Sidong Guo

Sidong Guo, Matthieu R. Bloch

Sensing privacy and communication confidentiality play fundamentally different but interconnected roles in adversarial wireless environments. Capturing this interplay within a single physical-layer framework is particularly challenging in integrated sensing and communication (ISAC) systems, where the same waveform simultaneously serves dual purposes. We study a secure ISAC system in which a monostatic transmitter simultaneously sends a confidential message to a legitimate receiver and senses an environmental state, while a passive adversary attempts both message decoding and state estimation. We partially characterize the fundamental trade-offs among three performance measures: the transmitter's secrecy rate, its detection exponent, and the adversary's detection exponent. Beyond the joint input distribution that governs overall performance, the trade-offs are further shaped by the transmitter's ability to extract keys via feedback and hide both the content and structure of the codewords via wiretap and resolvability codes. We derive an achievable region, and illustrate the resulting design trade-offs through a numerical example.

Sidong Guo, Tyler Kann, Teodora Baluta et al.

We study the problem of multi-bit watermarking for large language models (LLMs). We introduce a block-autoregressive model inspired by multi-token prediction, in which the encoder has limited non-causal access to token distributions within each block. This formulation enables an information-theoretic characterization of multi-bit watermarking capacity, by which the knowledge of LLM cover statistics is leveraged to enable a multi-bit covert embedding. We study the information-theoretic limits of the model by combining Gelfand-Pinsker and channel synthesis coding techniques and obtain an exact characterization of the capacity. The embedding strategy is further optimized across blocks using a constrained Markov decision process (CMDP) and we develop an explicit algorithm based on polar codes following the information-theoretic principles. Our algorithm achieves a bit-error rate below 10 percent with a rate of 0.375 bits/token over short token lengths with negligible perplexity and distortion degradation.