|| Two-Level Private Information Retrieval
||Ruida Zhou, Chao Tian, Texas A&M University, United States; Hua Sun, University of North Texas, United States; James S. Plank, University of Tennessee, United States|
||D4-S6-T4: Private Information Retreival II
||Thursday, 15 July, 23:40 - 00:00
||Friday, 16 July, 00:00 - 00:20
In the conventional robust $T$-colluding private information retrieval (PIR) system, the user needs to retrieve one of the possible messages while keeping the identity of the requested message private from any $T$ colluding servers. Motivated by the possible heterogeneous privacy requirements for different messages, we consider the $(N, T_1:K_1, T_2:K_2)$ two-level PIR system, where $K_1$ messages need to be retrieved privately against $T_1$ colluding servers, and all the messages need to be retrieved privately against $T_2$ colluding servers where $T_2\leq T_1$. We obtain a lower bound to the capacity by proposing a novel coding scheme, namely the non-uniform successive cancellation scheme. A capacity upper bound is also derived. The gap between the upper bound and the lower bound is analyzed, and shown to vanish when $T_1=T_2$.