks."
This is, of course, partly because the applied research institute works predominantly with industrial partners looking to commercialise applications; in this instance, in sectors such as automotive and lighting. Like Prof Haas, Paraskevopoulos believes the initial focus should be on industrial applications and suggests interference prone conference rooms, fair trade booths and hospitals would be ideal places to showcase the technology.
HHI recently demonstrated a bidirectional, real time line of sight VLC system (see fig 1). Currently working in half duplex mode, the system relies on rate adaptive OFDM and is implemented with feedback via the reverse link. The transceivers, equipped with proprietary VLC transmitter and receiver modules, offer a bandwidth of up to 180MHz and links to 1000BaseT interfaces.
At a typical working distance of 2m between the ceiling unit and table top, and in a circle of about 60cm in diameter, the system allows a data rate of 200Mbit/s per user. Second generation prototypes, with smaller form factors, are on the way.
Paraskevopoulos also told New Electronics that HHI prefers to pursue the vision of 'optical Wi-Fi'. "We, and the scientific community, have yet to accept the term Li-Fi completely. But we certainly share most of the underlying technologies and targets as outlined by Prof Haas' group."
On one aspect, the researchers are in total agreement: the technology needs to be standardised before it can really take off, from the front end 'luminaires' to the multipoint to multipoint functionality. However, with bodies such as the IEEE 802.15.7 and the ITU's g.hn home networking group likely to be involved, it may be a while before this bright idea is ready to shine.
