[Paper Review] Reliable Communications with Asymmetric Codebooks: An Information Theoretic Analysis of Robust Signal Hashing

Yücel Altuğ, M. Kıvanç Mıhçak+2 more|ArXiv.org|Sep 11, 2008

Wireless Communication Security Techniques10 references3 citations

TL;DR

This paper introduces a novel communication model—reliable communications with asymmetric codebooks—where the encoder and decoder use statistically related but distinct codebooks, modeling privacy-preserving signal hashing. It establishes that the system's operational capacity equals the mutual information I(U;Y), and proves a fundamental rate limit due to codebook asymmetry, quantified as a gap from the classical binary symmetric channel capacity.

ABSTRACT

In this paper, a generalization of the traditional point-to-point to communication setup, which is named as "reliable communications with asymmetric codebooks", is proposed. Under the assumption of independent identically distributed (i.i.d) encoder codewords, it is proven that the operational capacity of the system is equal to the information capacity of the system, which is given by $\max_{p(x)} I(U;Y)$, where $X, U$ and $Y$ denote the individual random elements of encoder codewords, decoder codewords and decoder inputs. The capacity result is derived in the "binary symmetric" case (which is an analogous formulation of the traditional "binary symmetric channel" for our case), as a function of the system parameters. A conceptually insightful inference is made by attributing the difference from the classical Shannon-type capacity of binary symmetric channel to the {\em gap} due to the codebook asymmetry.

Motivation & Objective

• To model reliable communication between two collaborating but privacy-conscious parties where the decoder does not have the encoder's codebook.
• To analyze the fundamental limits of information transmission under codebook asymmetry, where the decoder's codebook is a perturbed version of the encoder's.
• To derive the operational capacity of such a system and relate it to the information capacity max_p(x) I(U;Y).
• To quantify the performance gap between this asymmetric setup and the classical binary symmetric channel capacity due to codebook mismatch.
• To establish a theoretical foundation for robust signal hashing in multimedia security, providing an upper bound on achievable transmission rates.

Proposed method

• Proposes a communication model where the encoder uses a codebook X, and the decoder uses a perturbed codebook U, with a known conditional distribution p(u|x).
• Models the system as a point-to-point channel with asymmetric codebooks, where the decoder's codewords are conditionally distributed given the encoder's.
• Applies information-theoretic tools to derive the operational capacity as max_p(x) I(U;Y), where Y is the decoder's channel input.
• Uses a binary symmetric case analysis to explicitly compute the capacity as a function of system parameters, including the perturbation distribution.
• Employs MAP decoding and conditional probability analysis to bound error probabilities, showing that asymmetry inherently limits performance.
• Derives a contradiction in error probability bounds to prove that the maximal error probability remains bounded away from zero, establishing a non-zero gap due to asymmetry.

Experimental results

Research questions

• RQ1 What is the fundamental limit of reliable communication when the encoder and decoder use statistically related but distinct codebooks?
• RQ2 How does codebook asymmetry affect the maximum achievable transmission rate compared to classical point-to-point communication?
• RQ3 Can the operational capacity of such a system be characterized in terms of mutual information I(U;Y)?
• RQ4 What is the quantitative gap between the capacity of this asymmetric system and the classical binary symmetric channel capacity?
• RQ5 How does the structure of the perturbation between codebooks influence the achievable rate and error probability?

Key findings

• The operational capacity of the asymmetric codebook system is equal to the information capacity max_p(x) I(U;Y), establishing a fundamental upper bound on reliable communication.
• In the binary symmetric case, the capacity is explicitly derived as a function of the perturbation parameters, showing a direct dependence on codebook mismatch.
• A non-zero gap exists between the system's capacity and the classical binary symmetric channel capacity, which is attributed to codebook asymmetry.
• The maximal probability of error remains bounded away from zero (at least 1 - 1/m), proving that perfect reliability is unattainable under asymmetry when m ≥ 2.
• The use of MAP decoding does not eliminate the performance gap, as the error probability lower bound is determined by the size of the equivalence class of messages.
• The analysis provides a theoretical foundation for robust signal hashing, showing that any such algorithm is fundamentally limited by the information-theoretic capacity derived here.