Probabilistic Batch Codes for Private Information Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Private Information Retrieval (PIR) systems, particularly the computational variant (CPIR), face significant challenges in supporting large databases and handling many requests due to high computational and network costs, limiting their deployment and efficiency.
Innovation Solution
The implementation of query compression techniques and probabilistic batch codes (PBCs) reduces network costs and increases throughput by compressing queries and amortizing computational costs, allowing for more efficient processing and support of larger databases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If CPIR is used to ensure privacy-preserving retrieval, then privacy guarantees are achieved, but computational cost increases significantly
Solution Approach 1:
The database is preprocessed during a setup phase to compute batch codes and store encoded versions of database elements. This preliminary action allows the server to answer queries more efficiently by simply retrieving and decoding precomputed values rather than performing expensive cryptographic operations on every element during query processing.
Solution Approach 2:
The database is divided into multiple blocks, and batch codes are computed for each block separately. This segmentation allows the server to process only the relevant blocks for a given query rather than the entire database, reducing computational overhead while maintaining privacy guarantees.
2Reliability
If traditional PIR schemes are used to answer queries, then privacy is preserved, but the server must perform computation on every database element
Solution Approach 1:
Instead of computing batch codes from scratch for each query, the system creates copies of the database elements in encoded form during setup. The server then retrieves these precomputed copies directly, avoiding repetitive computation while preserving privacy through the cryptographic properties of the batch code construction.
3Quantity of substance
If PIR systems support large databases, then more data can be retrieved privately, but network costs and processing time increase
Solution Approach 1:
Batch codes are precomputed and stored during a setup phase, allowing the server to answer queries by simply retrieving preprocessed data. This eliminates the need for time-consuming computations during query processing, enabling efficient handling of large databases without proportional increases in query response time.
4Productivity
If PIR systems handle many requests, then service capacity increases, but computational overhead accumulates
Solution Approach 1:
The batch code structure allows the server to answer multiple queries by retrieving precomputed values and performing only lightweight decoding operations. This amortizes the computational cost across many requests, as the expensive preprocessing is done once during setup rather than repeated for each query.
Data Source
AI summary
Systems, methods, and computer-executable instructions for reducing amortized computational costs for a query that includes receiving at least two indexes for elements stored in an n-element database. The n-element database is encoded into at least three buckets. Each element is stored within at least two buckets. Each bucket stores a proper subset of the n-elements. For each of the two indexes, a bucket is determined to retrieve the element at the index. The determined buckets are queried to retrieve the elements. The elements at the indexes are retrieved based on the querying the determined buckets.