TV white space (TVWS) prototypes initially relied on spectrum sensing and dynamic spectrum access (DSA). The idea was that secondary users could share spectrum with primary users (TV stations and wireless microphones) and avoid interference by detecting and using only the available channels.

However, after several years of debates and trials, the FCC decided in 2009 to replace the RF sensing approach with geo-location white space databases (WSDBs). This eliminated the requirement for white space devices (WSDs) to detect TV and microphone signals. The WSDBs now maintain a list of frequencies available for use by the WSDs at different geographical locations.

Today, while TVWS commercial products that use WSDBs are being deployed, research continues on RF sensing technologies for spectrum sharing.

Limitations of the WSDB approach

Microsoft and the Chinese University of Hong Kong1 recently published some results on dynamic spectrum access for indoor applications that compared the WSDB approach with actual RF spectrum sensing. The system detected which frequencies were available for use at given locations based on “a limited number of sensors” to achieve “satisfactory performance at a reasonable cost”.

According to the report, “there is often more spectrum available for use than indicated by the geo-location database” and, “on average we found 40% more TV spectrum to be available indoors”.

Considering that the number of TVWS channels appears extremely limited in some cities when using the current WSDB approach, these results could be very significant to regulators in countries that are currently undergoing consultations for future TVWS regulations. As it stands today, the regulations do not have any special considerations for indoor scenarios.

Beyond the TVWS spectrum

The overwhelming popularity of mobile devices continues to push the capacity of the radio spectrum. As a result, spectrum sharing is being considered not only for the TV spectrum. For example, the Telecom Regulatory Authority of India (TRAI) recently proposed that the trading of spectrum in all bands and technologies should be allowed. According to The Times of India (Oct 14), the DoT is expected to give a go-ahead for operators to trade and share cellular spectrum.

Although the mechanism for sharing will likely be some licensed shared access (LSA) or authorised shared access (ASA) mechanism, it is interesting to consider a future landscape that empowers true dynamic spectrum sensing and cognitive radio (CR) technologies for optimal spectrum utilization.

Redefining spectrum efficiency

According to research published in 2010 at the ICST Conference on Cognitive Radio2, the sharing opportunities in the cellular spectrum are very significant. Radio spectrum measurements were conducted in France and in the Czech Republic between 400 MHz and 3 GHz. The results indicated that the level of utilization in the GSM 900 and 1800 bands would vary between 38% and 48%, depending on the regions.

Similar results were published in a 2013 spectrum survey conducted in UK3. The level of occupancy for the GSM 900 and 1800 bands would be about 33% and 24%, respectively. This leads us to conclude that beyond the usual bit/s/Hz spectrum efficiency, true spectrum efficiency would require some form of spectrum sensing and cognitive radio technology.

Promising results

To demonstrate the exploitation of unused spectrum in the cellular bands, Nokia and the Cognitive Radio Trial Environment (CORE) consortium in Finland recently conducted an ASA demonstration using 4G LTE handsets. According to the CORE lab, Nokia’s TD-LTE test network, which switched frequencies dynamically based an ASA repository, resulted in a system capacity increase of 18%4.

Considering that RF and baseband systems-on-a-chip (SoCs) are becoming more advanced and more affordable every year, it is conceivable that, in the near future, the technology will be available to enable the wise use of white spaces across the entire spectrum. For example, the current software-defined radio (SDR) platform from Nutaq is capable of covering the spectrum from 300 MHz to 3.8 GHz. Its massive digital signal processing power enables it to implement any of the 2G/3G/4G cellular standards, which makes it an ideal platform for achieving further progress in white space technologies.

References

1. “CUHK Faculty of Engineering and Microsoft Research of Hong Kong Jointly Develop WISER to Optimize Indoor Wireless Connectivity Using TV White Spaces” Communications and Public Relations Office, The Chinese University of Hong Kong, 30 September, 2013.; “Got bandwidth? Microsoft Research and university researchers team up to find white spaces”, Microsoft News Center, 30 Sept 2013; “Exploring Indoor White Spaces in Metropolises”, MobiCom 2013, 30 September 2013.

2. « Survey on Spectrum Utilization in Europe: Measurements, Analyses and Observations », Václav Valenta, Roman Maršálek, Geneviève Baudoin, Martine Villegas, Martha Suarez, Fabien Robert, 5th International ICST Conference on Cognitive Radio Oriented Wireless Networks and Communications, France, June 2010.

3. Spectrum Occupancy Survey In HULL-UK For Cognitive Radio Applications: Measurement & Analysis, Meftah Mehdawi, N. Riley, K. Paulson, A. Fanan, M. Ammar, INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME 2, ISSUE 4, APRIL 2013.

4. Nokia demonstrates: White Space ploy can get more 4G handsets into same spectrum, Bill Ray, Sept 30, 2013, http://www.theregister.co.uk/2013/09/30/nokia_demonstrates_lte_can_share_nicely/
http://core.willab.fi/sites/default/files/ASA_concept_press_release_long.pdf