Cognitive Interference Channels with Confidential Messages

TL;DR

This paper characterizes the capacity-equivocation region of the cognitive interference channel with confidential messages, providing explicit formulas for the discrete memoryless and Gaussian cases, and establishing a new capacity theorem for interference channels.

Contribution

It introduces a single-letter expression for the capacity-equivocation region and derives explicit results for Gaussian channels, advancing understanding of secure communication in cognitive interference channels.

Findings

Derived the capacity-equivocation region for discrete memoryless channels.

Explicit capacity formulas for Gaussian cognitive interference channels.

Established a new capacity theorem for interference channels without secrecy constraints.

Abstract

The cognitive interference channel with confidential messages is studied. Similarly to the classical two-user interference channel, the cognitive interference channel consists of two transmitters whose signals interfere at the two receivers. It is assumed that there is a common message source (message 1) known to both transmitters, and an additional independent message source (message 2) known only to the cognitive transmitter (transmitter 2). The cognitive receiver (receiver 2) needs to decode both messages, while the non-cognitive receiver (receiver 1) should decode only the common message. Furthermore, message 2 is assumed to be a confidential message which needs to be kept as secret as possible from receiver 1, which is viewed as an eavesdropper with regard to message 2. The level of secrecy is measured by the equivocation rate. A single-letter expression for the capacity-equivocation region of the discrete memoryless cognitive interference channel is established and is further explicitly derived for the Gaussian case. Moreover, particularizing the capacity-equivocation region to the case without a secrecy constraint, establishes a new capacity theorem for a class of interference channels, by providing a converse theorem.