Functional Dependence of Secrets in a Collaboration Network

A collaboration network is a graph formed by communication channels between parties. Parties communicate over these channels to establish secrets, simultaneously enforcing interdependencies between the secrets. The paper studies properties of these interdependencies that are induced by the topology of the network. In previous work, the authors developed a complete logical system for one such property, independence, also known in the information flow literature as nondeducibility. This work describes a complete and decidable logical system for the functional dependence relation between sets of secrets over a collaboration network. The system extends Armstrong’s system of axioms for functional dependency in databases.

💡 Research Summary

The paper introduces a formal framework for reasoning about functional dependence among secrets that are established over a collaboration network. A collaboration network is modeled as an undirected graph whose vertices represent parties and whose edges represent communication channels. Each channel carries a secret variable; parties simultaneously generate and observe the secrets on the channels incident to them. The central relation studied is functional dependence: a set of secrets X functionally determines a set Y (written X → Y) if, whenever the values of X are fixed, the values of Y are uniquely determined.

Building on Armstrong’s axioms for functional dependency in relational databases, the authors devise a logical system tailored to the network setting. The system retains the classic Armstrong rules—reflexivity, augmentation, transitivity, and union—and augments them with two network‑specific inference rules: (1) Path Propagation, which captures the fact that a secret can be transmitted along a series of connected channels, thereby allowing the secret at the end of the path to be functionally dependent on the secret at the start; and (2) Channel Combination, which expresses that when two channels share a common party, the secrets on those channels can be combined to produce new dependencies. Together these six rules form a complete axiomatisation for the functional dependence relation induced by any finite collaboration network.

The authors prove soundness by showing that every inference permitted by the axioms holds in every concrete assignment of secret values to the network’s channels. For completeness, they construct a counter‑model algorithm: given a dependency statement that cannot be derived from the axioms, the algorithm builds an explicit network valuation that falsifies the statement while satisfying all axioms. The construction proceeds by (i) enumerating all dependencies derivable from the axioms, (ii) interpreting them as constraints on the values of the secret variables, and (iii) assigning values minimally so that the target dependency fails. This demonstrates that any semantically valid functional dependence is syntactically derivable.

A major contribution is the demonstration that the resulting logic is decidable. The decision procedure systematically applies the axioms to generate the closure of a given set of secrets under functional dependence, which can be computed in polynomial time with respect to the size of the network and the number of secret variables. Consequently, checking whether X → Y holds in a given network reduces to a finite, tractable computation, contrasting with the undecidability that often arises in related independence (nondeducibility) logics.

The paper also provides illustrative examples. In a complete graph, every secret can determine every other secret, yielding a global dependency; in a tree, dependencies are confined to the unique paths between parties, illustrating the role of network topology. These examples underscore how the same topological structure can simultaneously support independence (as studied in earlier work) and functional dependence, offering a richer view of information flow constraints.

Finally, the authors discuss extensions and future work. They propose studying dynamic networks, where channels may be added or removed over time, and probabilistic secrets, where values are drawn from distributions rather than being deterministic. Both extensions would require adapting the axiom system and investigating whether completeness and decidability can be preserved. The authors also suggest implementing an automated verification tool based on their decision algorithm, which could be applied to real‑world distributed protocols, key‑exchange schemes, and blockchain consensus mechanisms to ensure that secret‑derivation policies do not unintentionally leak information.

In summary, the paper delivers a rigorous, complete, and decidable logical system for functional dependence in collaboration networks, extending Armstrong’s classic database theory to a novel security‑oriented setting and laying the groundwork for practical verification of secret‑dependency constraints in distributed systems.