The momentum around dynamic spectrum usage continues to gather, and in most cases, it is initially targeted at the white spaces in TV spectrum. This week alone, two of the most active proponents of mobile services in these frequencies ‘ UK regulator Ofcom and Microsoft ‘ announced new trials, and the latter claimed a new, efficient approach to finding vacant channels.
Ofcom, as part of a broader outlining of its ambitious spectrum plan this week, revealed the companies which will take part in the pilot it first announced in April. Over the next six months, a score of public and private organizations will take part, testing various applications and use cases.
Many of these will come within the two main categories that have attracted attention to the white spaces, but which could come into conflict with one another in terms of access to frequencies. These are connectivity for the internet of things and broadband access for underserved, particularly rural areas.
For the first, the spaces would provide a low cost and brand new resource to support the expected mushrooming of machine-to-machine applications which require broadband-class connections, such as smart meters or advanced road traffic management. For the second, the low frequency spectrum offers cost effective coverage of rural areas because of its range and unlicensed nature.
Among the participants will be fixed line incumbent BT, which has already conducted white space tests of rural broadband; ISP Click4internet, also focused on rural access; Neul2, which has created a platform and would-be standard, especially geared to M2M; and the Department for Transport, which is interested in transmitting data on traffic conditions and is running a test of this on the road from the eastern port town of Felixstowe to the hub of UK white space activity, Cambridge.
Microsoft is also prominent, as it has been in the UK’s white space efforts ever since Ofcom became the second major regulator after the US FCC to open up the licence-exempt frequencies for mobile broadband. Its latest activity will be a pilot of free Wi-Fi access, but not for underserved rural areas but in the Scottish city of Glasgow, which is said to have the lowest rate of broadband uptake of any UK city.
Microsoft has also described new white spaces projects in China, where it has been working with university researchers on techniques to make it easier to discover vacant channels (important, since mobile signals must avoid interference with incumbent users like broadcasters and wireless microphones). The technology, developed by Microsoft Research and the Chinese University of Hong Kong, concentrates on identifying available white space channels indoors. It is dubbed WISER (White-space Indoor Spectrum EnhanceR) and its creators claim it can identify 30% to 50% more white space spectrum than alternative methods, most of which have been designed for outdoor use. A six-month analysis of white space spectrum in Hong Kong locations by Microsoft and its partners found that there is about 40% more of this capacity available indoors than outdoors.
WISER relies on a cluster of RF sensors, which sample the airwaves in a building to find vacant spectrum. The data, and the locations of wireless access points and self-reporting clients, is stored in a geolocation database. The WISER algorithms are used to profile the building and find optimal positions for sensors. In the prototype, the sensors use computer hosted software radios based on the USRP (Universal Software Radio Peripheral) standard, but Microsoft says they could have been ‘any spectrum analyzer with a low noise floor’.
The technology’s most important contribution to the emerging white spaces sector, apart from its indoor functionality, is to avoid the need for the wireless devices to do spectrum sensing themselves. The device just needs to report its position to the geolocation database using any indoor technique such as Wi-Fi or Bluetooth, and the database responds with the channels available for the client to connect to the internet. Microsoft sees WISER being embedded in Wi-Fi access points or in sensors which a building owner could plug into wall outlets, backed by powerline or Wi-Fi.
The FCC allows two methods for detecting free channels: spectrum sensing, by the device radio, or querying a web-based geolocation database. According to the WISER research paper, the latter is more commonly used because spectrum sensing is costly, and at low thresholds, it can be inaccurate with off-the-shelf hardware. However, the team claims the geolocation method has ‘inherent inefficiency’ because is uses propagation modeling rather than direct measurements to identify available spectrum ‘and hence, is very conservative in the channels it returns for a given location’. Therefore, the goal of WISER is to reduce the cost of the sensor approach and improve its accuracy.
“The geolocation approach has been studied extensively. While it does work, it has severe limitations,” said Minghua Chen, a CUHK professor in the department of information engineering. “As an alternative, we have developed a new system, WISER, that uses sensing technology in a more efficient and cost effective way than ever before.”