Untangling RFID Privacy Models

Table 2

Comparison of the presented privacy models. “✓” (resp., “✗”) means that the feature is (resp., is not) given to the adversary 𝒜. “N/A” means that the feature is not applicable in the model.

Feature	Model
	Avoine	JW	Vaudenay	LBM	DMR	CCEG	DLYZ	HPVP
Interaction with all the tags	Only 2 tags	not 𝒯^∗𝑏⊕1	✓	✓	not corrupted agents	✓	not all clean tags	all-but-one
Choice of the challenge tags	✗	✓	✓	✓	✓	✓	✓	✓
Attack on incomplete executions	Both	✓	✓	✗	✓	✓	✓	✓
CORRUPT challenge tags	Only 𝒯	Only 𝒯^∗𝑏	✓	✓	✗	✓	✗	✓
CORRUPT nonchallenge tags	N/A	✓	N/A	N/A	✓	N/A	✓	N/A
CORRUPT any tag	✗	✗	✓	✓	✗	✓	✗	✓
𝖭𝖠𝖱𝖱𝖮𝖶/𝖶𝖨𝖣𝖤	𝖭𝖠𝖱𝖱𝖮𝖶	𝖶𝖨𝖣𝖤	both	𝖶𝖨𝖣𝖤	𝖭𝖠𝖱𝖱𝖮𝖶	𝖶𝖨𝖣𝖤	𝖶𝖨𝖣𝖤	Both
Channels asymmetry	✓	✗	✗	✗	✗	✗	✗	✗
Protocols analyzable	3-pass with independent secrets	SK based	all	all	all	all	(2𝛾+1)-pass	All
Tag-init protocols analyzable	✗	✓	✗	✗	✓	✗	✗	✗