Performance Tradeoff in a Unified System of Communications and Passive Radar: A Secrecy Capacity Approach

TL;DR

This paper explores the balance between passive radar and communication system performance by optimizing secrecy capacity and SINR, proposing methods for both overlapping and non-overlapping resource scenarios.

Contribution

It introduces novel optimization techniques for maximizing SINR while maintaining a secrecy rate threshold in a unified radar-communication system.

Findings

Overlapping resource scenario outperforms non-overlapping in secrecy capacity.

Semi-analytical approach offers computational advantages.

Tradeoff boundaries between radar and communication performance are characterized.

Abstract

In a unified system of passive radar and communication systems of joint transmitter platform, information intended for a communication receiver may be eavesdropped by a passive radar receiver (RR), thereby undermining the security of communications system. To minimize this information security risk, in this paper, we propose a unified passive radar and communications system wherein the signal-to-interference and noise ratio (SINR) at the RR is maximized while ensuring that the information secrecy rate is above a certain threshold value. We consider both scenarios wherein transmissions of the radar waveform and information signals are scheduled with the disjoint (non-overlapping case) as well as with the same set of resources (overlapping case). In both cases, the underlying optimization problems are non-convex. In the former case, we propose alternating optimization (AO) techniques that employ semidefinite programming and computationally efficient semi-analytical approaches. In the latter case, AO method based on semi-definite relaxation approach is proposed to solve the optimization problem. By changing the threshold value of the information secrecy rate, we then characterize the performance tradeoff between passive radar and communication systems with the boundaries of the SINR-secrecy capacity regions. The performance comparison of the proposed optimization methods demonstrate the importance of the semi-analytical approach and the advantage of overlapping case over non-overlapping one.