Tom and I brought our trusty Anritsu MS2721B handheld spectrum analyser with us to Chicago (this is the same analyser that survived our 3000-plus miles spectrum-measurements roadtrip late last year). As part of the DySPAN demonstrations preparation, we wanted to take a look at spectrum usage in the vicinity of the Knickerbocker Hotel, the venue for this year’s conference, to help us plan our demonstrations strategy. The set-up here is quite straightforward; the analyser is connected to a laptop using an Ethernet cable and a simple magmount whip antenna is connected to the analyser. The analyser is controlled from the laptop using standard commands for programmable instruments (SCPI) issued by a Matlab script (we have a Python version of this script also). Traces are stored as individual files (as we’ve found that gives us some hope in case of a failure along the way) and can then be plotted in a variety of ways. The figures below are 3D representations of the bands we have been looking at for the past few days with a colorbar indicating the received power in dBm.

This is a view of the 400-500MHz band from inside the third floor of the Knickerbocker Hotel in Chicago:

At one point, we detected a very strong signal between 464MHz and 473MHz that briefly maxed out the analyser and resulted in messing up our colour-scaling for this plot.

Here’s a quick look at the 500-600MHz band. This is occupied by various analogue and digital TV signals. One particularly strong DTV signal, WYCC-DT (according to a channel guide), can be spotted on channel 21 (512-518MHz). This is being broadcast from the top of the Hancock building only a couple of blocks away from our hotel.

The ‘700MHz band’ (698-806MHz) looks like this from our location:

Keith and I spent the last couple of days in Chicago in preparation for this year’s DySPAN 2008. Our particular goals were to get to know the local crew at IIT, including one of the esteemed general co-chairs of the conference, Dennis Roberson. During our visit, we met some of his graduate students, and they showed us the spectrum collection system they have set up on top of the 22-story building, the tallest in that area of Chicago.

The top of the building houses both a great view of Chicago:

As well as a great view of the spectrum:

They’re doing a great job there and have been able to capture months of spectrum usage in Chicago. Because of the length of time that their measurement system has to offer, they can start seeing and analyzing interesting patterns, periodicities, and anomalies. We hope we can work with them to build a common format for storing and presenting both their data and our data in some way that that the rest of the community can then use and develop tools around. I hope a common format would enable other researchers to make use of the efforts we (and others) have put in to taking these kinds of measurements.

Our main purpose for being in Chicago was to plan for the demonstration at DySPAN 2008, which included checking out the Millennium Knickerbocker Hotel. With what Keith and I are lining up for the venue and the great proposals we have received for demonstrations, this year’s conference should easily keep everyone busy and excited.

In a recent conversation, the topic of where cognitive radio and software-defined radio are in terms of estimated positions on Gartner’s Hype Cycle arose. During the emergence of any new technology, five stages or categories of development can be observed: The technology trigger represents the breakthrough (e.g. Mitola’s work in the late 1990s kicking off the concept of cognitive radio). Following this is the peak of inflated expectations signifying the region of time dominated by “over enthusiasm and unrealistic projections…a flurry of publicized activity by technology leaders..the only enterprises making money at this stage are conference organizers and magazine publishers.”

The trough of disillusionment is the point at which the technology becomes unfashionable and abandoned by the press because the overinflated expectations are not being met. Solid hard work and experimentation by an increasingly diverse range of organizations helping to gain a real understanding of the technology’s applicability, risks and benefits brings us to the slope of enlightenment category. During the plateau of productivity, the real-world benefits of the technology are demonstrated, accepted and serious revenue-generating applications are finally out in the market place.

There is currently a large number of active and recent calls for papers for various conferences, journals and books relating to cognitive radio (e.g. CROWNCOM,CogCom,DySPAN,Milcom,Globecom, PIMRC,WCCI, SDRForum, WiCom, and VTC are just some of the conferences featuring cognitive radio-related technologies). Significant press has also been devoted to the topic by various print and online media bodies. However, there are no publically releasable and really groundbreaking applications that use cognitive radio technology yet other than incremental enhancements at the infrastructure level (this includes dynamic spectrum access techniques). Tom and I also got a variety of views from our travels late last year and various meetings so far this year which cover the range of opinions from being at peak visibility to being right down in the trough of overflated expectations that are not being met. Cognitive radio as a technology, is therefore somewhere between the peak and trough on the hype cycle.

Software-defined radio appears to be on the slope of enlightenment however (thankfully). VANU Inc. and their field trials, BitWave, PrismTech, PicoChip are some of the companies generating news with robust market offerings backing the claims up.

Solid research and experimentation work is required to help make greater technological leaps and to develop innovative cognitive radio and network-based applications (analagous to creating applications like FaceBook, YouTube, Ebay, etc., that make use of the network infrastructure instead of focusing all of the research effort on the infrastructure itself). Reducing the time spent in the upcoming trough by helping to push the experimental work was one of the main motivating factors for arranging the demonstrations at IEEE DySPAN 2007 in Dublin and is still one of the driving forces behind this years demonstrations in Chicago for DySPAN ‘08 where I’m co-chairing the demonstrations again with Tom.

Now that the FCC 700MHz band auction is finally over after 261 bidding rounds and $19.6 billion later in addition to the NTIA spectrum-sharing innovation testbed, trough-time may be shortened even further, in the US at least.

Google news alerts sometimes raises some interesting angles – perhaps this blogger knows something we don’t…