[Paper Review] Message and state cooperation in multiple access channels

Haim H. Permuter, Shlomo+2 more|arXiv (Cornell University)|Jun 10, 2010

Wireless Communication Security Techniques18 references4 citations

TL;DR

This paper introduces a novel framework for multiple access channels (MAC) with cooperating encoders where each encoder has partial state information and the decoder has full state knowledge. By employing double-binning to simultaneously achieve state coordination and private message sharing over a finite-capacity cooperation link, the authors derive the capacity region, demonstrating that joint message-state cooperation strictly outperforms separate cooperation strategies in terms of achievable rates.

ABSTRACT

We investigate the capacity of a multiple access channel with cooperating encoders where partial state information is known to each encoder and full state information is known to the decoder. The cooperation between the encoders has a two-fold purpose: to generate empirical state coordination between the encoders, and to share information about the private messages that each encoder has. For two-way cooperation, this two-fold purpose is achieved by double-binning, where the first layer of binning is used to generate the state coordination similarly to the two-way source coding, and the second layer of binning is used to transmit information about the private messages. The complete result provides the framework and perspective for addressing a complex level of cooperation that mixes states and messages in an optimal way.

Motivation & Objective

• To characterize the capacity region of a multiple access channel where encoders have partial state information and cooperate via a finite-capacity link.
• To unify two forms of cooperation—state coordination and private message sharing—within a single coding framework.
• To show that joint message-state cooperation via double-binning achieves a strictly larger capacity region than separate message and state cooperation.
• To extend prior results on cooperating encoders and state-dependent channels by incorporating both message and state cooperation in a unified model.

Proposed method

• Double-binning is used: the first layer enables empirical coordination of state information between encoders, and the second layer enables transmission of private message information.
• The cooperation link is used to share both state-related actions (for coordination) and message-related data (for message sharing).
• The auxiliary random variables U and V are introduced: U is a function of the state sequence only, and V is a function of the first encoder's message, ensuring statistical independence between state and message-related cooperation.
• The capacity region is derived using a joint distribution of the form P(s,u)P(v)P(x₁|s,u,v)P(x₂|u,v)P(y|x₁,x₂,s), which models the dependence structure of the system.
• The achievability proof constructs codebooks multiplexed over the auxiliary variable U, enabling coordinated transmission based on state and message cooperation.
• The converse proof identifies U as a function of the state sequence and V as a function of the message, ensuring the tightness of the outer bound.

Experimental results

Research questions

• RQ1 What is the capacity region of a MAC with cooperating encoders when each encoder has partial state information and the decoder has full state knowledge?
• RQ2 Can joint message and state cooperation via a single cooperation link outperform separate message and state cooperation strategies?
• RQ3 How should the cooperation link be utilized to simultaneously achieve state coordination and private message sharing?
• RQ4 What is the optimal coding strategy that balances state coordination and message exchange in a two-user MAC?
• RQ5 Does the use of double-binning enable a strictly larger capacity region compared to splitting cooperation into separate message and state links?

Key findings

• The capacity region is characterized by a set of mutual information inequalities involving auxiliary random variables U and V, with constraints on cooperation rates for state and message sharing.
• Joint message and state cooperation via double-binning achieves a strictly larger capacity region than separate cooperation, as demonstrated in Figure 13 where C₁₂ = 0.5 outperforms C₁₂ᵐ = C₁₂ˢ = 0.25.
• The optimal cooperation strategy requires the use of double-binning: the first binning layer ensures empirical coordination of state information, and the second layer enables message exchange.
• The auxiliary variable U is identified as a function of the state sequence only, and V as a function of the first encoder’s message, ensuring that (U, Sⁿ) is independent of V.
• When cooperation is split into separate message and state links, the resulting capacity region is strictly suboptimal, as shown by the comparison in Figure 13.
• The converse proof establishes that the derived region is tight, confirming it as the complete capacity region for the model.