IEEE Consumer Electronics Magazine - January/February 2022 - 8

Smart Healthcare Security
features are concealed amongst the chaff points.
This vault can then be exchanged with other individuals
with an interest in traveling. For example,
Bob may attempt to unlock the vault using
his features contained in the set B ¼fBerlin;
LosAngeles; NewYorkg and will achieve this as
long as jA
T
Bj > 1 þ n,where n is the degree of
the polynomial. Any other user who does not
meet this requirement will be unable to acquire
Alice's contact information.
Figure 1. Main steps of the fuzzy vault construction
process.
also no chaff point may imitate a legitimate point
C ¼fðCx;CyÞjCx =2 A; Cy 6¼ PðxÞg. 4) Combine
the sets R and C together then order the resulting
set in ascending order. Any attempt to reconstruct
the polynomial, and by extension the
secret S, will require knowledge of n þ 1 features
(elements) present within the set A. The fuzzy
vault construction steps are shown in Figure 1.
An illustrative use case of the fuzzy vault
scheme would be the following. Alice wishes to
exchange her contact details (i.e., phone/mobile
number or email address) on travel forums with
individuals who share the same interest with
regards to international travel. Since Alice does
not want everyone to have access to her contact
information she may distribute them using a
fuzzy vault, by concealing within the vault her
contact information as a secret to be recovered.
Only those who have similar travel interests may
successfully recover Alice's contact information.
Alice will start by encoding her contact information
within the coefficients of a polynomial PðXÞ,
which has a degree that determines the error tolerance
of the vault. The smaller the degree the
fewer overlapping features are required for a successful
unlock, whereas the higher the degree the
more overlapping features are needed. Alice will
then construct a set of elements A which represents
travel locations (i.e., features) around the
world; A ¼fManchester; Berlin; LosAngeles; :::g.
These features will be mapped onto PðXÞ followed
by the generation of chaff points which
must not fall upon the polynomial. The vault is
then shuffled to ensure that the legitimate
8
FUZZY VAULT SECURITY
VULNERABILITY
Two pioneer solutions use the fuzzy vault or
a similar construction known as PSKA (Physiological
Secure Key Agreement)8 and OPFKA
(Ordered-Physiological-Feature-Based Key Agreement).9
These schemes enable key agreement
between BAN nodes that are equipped with either
ECG or PPG (photoplethysmogram) sensors, as
these physiological signals are used to conceal
the symmetric key to be agreed upon. While these
schemes provide a novel way of agreeing upon a
key accompanied with a high-security level, they
do not account for the possibility of the physiological
signals being captured remotely off the
body, without knowledge or consent of the
patient. This remote capture of physiological signals
may allow an adversary to successfully
impersonate a genuine body sensor and enable
them to carry out further attacks within the BAN,
such as data manipulation in which they send
fake readings that can mislead medical professionals
and have serious consequences for the
patient. Kim et al.10 investigated the potential of
using prediction filters, such as a Kalman filter, to
unlock PSKA-based fuzzy vaults using leaked
physiological signals such as PPG. As stated by
Lee et al.,11 the remote capture of PPG signals
using ultrawide band radars is possible, thanks to
advancements in remote sensing technology.
This presents a serious security risk for devices
aiming to achieve authentication and key agreement
using PSKA or similar schemes as the key
could be unlocked by an adversary and neither
the user nor the network administrator would be
aware of the intrusion. In the work by Zhao
et al.,12 a concern is raised regarding the security
of PSKA vaults with the advent of remote sensor
IEEE Consumer Electronics Magazine
IEEE Consumer Electronics Magazine - January/February 2022

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - January/February 2022
