Spectrum Sharing between Radar and Communication Systems

Sponsors:

1. NSF ECCS1443967 EARS Collaborative Research Lets share CommRad – Spectrum Sharing Between Communications and Radar Systems

Problem Statement and Motivation 

• Increased demand for wireless services necessitates the spectrum co-existence between radar and communication systems.
• How an unaltered radar pulse affects the performance of an uncoded communication RX in terms of the Symbol Error Rate (SER).
• Optimize a signal constellation that achieves lowest SER:
  • Weak radar interference (INR << SNR):
    • Radar interference is treated as Gaussian noise.
    • Optimal constellation has a shape of concentric hexagons.
  • Moderate radar interference (INR ≈ SNR):
    • Yields the highest SER.
  • Strong radar interference (INR >> SNR >> 1):
    • Radar signal is completely canceled along with part of comm signal resulting in an irreducible error floor.
    • Optimal constellation has a shape of an unevenly-spaced PAM.

Technical Approach

• Derive the SER for the optimal ML decoder and its approximations using the communication theoretic approach.

Key Achievements and Future Goals

• N. Nartasilpa, D. Tuninetti, N. Devroye, and D. Erricolo, “Let’s share CommRad: effect of radar interference on an uncoded data commu- nication system,” in Proc. of IEEE RadarCon, May 2016.
• N. Nartasilpa, D. Tuninetti, N. Devroye, and D. Erricolo, “On the error rate of a communication system suffering from additive radar interfer- ence,” in Proc. of IEEE GlobeCom, Dec 2016.
• N. Nartasilpa, D. Tuninetti, N. Devroye, and D. Erricolo, “Signal constellation optimization in the presence of radar interference and Gaussian noise,” in submitted to Proc. of IEEE ICC, May 2017.
• Extensions to more practical models such as fading channels, OFDM channels, MIMO channels, and channel-coded systems.