NDSS 2025 – Iris: Dynamic Privacy Preserving Search In Authenticated Chord Peer-To-Peer Networks

Why It Matters

Iris makes structured P2P overlays viable for privacy‑sensitive applications, lowering the barrier for secure decentralized services. Its tunable overhead ensures organizations can balance performance and confidentiality according to risk profiles.

Key Takeaways

• Iris adds privacy without altering Chord protocol
• Alpha‑delta privacy extends k‑anonymity for iterative searches
• Overhead scales linearly with desired privacy level
• Prototype implemented in Matlab demonstrates tunable performance
• No network‑wide adoption required for privacy benefits

Pulse Analysis

Structured peer‑to‑peer overlays such as Chord have long been prized for their scalability and deterministic routing, yet each hop in the lookup process reveals the target key to intermediate nodes. This exposure makes traditional Chord unsuitable for scenarios where query confidentiality is paramount—think intellectual property sharing, confidential data stores, or regulated financial transactions. Researchers have therefore sought mechanisms that can mask the query path without dismantling the underlying routing guarantees that give Chord its efficiency.

Iris addresses this gap by embedding a privacy layer that operates transparently atop the standard Chord protocol. The core contribution is the alpha‑delta privacy model, a formal notion inspired by k‑anonymity that quantifies the indistinguishability of a query across all iterative hops, even against colluding adversaries. Because Iris does not require peers to modify their routing logic or announce the privacy feature, it can be deployed incrementally. The authors back their design with a rigorous security analysis and demonstrate, via a Matlab prototype, that the added latency and bandwidth costs are modest and directly proportional to the selected privacy parameters.

Performance testing shows that when privacy is disabled, Iris incurs virtually no overhead, while enabling higher alpha‑delta levels introduces a predictable, linear increase in message complexity. This tunability empowers system architects to align privacy settings with service‑level agreements and regulatory constraints. As decentralized applications continue to mature—especially in finance, healthcare, and edge computing—the ability to preserve query confidentiality without overhauling network infrastructure positions Iris as a practical building block for the next generation of secure P2P services.