Communication in the Presence of a State-Aware Adversary
Amitalok J. Budkuley, Bikash Kumar Dey, Vinod M. Prabhakaran

TL;DR
This paper characterizes the capacity of state-dependent channels with a malicious, state-aware jammer, showing that the worst-case jamming strategy is memoryless and that the capacity can be explicitly determined for Gaussian channels.
Contribution
It provides a capacity characterization for state-dependent AVCs with a malicious adversary, demonstrating the adversary's optimal strategies and their impact on capacity.
Findings
Capacity is unaffected by the adversary's non-causal knowledge of the state.
The worst-case jamming strategy is memoryless, depending only on the current state.
For Gaussian channels, the capacity equals that of a standard AWGN channel with additive noise.
Abstract
We study communication systems over state-dependent channels in the presence of a malicious state-aware jamming adversary. The channel has a memoryless state with an underlying distribution. The adversary introduces a jamming signal into the channel. The state sequence is known non-causally to both the encoder and the adversary. Taking an Arbitrarily Varying Channel (AVC) approach, we consider two setups, namely, the discrete memoryless Gel'fand-Pinsker (GP) AVC and the additive white Gaussian Dirty Paper (DP) AVC. We determine the randomized coding capacity of both the AVCs under a maximum probability of error criterion. Similar to other randomized coding setups, we show that the capacity is the same even under the average probability of error criterion. Even with non-causal knowledge of the state, we prove that the state-aware adversary cannot affect the rate any worse than when it…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
