Reduced Complexity Cognitive Radio Systems

Researchers are proposing a variety of increasingly complex methods of implementing cognitive radio, which incorporate software defined radio, dynamic spectrum management, and intelligence. The drawback of this complexity is that it is predicted that CR is still years away from implementation. The purpose of this project is to initially take a top level examination of the solutions proposed, to try and understand whether such complexity is justified, and what benefits they bring to overcome the current regulatory constrained spectral assignment process.

It is foreseen that it should be possible to develop reduced complexity strategies that will deliver much of the functionality of proposed systems by better exploiting distributed artificial intelligence. Such reduced complexity CR will enable more rapid adoption and wider use in systems where CR is currently not being considered due to prohibitive complexity.

For example, even limited intelligence, coupled with the ability to dynamically assign spectrum and respond to changing spectrum use, is likely to be sufficient for many implementations, removing the need for a full software defined radio. Such simplified devices are still likely to give rise to complex behaviour, which will be characterised through simulation and analysis. The application of distributed artificial intelligence strategies such as reinforcement learning will be investigated.

Set theory and Markov analysis will be particularly important analytical tools. It is expected that proposed solutions will be applied to communications architectures incorporating terrestrial ad hoc and high altitude platform nodes.

The project will take note of existing standardisation activities underway, such as IEEE 802.22, IEEE802.16h and IEEE SSC41. This work will integrate closely with other activities within the Group.

Key objectives

To understand the benefits/drawbacks of the different cross layer components of cognitive radio, including software defined radio, dynamic spectrum management and node intelligence
To simulate/analyse low complexity CR solutions with representative communications architectures, and compare with existing solutions
To understand the impact on standardisation activities underway that incorporate CR (IEEE 802.22, IEEE 802.16h, IEEE SSC41)

Outputs

Low complexity ‘intelligent’ CR strategies
Review of benefits/drawbacks of different components of CR
Simulations and analysis showing likely performance with different communications architectures, and impact on other (including primary) users
Contributions to conference and journal papers.

Reduced Complexity Cognitive Radio Systems

Key objectives

Outputs

Members

Dates

Research