The Capacity of Single-Server Weakly-Private Information Retrieval
Paper in proceeding, 2020

Weakly-private information retrieval (WPIR) is a variant of the private information retrieval problem in which a user wants to efficiently retrieve a file stored across a set of servers while tolerating some information leakage on the identity of the requested file to the servers. In this paper, we consider WPIR from a single-server database where the information leakage is measured in terms of the mutual information (MI) or maximal leakage (MaxL) privacy metrics. In particular, we establish a connection between the WPIR problem and rate-distortion theory, and fully characterize the optimal tradeoff between the download cost and the allowed information leakage under the MI and MaxL metrics, settling the single-server WPIR capacity.

Author

Hsuan Yin Lin

Simula UiB

Siddhartha Kumar

Simula UiB

Eirik Rosnes

Simula UiB

Alexandre Graell I Amat

Simula UiB

Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks

Eitan Yaakobi

Technion – Israel Institute of Technology

Published in

IEEE International Symposium on Information Theory - Proceedings

21578095 (ISSN)

Vol. 2020-June p. 1053-1058 art. no 9174048
9781728164328 (ISBN)

Conference

2020 IEEE International Symposium on Information Theory, ISIT 2020
Los Angeles, USA, 2020-07-20 - 2020-07-25

Research Project(s)

Rethinking Distributed Storage for Data Storage and Wireless Content Delivery

Swedish Research Council (VR) (2016-04253), 2016-01-01 -- 2019-12-31.

Categorizing

Subject Categories (SSIF 2011)

Telecommunications

Information Studies

Communication Systems

Identifiers

DOI

10.1109/ISIT44484.2020.9174048

More information

Latest update

12/20/2021