Achievable Secrecy Rate Regions for Relay Networks

Amir Salimi, Jörg Kliewer

In this paper we consider a class of degraded relay networks with multiple receivers, address the capacity region for the multicast case, and consider information-theoretic security aspects. Exemplary, we study a degraded relay network, which is comprised of one source, an arbitrary number of relay nodes, and two receiver nodes in a degraded setting. Based on a simple combination of superposition encoding and decode-and-forward at the relay nodes we provide an achievable rate region for this network. We then consider a related setup where the weaker receiver node now plays the role of an eavesdropper who is able to listen to the transmission of the source and all relays in the network. Based on the result for the degraded relay network case we then provide an achievable secrecy rate for this setup.

Faezeh Sharifi, Soroush Akhlaghi

Yulong Zou, Jia Rong Zhu

Xiang He, Aylin Yener

We consider the secrecy rate of a relay network where an eavesdropper is co-located with the relay node. This exemplifies a scenario where the relay node is not malicious by nature, but is located in an "untrusted region", and hence is potentially compromised. Given that the aim now is to keep the relay node completely oblivious to the information sent from the source to the destination, an interesting question is whether the relay node should be deployed at all. We investigate this question for two types of relay networks with orthogonal components. For the first model, we find the equivocation capacity region and prove the relay node should not be deployed. For the second model, we present an achievable secrecy rate based on compress-and-forward, and conclude that the relay node is potentially useful as it can, without being able to decode the source data, facilitate secret communication between the source and the destination that would not be possible without the relay.

Nanrun Zhou, Xun Chen, Chisheng Li +1 more

Liang Chen

This paper studies the transmission rates and physical layer security of cooperative relay networks. For any relay network, there are natural information theoretic cut-set bounds, which upper bound the transmission rates. We further tighten the outer bounds on rate regions. We also study the achievable rate regions in consideration of physical layer security. We analyze two cross-group cooperative strategies with decode- and-forward relaying and compress-and-forward relaying respectively. We can use the decode-and-forward scheme only if the users which have both good communication channels from the source and high levels of security clearance are selected to serve as relays; otherwise, we cannot achieve physical layer security. While the compress-and-forward scheme is used, physical layer security can be achieved even though the relays have lower security clearance than the destinations.