Succinct Encrypted Filters for Low-Bandwidth Set Membership
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing private set membership systems require clients to store and manage large encrypted identifier sets, exceeding their bandwidth and storage capabilities, while maintaining privacy and security.
Innovation Solution
Implementing succinct filters that are smaller in size than the encrypted identifier sets, allowing clients to perform approximate membership queries using a server-generated filter, with optional cryptographic protocols for confirmation, reducing bandwidth and storage requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If Bloom filters are used for private set membership, then communication overhead is reduced, but false positive rate increases and precision is lost
Solution Approach 1:
The Bloom filter is divided into multiple independent segments or sections. Each segment maintains its own bit array and hash function set, allowing the system to process queries in parallel across segments. This segmentation reduces the false positive rate for each individual segment while maintaining overall efficiency, as the false positives are distributed across multiple smaller units rather than concentrated in a single large filter.
Solution Approach 2:
The patent combines Bloom filters with other data structures or filtering mechanisms to create a composite filtering system. This hybrid approach integrates the space efficiency of Bloom filters with the precision of alternative structures, achieving a balance between low communication overhead and reduced false positive rates through the synergistic properties of the combined system.
2Quantity of substance
If traditional Bloom filters are used, then space efficiency is achieved, but precision is lost due to false positives
Solution Approach 1:
By segmenting the Bloom filter into multiple smaller filters, each with its own bit array, the system maintains space efficiency while reducing the probability of false positives. The segmented structure allows for more granular control over false positive rates without significantly increasing overall space requirements, as each segment can be optimized independently.
Solution Approach 2:
The patent modifies key parameters of the Bloom filter, such as the number of hash functions, bit array size, and segment count, to optimize the balance between space efficiency and precision. By adjusting these parameters dynamically or through careful selection, the system achieves better false positive rates while maintaining compact representation suitable for bandwidth-constrained environments.
3Quantity of substance
If set difference operations are performed on Bloom filters, then communication bandwidth is saved, but precision deteriorates due to false positives
Solution Approach 1:
Set difference operations are performed on segmented Bloom filters rather than single monolithic filters. This segmentation allows the system to identify and eliminate false positives more effectively, as errors are localized to specific segments. The communication bandwidth savings are maintained while precision improves through the ability to selectively process and validate results from individual segments.
Solution Approach 2:
The patent introduces intermediary verification mechanisms or additional filtering stages that act as mediators between the Bloom filter difference operation and the final result. These intermediaries help validate whether detected differences are genuine or false positives, thereby improving precision without requiring additional communication bandwidth for the core set difference operation.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
A method (600) includes obtaining, from a server (140), a filter (400) including a set of encrypted identifiers (152E) each encrypted with a server key (162). The method includes obtaining a request (170) that requests determination of whether a query identifier (172) is a member of a set of identifiers (152) corresponding to the set of encrypted identifiers. The method also includes transmitting an encryption request (212) to the server that requests the server encrypt the query identifier. The method includes receiving, from the server, an encrypted query identifier (172E) including the query identifier encrypted by the server key and determining, using the filter, whether the encrypted query identifier is not a member of the set of encrypted identifiers. When the encrypted query identifier is not a member, the method includes reporting that the query identifier is not a member of the set of identifiers.