Fading Cognitive Multiple-Access Channels With Confidential Messages

The fading cognitive multiple-access channel with confidential messages (CMAC-CM) is investigated, in which two users attempt to transmit common information to a destination and user 1 also has confidential information intended for the destination. User 1 views user 2 as an eavesdropper and wishes to keep its confidential information as secret as possible from user 2. The multiple-access channel (both the user-to-user channel and the user-to-destination channel) is corrupted by multiplicative fading gain coefficients in addition to additive white Gaussian noise. The channel state information (CSI) is assumed to be known at both the users and the destination. A parallel CMAC-CM with independent subchannels is first studied. The secrecy capacity region of the parallel CMAC-CM is established, which yields the secrecy capacity region of the parallel CMAC-CM with degraded subchannels. Next, the secrecy capacity region is established for the parallel Gaussian CMAC-CM, which is used to study the fading CMAC-CM. When both users know the CSI, they can dynamically change their transmission powers with the channel realization to achieve the optimal performance. The closed-form power allocation function that achieves every boundary point of the secrecy capacity region is derived.

💡 Research Summary

This paper investigates a novel communication scenario called the fading cognitive multiple‑access channel with confidential messages (CMAC‑CM). In this model two users simultaneously transmit a common message to a single destination, while user 1 also wishes to send a private (confidential) message that must be kept secret from user 2, which is treated as an internal eavesdropper. Both the user‑to‑user link (the “cognitive” link) and the user‑to‑destination links are subject to multiplicative fading gains and additive white Gaussian noise (AWGN). Crucially, the full channel state information (CSI) is assumed to be known instantaneously at both transmitters and at the receiver.

The authors first consider a parallel CMAC‑CM consisting of a set of independent sub‑channels. For each sub‑channel they derive the secrecy capacity region, i.e., the set of all achievable rate pairs ((R_{c},R_{s})) where (R_{c}) is the common rate and (R_{s}) the confidential rate. By treating the sub‑channels independently and then time‑sharing across them, they obtain the overall secrecy capacity region of the parallel system. When the sub‑channels are degraded (user 2’s observation is always a degraded version of user 1’s), the region simplifies to a single‑letter expression that can be evaluated in closed form.

Building on this result, the paper specializes to the parallel Gaussian CMAC‑CM, where each sub‑channel is a real (or complex) Gaussian MAC with additive noise. The presence of secrecy constraints couples the power allocation across sub‑channels: the transmitter must balance the need to boost the confidential signal at the destination against the risk of leaking information to user 2. Using Lagrange‑multiplier techniques and the Karush‑Kuhn‑Tucker (KKT) conditions, the authors derive the optimal power‑allocation policy that achieves every boundary point of the secrecy capacity region. The solution resembles a water‑filling algorithm but includes an extra “secrecy weight” that depends on the instantaneous SNRs of both the legitimate and eavesdropping links. In particular, power is allocated to a sub‑channel only when the legitimate link is strong relative to the eavesdropper’s link; otherwise the power is reduced or set to zero.

The parallel Gaussian results are then lifted to the fading CMAC‑CM by averaging over the fading distribution. Because CSI is available at both users, the optimal power‑allocation policy can be applied on a per‑realization basis, leading to an “opportunistic secrecy” strategy: user 1 transmits confidential information aggressively when user 2’s channel is weak and backs off when user 2’s channel improves. The authors provide explicit closed‑form expressions for the optimal power functions for a wide class of fading models (e.g., Rayleigh, Nakagami).

Numerical simulations illustrate the theoretical findings. The proposed secrecy‑aware power allocation substantially enlarges the achievable secrecy region compared with naïve schemes such as equal power allocation or conventional water‑filling that ignores the eavesdropper. The gain is especially pronounced at moderate SNRs and for highly variable fading. The paper also discusses the impact of imperfect or delayed CSI, showing that the derived policies are robust and can be approximated with modest performance loss.

In summary, the work makes three key contributions: (1) it establishes the secrecy capacity region for a parallel CMAC‑CM with independent sub‑channels, (2) it derives the optimal power‑allocation policy for the parallel Gaussian CMAC‑CM, and (3) it translates these results to the fading CMAC‑CM, providing closed‑form power control laws that achieve any point on the secrecy capacity boundary. The analysis bridges cognitive multiple‑access networking and physical‑layer security, offering practical guidelines for designing secure, power‑adaptive wireless systems where users may simultaneously act as collaborators and potential eavesdroppers.