Private Information Retrieval with Stateful Client Sublinear Operations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing Private Information Retrieval (PIR) protocols face inefficiencies due to the requirement of stateless clients in single-server systems, leading to high computational costs and inability to handle parallel queries, while multi-server systems are infeasible in real-world scenarios where servers may collude, and existing solutions either sacrifice parallel access or require client-to-client communication.
Innovation Solution
A single-server PIR system that allows clients to maintain independent asynchronous states, reducing the number of encrypted operations by using a private batched sum retrieval method and pseudorandom partitioning to retrieve data blocks without revealing access patterns, enabling efficient and parallel querying.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional PIR schemes are used with stateless clients, then privacy is maintained, but computational cost increases to Ω(n) encrypted operations
Solution Approach 1:
The system performs preliminary actions by having clients initialize local state buffers before querying. This pre-computation of sum buffers and pseudorandom partitioning keys enables subsequent queries to be answered with sublinear computational cost, resolving the contradiction between maintaining privacy and reducing computational operations from Ω(n) to O(1) per query.
Solution Approach 2:
The system transitions from static stateless client operation to dynamic stateful operation where clients maintain and update local buffers. The state evolves through initialization, querying, and periodic re-initialization phases, allowing the system to adapt computational work between client preprocessing and server response, thereby improving efficiency while preserving privacy.
2Device complexity
If single-server PIR is used with stateless clients, then implementation is simplified, but parallel querying capability is lost
Solution Approach 1:
The introduction of stateful clients with local buffers fundamentally changes the interaction dynamics between client and server. Clients can independently manage their query sequences and maintain state without server coordination, enabling parallel queries while keeping the server implementation relatively simple. The state acts as a local cache that decouples concurrent query operations.
3Adaptability or versatility
If multi-server PIR is used, then parallel querying is enabled, but server collusion becomes a security risk
Solution Approach 1:
The patent extracts the parallel querying capability from the server architecture and relocates it to the client side through stateful operation. Instead of using multiple servers to enable parallel access, a single server works with stateful clients who maintain local state, thereby eliminating the collusion risk while preserving parallel querying capability.
4Reliability
If clients download all content from storage devices, then access patterns are not revealed, but retrieval time becomes too long
Solution Approach 1:
Instead of downloading all content (excessive action), the system uses partial action by downloading only the specific data block needed. The stateful client mechanism with local buffers and pseudorandom partitioning enables this selective retrieval while maintaining access pattern privacy, thereby resolving the contradiction between privacy and retrieval time.
Data Source
AI summary
A method includes initializing a client state on a client device be executing a private batched sum retrieval instruction to compute c sums O of data blocks from an untrusted storage device. Each computed sum O stored on memory hardware of the client device and including a sum of a corresponding subset S of exactly k data blocks. The method also includes a query instruction to retrieve a query block Bq stored on the untrusted storage device by iterating through each of the c sums O of data blocks to identify one of the c sums O that does not include the query block Bq, instructing a service to pseudorandomly partition the untrusted storage device into partitions and sum the data blocks in each partition to determine a corresponding encrypted data block sum.