Wednesday, 12 October 2016
LiFi or Visible Light Communications (VLC) evolves
Even as we increase data rates and network capacity in Wi-Fi with 802.11ac, 11ad and then 11ax, it’s worth keeping an eye on other communication technologies that may become useful in the enterprise. Visible light communication (VLC) is making progress in the background, and while it is barely out of the lab, it may take off in a few years. The startups working with enterprise VLC are adopting the term “LiFi” (although it has—today—no connection with Wi-Fi). ￼ Like all new technologies, there are many possibilities—and the number of potential uses is impressive. Applications include vehicle-to-vehicle, underwater communications, streetlamps, aircraft cabins, and industrial uses where Wi-Fi cannot be used due to RF interference. Attributes of LiFi The most attractive attribute of LiFi is its bandwidth. The visible-light portion of the spectrum spans around 300,000 GHz—six orders of magnitude more than we now use for Wi-Fi in the 2.4 and 5 GHz bands—so the data rates and capacity that can eventually be obtained are almost unimaginable. Most light communication to date has used lasers and fiber-optics. It’s very effective, it's but expensive—although work on Silicon Photonics promises to bring down the cost and perhaps spur another wave of adoption in the data center, displacing copper. But VLC in the enterprise is based on LEDs, not lasers, modulating lightbulbs to communicate with traditional clients such as laptop PCs. LiFi products The most visible proponent is pureLiFi, a startup associated with the University of Edinburgh. Their second-generation product set comprises a ceiling-mounted unit, like a Wi-Fi access point, and a matching PC dongle. The ceiling unit is Ethernet-connected to the enterprise LAN, and its light shines on the photo-diode sensor of a dongle on the PC below it. It also contains a photo-diode to receive the uplink connection from the dongle’s LED. Claimed rates are only 40 Mbps full-duplex over a few meters, with inter-access point handover, so we are clearly at the beginning of the learning curve compared to the theoretically achievable performance. RELATED 802.11ad is the fastest Wi-Fi that you might not ever use Is Wi-Fi finally ‘fast enough?’ Scientists developing solar panel that doubles as a Li-Fi receiver on IDG Answers What does Microsoft Hololens Enterprise Edition offer business that other... This is an interesting start, given the early state of the technology, which is still pre-standard. (In the standards world, the term “VLC” is already claimed by IEEE 802.15.7, and there are moves afoot to start work on LiFi in both 802.15 and the IEEE 802.11 group. LiFi could develop a common MAC with Wi-Fi, which would surely broaden its interest: consider a device that could seamlessly move a connection between 5GHz - 802.11ac, 60GHz - 802.11ad and light.) Nevertheless, it is hard to believe that replicating Wi-Fi topology (ceiling-mounted access points communicating with desk-level mobile devices and PCs) is the best application of LiFi in the enterprise. Surely it would be better to find situations Wi-Fi cannot satisfy and hope to build from those niches to general applications. Proponents have certainly made an effort to identify where differences between Wi-Fi and light might be usefully exploited. Whereas Wi-Fi signals are notoriously difficult to arrest—much of our focus in WLAN design goes to limiting interference from cell to cell—walls block light quite effectively. Also, light communication offers a solution for situations where the Wi-Fi spectrum is congested or suffers interference. But can these differences be exploited for a commercially successful solution? Also, the LiFi access points will need clients. Dongles are clearly a limited solution, but curiously, many of the devices we see in the enterprise—phones and tablets—already have light sensors (the cameras) and transmitters (flashlights). If the installed base of phones and tablets can be made to support LiFi, the barriers to adoption become dramatically lower. Early-stage technology is so difficult to evaluate. To the visionaries, eventual benefits are obvious, but can the industry follow the classic path to commercialization where it exploits vigorous niche opportunities before moving into broader applications? And how long will it take? The answer to the first is that while the possibilities are intriguing, the product stepping stones have not yet emerged. And the second is always “longer than you think.” There’s plenty of time to see how this develops.