Prerequisites & Notation

Before You Begin

This chapter introduces information-theoretic demand privacy in coded caching. Prerequisites: the MAN scheme, D2D caching, and basic information-theoretic security (Shannon's perfect secrecy concept).

MAN coded caching (Ch 2)(Review ch02)
Self-check: Can you state the MAN rate formula and the XOR delivery pattern?
D2D caching and scaling law (Ch 10)(Review ch10)
Self-check: Can you state the $\Theta(M/N)$ per-user scaling?
Shannon's perfect secrecy(Review ch20)
Self-check: What is the key-rate requirement for one-time pad encryption?
Mutual information and KL-divergence(Review ch01)
Self-check: What does $I(X;Y) = 0$ mean operationally?
Information leakage vs computational privacy
Self-check: Why is information-theoretic privacy stronger than computational (cryptographic) privacy?

Notation for This Chapter

Symbols for demand privacy analysis.

Symbol	Meaning	Introduced
$\mathbf{d} = (d_1, \ldots, d_K)$	Demand vector; each $d_k$ is user $k$ 's requested file index	s01
$X_{\mathbf{d}}$	Broadcast message sent by the server under demand $\mathbf{d}$	s01
$I(\mathbf{d}_{-k}; X_{\mathbf{d}} \| \mathcal{Z}_k, d_k)$	Leakage: how much user $k$ learns about others' demands	s01
$\mathcal{K}$	Shared randomness (key) distributed among users	s02
$z$	Colluding coalition size in D2D private caching	s03
$t$	Caching gain $t = KM/N$ (unchanged from non-private MAN)	s01
$R_{\text{priv}}$	Private delivery rate (same as MAN for shared-link)	s02

← Ch 11 The Demand Privacy Model