MRT has an article up on cognitive radio and public safety. It touches on a number of different developments over the last year which should make cognitive radio and frequency agility in general easier to achieve – Bitwave’s flexible RFIC, Harris’s multiband public safety radio, the Shared Spectrum / Harris partnership.
It also mentions a $500 cognitive radio being developed by M/A COM and Shared Spectrum. Giving a little more context, I’m virtually certain that refers to the DARPA WNAN/WANN/WAND program where the networking part (WAND) is led by BBN and the hardware node is being designed by M/A COM. As part of the program there’s a mandate to include Shared Spectrum’s xG technology in the deployed solution.
Washington Technology announced that Shared Spectrum’s xG technology (DSA) is going to be ported to the Harris Falcon III radio for proof-of-concept and feasibility analysis as part of an extension of the DARPA xG program.
That’s interesting because I thought that was already going on as I had heard some second-hand reports of initial performance testing results from a couple different sources. So maybe it’s progressed far enough now that they’re wanting to advertise a little.
Because a) I sounded like I knew what I was talking about and perhaps more importantly b) I had to get off the teleconference early so I couldn’t say no, I got assigned to revise the SDR Forum Cognitive Radio Working Group’s definition of cognitive radio and the surrounding discussion in a nomenclature document which should come out sometime soon.
Thinking this might be of interest to readers of this blog, I’ve reproduced my contribution in the following.
Cognitive radio refers to both a device and to an engineering paradigm for designing wireless systems. Because cognition is normally associated with human thought processes, the cognitive radio community has adopted several terms from human psychology to describe cognitive radio whose meaning is unclear in an engineering setting. To resolve this, the following also defines these related terms in a manner applicable to wireless engineering.
Cognitive Radio (design paradigm)
An approach to wireless engineering wherein the radio, radio network, or wireless system is endowed with cognition and agency to intelligently adapt operational aspects of the radio, radio network, or wireless system.
The capacity to perceive, retain, and reason about information.
Typical types of information used in a cognitive radio include location, environmental information, and internal states.
The capacity to make and implement choices.
Exhibiting behavior consistent with a purposeful goal.
While a system could be cognitive without exhibiting agency (e.g., a brain in a jar), or could have cognition and agency without intelligence (e.g., a person who makes all of his/her choices by a flip of a coin), all three aspects are critical to the cognitive radio design paradigm.
The process of acquiring, classifying, and organizing information.
Note that there are many different potential sources from which and kinds of information that may be acquired. Some sources may be internal (e.g., a measurement of an amplifier bias current); some may be external (e.g., information from a networked database); some information may be about itself (e.g., the radio’s own location); and some information may be about other radios (e.g., the interference experienced by another radio).
The application of logic and analysis to information.
Using these definitions, the cognitive radio design paradigm can be equivalently defined as follows.
Cognitive Radio (design paradigm)
An approach to wireless engineering wherein the radio, radio network, or wireless system is endowed with the capacities to:
- acquire, classify, and organize information (cognitive – perceive)
- retain information (cognitive – retain)
- apply logic and analysis to information (cognitive – reason)
- make and implement choices (agency) about operational aspects of the radio, network, or wireless system in a manner consistent with a purposeful goal operational aspects of the radio, network, or wireless system (intelligent).
Because there are far too many ways that the cognitive radio paradigm can be applied to list all possible implementations, the following only defines the three classes of implementations most commonly discussed at the time this document was created.
Cognitive radio (device)
- A radio designed according to the cognitive radio engineering paradigm.
- Cognitive radio as defined in (2) that utilizes Software Defined Radio, Adaptive Radio, and other technologies
- A radio, radio network, or wireless system designed according to the cognitive radio engineering paradigm.
- A radio endowed with the capacities: to acquire, classify, retain, and organize information, to apply logic and analysis to information, and to make and implement choices about operational aspects of the radio in a manner consistent with a purposeful goal.
Note that cognitive radio does not explicitly refer to a specific realization of a radio. A mobile could be a cognitive radio; a base station could be a cognitive radio; a mesh node could be a cognitive radio; etc..
- A network designed according to the cognitive radio engineering paradigm.
- A network endowed with the capacities: to acquire, classify, retain, and organize information, to apply logic and analysis to information, and to make and implement choices about operational aspects of the network in a manner consistent with a purposeful goal.
Example: An enterprise WiFi network wherein “thin” access points take in sensing information which is then passed to a networked controller; which then assigns channels to the access points.
Note that the centralization of the capacities for cognition and agency is not critical to the concept of a cognitive network. Instead these capacities could be implemented as distributed processes.
Cognitive radio network: A network of cognitive radios
Example: An enterprise WiFi network wherein each access point is a cognitive radio.
It is important to note that having a capacity does not imply that the capability is always used. For instance a mobile cognitive radio might have its operation directed by a network at some times and be self-directed at other times. In all cases, the mobile cognitive radio remains a cognitive radio because it has the requisite capacities even when not actively exercised.
ETSI announced that they’re starting a technical committee to examine standardization and development of SDR and cognitive radio. The first meeting of the committee will be March 19-20, 2008 at ETSI Headquarters, Sophia Antipolis in France.
Yesterday, NICT (National Institute of Information and Communications Technology – Japan) issued a press release announcing they had put together a cognitive radio capable of automatically selecting to connect with networks supporting various flavors of WiFi and WCDMA. Waveform processing is performed on an unspecified FPGA which can be reconfigured in 200 ms; network detection takes 1 second; the RF chain goes between 400 MHz and 6 GHz with 6 different center frequencies.
- Network selection isn’t new (Intel holds some key patents on this)
- This is really more in line with the UMA effort than with cogntiive radio (not that they need to be mutually exclusive).
- It’s nice to see (quasi) commercial interest. Most CR work is defense focused.
- You have to expect that the simplest applications will be tackled first
- presumably the platform can be extended.
- FPGAs are likely not an immediately fieldable solution for handsets (which this seems targeted towards) due to power consumption considerations. However, both Xilinx and Altera are working to drive down power consumption, so this might be a viable platform in a few years.
The NTIA formally issued a Notice of Solicitation of Participation in today’s Federal Register and request for comments on implementation of a Spectrum Sharing Innovation Test-Bed and looking for expressions of interest in participating. Responses are due by Feb 29; the following are key excerpts from the solicitation.
Test-Bed Frequency Band: 410-420 MHz.
Authorization of Test-Bed Operations: FCC Part 5 Experimental Radio Service Rules.
Limitations on Test-Bed Operations: Frequency and/or geographic limitations may be identified as necessary
Protection of Incumbent Spectrum Users: To address potential interference to incumbent spectrum users the Test-Bed employing DSA equipment will be performed in three phases:
Phase 1 - Equipment Characterization. Equipment employing DSA techniques will be sent to the NTIA Institute for Telecommunication Sciences in Boulder, Colorado and characterization measurements of the DSA capabilities in response to simulated environmental signals will be performed.
Phase 2 - Evaluation of Capabilities. After successful completion of Phase 1, the DSA capabilities of the equipment in the geographic area of the Test-Bed will be evaluated.
Phase 3 - Field Operation Evaluation. After successful completion of Phase 2, the DSA equipment will be permitted to transmit in an actual radio frequency signal environment. An automatic signal logging capability will be used during the operation of the Test-Bed to help resolve interference events if they occur. A point-of-contact will also be established to stop Test-Bed operations if interference is reported.
Planning and Evaluation of Test-Bed: A flexible peer review process open to the public will be employed.\13\ Federal and non-federal users will have an opportunity to participate in the development of test plans, review status reports, and review the final report on the results of the Test-Bed.