Skip to content

NSF NeTS grant 1618450

Spectrum Sharing Systems for Wireless Networks: Performance and privacy challenges

The rapid increase in the quantity and capability of consumer mobile wireless devices has accelerated the growth in demand for radio frequency spectrum. With limited opportunities to open up new unencumbered bands for spectrum hungry mobile wireless services, regulators are turning to spectrum sharing among increasingly broad system types as their only recourse. Inherent challenges and complexities involved with implementing practical spectrum sharing schemes are exemplified by two recent examples in the United States. First, in January 2015, the US Federal Communications Commission (FCC) auctioned the band 1695-1710 MHz, making it available for use by cellular systems which will be required to share the spectrum with incumbent meteorological satellite (METSAT) services already in the band. Cellular operators will need to demonstrate that they can preclude harmful interference to fixed METSAT earth station locations. According to another recent FCC ruling, the 3550-3700 MHz band will be opened up to new spectrum uses through advanced shared spectrum access systems. The proposed systems are intended to protect a tier of incumbent, mostly government, primary users from harmful interference while dynamically assigning spectrum resources to lower tier, or secondary user, networks. Incumbent users, e.g. the Department of Defense operating radars in this band, have raised concerns about maintaining the privacy of their operations with these systems.

Project Goals:

Both examples above broach the general problem of how to design effective and practical spectrum sharing systems. This proposal explores a general framework for architecting such systems and examining design trades rigorously. The design goals of a spectrum sharing system will fall into three categories, namely, performance, privacy and complexity. The proposal will conduct formal optimization analysis to study the allocation of resources (time, frequency and power) among the various networks in terms of the resulting utility of the spectrum for those networks to maximize performance subject to interference constraints. Since the operational information about wireless systems needed to effect spectrum sharing, such as locations, frequencies, time of use, and susceptibility to interference, may be considered sensitive, this proposal will answer whether, under what conditions, and how a spectrum sharing system can be designed to protect a critical level of privacy for the systems it enables. To do so, the proposal will consider the scope of operational scenarios, adversary exploitation techniques, and obfuscation strategies to protect privacy, developing relevant metrics for the analysis.

Personnel:

Konstantinos Psounis, Professor
Matt Clark, PhD (now at Aerospace Corporation)
Weng Ao, PhD (now at Qualcomm)
Po-Han Huang, PhD student
Lillian Clark, PhD student

Activities:

Maximizing the performance of the spectrum sharing system subject to a privacy constraint, meaning subject to satisfying a privacy requirement of the incumbent/primary user, e.g. the military.
Spectrum sharing in the presence of a sensing infrastructure. This is motivated by the request of incumbent users like the military who prefer not to share any information with the spectrum sharing system. Instead, the system would rely on passive sensing.
Spectrum sharing system problem in the context of radars, where the physical layer of the radar has been considered to optimize the ability to protect the radar privacy.
Optimize/maximize the privacy of the incumbent user subject to a performance constraint for the secondary user. This is particularly interesting because traditionally literature attempts to maximize the performance of the secondary user subject to a privacy constraint, whereas from the point of view of the primary/incumbent user, the goal might be to maximize its privacy subject to a performance constraint (minimum performance achieved) by the secondary user.

Outcomes:

  1. On Radar Privacy in Shared Spectrum Scenarios, A. Dimas, M. Clark, B. Li, K. Psounis and A. Petropulu, in Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP)}, Brighton, UK, May 2019.
  2. Privacy utility trades in wireless data via optimization and learning, L. Clark, M. Clark, K. Psounis, P. Kairouz, in Proceedings of Information Theory and Applications Workshop (ITA), San Diego, California, USA, February 2019.
  3. Trading Utility for Privacy in Shared Spectrum Access Systems, M. Clark and K. Psounis, IEEE/ACM Transactions on Networking, Vol. 26, Issue 1, February 2018.
  4. Equal Interference Power Allocation for Efficient Shared Spectrum Resource Scheduling, Matthew Clark and Konstantinos Psounis, IEEE Transactions on Wireless Communications, Vol. 16, Issue 1, January 2017.
  5. Achievable Privacy-Performance Tradeoffs for Spectrum Sharing with a Sensing Infrastructure, M. Clark and K. Psounis, in Proceedings of the 14th Annual Conference on Wireless On-Demand Network Systems and Services (IFIP WONS), Isola, France, February 2018.
  6. Spectrum Sharing Between Radar and Communication systems: Can the Privacy of the Radar be Preserved?, A. Dimas, B. Li, M. Clark, K. Psounis, A. Petropulu, in Proceedings of the Asilomar Conference on Signals, Systems and Computers, Pacific Grove, California, USA, October 2017.
  7. Designing Sensor Networks to Protect Primary Users in Spectrum Access Systems, M. Clark and K. Psounis, in Proceedings of the 13th Annual Conference on Wireless On-Demand Network Systems and Services (IFIP WONS), Jackson, WY, February 2017.
  8. Can the Privacy of Primary Networks in Shared Spectrum be Protected?, M. Clark and K. Psounis, in Proceedings of IEEE INFOCOM, San Fransisco, April 2016. (acceptance rate 18.2%)
  9. Efficient Resource Scheduling for a Secondary Network in Shared Spectrum, Matthew Clark and Konstantinos Psounis, in Proceedings of IEEE INFOCOM, Hong-Kong, April 2015. (acceptance rate 19.0%)