TPRC has ended
Back To Schedule
Sunday, September 29 • 9:00am - 9:35am
A Quantitative Analysis on the Impact of Spectrum Access Policy Flexibility and DSA System Capability on Spectrum Sharing Potential

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Download Paper

While Dynamic Spectrum Access (DSA) is premised upon the existence of technologies and policies that enable flexible access to spectrum, a quantitative understanding of the coupling among DSA technologies, policy definition, and spectrum sharing potential remains largely unexplored. Over ten years of technology research, development, and prototype testing has resulted in algorithms and system concept advances, including the ability to deduce policy constraints (and permissions) in situ. Knowing the policy, however, is different than selecting operating parameters (e.g., transmission power) that lead to policy compliance (e.g., interference mitigation). The primary hurdle to in situ compliance is situational awareness uncertainty, which results from the inherently stochastic nature of wireless communications and spectrum user activity.

Situational uncertainty is currently mitigated by policy specifications that impose universal operating constraints and limit efficient spectrum sharing. For example, TV Band Device policies were derived in a manner similar to policies for non-DSA systems. Specifically, numerous models of hypothetical scenarios ? including plausible ?worst-case? conditions ? were developed and analyzed. Judgments were made as to the mechanisms (e.g., sensing, GPS-aided geolocation) to be used for situational awareness as well as their corresponding capabilities (e.g., detector sensitivity, location accuracy). Those factors were then specified in regulatory policy as universal constraints on device design, awareness mechanisms, and performance limitations. The result is limited spectrum sharing ability in the majority of operating conditions in an effort to ensure risk mitigation of significant impacts associated with rare situations.

This paper asserts and tests the hypothesis that improved spectrum sharing is possible by incorporating in situ uncertainty evaluations in DSA situational awareness and developing flexible policy specifications that regulate DSA behavior as a function of awareness. The assertions are based upon preliminary results of ongoing research into the relationship between DSA situational awareness uncertainty and potential DSA performance as measured by capacity gain, interference mitigation, monetary cost, and other metrics. A DSA situational awareness and decision model is developed using probabilistic Structural Causal Modeling (SCM) and multiattribute decision theory. The probabilistic SCM provides a formal system for representing cause and effect of DSA system operations and is built upon well-established engineering models for wireless communications. The multiattribute decision model provides a formal basis for in situ DSA decision-making that combines user goals and situational uncertainty. The model can also be used by engineers and policy-makers for evaluating the impact of policies, DSA awareness mechanisms, and user preferences on spectrum sharing and DSA system performance.

The paper provides an overview of the model but focuses on quantitative evaluations of spectrum sharing and DSA system performance as a function of policy flexibility and DSA system capabilities. Characteristics of the various model components and their significance are presented through a discussion of model development. The impact of spectrum policy on spectrum sharing and DSA system performance are quantified in the use cases in terms of capacity and interference.

Preliminary results collected to date indicate a clear gain in spectrum sharing as a function of increased policy flexibility and increased DSA system situational awareness. Policies that define specific DSA mechanisms or design constraints provide the least amount of policy flexibility and spectrum sharing potential; they similarly limit the design options available to system developers. Policies that define the desired effect (e.g., capacity goal) or impact limitations (e.g., maximum interference temperature) afford the greatest flexibility to a DSA system and thus the greatest potential for spectrum sharing. A DSA system?s ability to use the spectrum sharing potential is then a function of the specific situation and the DSA system?s degree of situational awareness uncertainty. The final paper will detail these findings as well as findings from ongoing research in the coming months.


John Chapin

Program Manager, DARPA

avatar for Todd Martin

Todd Martin

George Mason University

Sunday September 29, 2013 9:00am - 9:35am PDT
GMUSL Room 329

Attendees (0)