Information Technology Reference
In-Depth Information
Example: Securing Sensor
Communication in PHM-CPS
of the deeply embedded nature of sensors and
their ability to monitor specific physiological
signals from their hosts. The result is a security
solution that combines signal processing along
with cryptographic primitives for secure key agree-
ment. Using PSKA, sensors do not require any
additional keying material to secure inter-sensor
communication, they can now be strapped on by
a patient to ensure secure communication in a
usable (plug-n-play, largely transparent) manner.
A malicious entity cannot eavesdrop on the com-
munication unless it has access to physiological
signals. A secure channel between sensors, PSKA's
brings out many new “allied” properties to security
which traditional (mostly involving manual key
pre-deployment) secure key agreement protocols
do not posses:
We have developed a novel CYPSec solution for
securing the communication in a PHMS called
the Physiological Signal based key Agreement
(PSKA) (Cherukuri, Venkatasubramanian, &
Gupta, 2003) (Venkatasubramanian, Banerjee,
& Gupta, Plethysmogram-based Secure Inter-
Sensor Communication in Body Area Networks,
2008) (Venkatasubramanian, Banerjee, & Gupta,
PSKAVenkatasubramanian, Banerjee, & Gupta,
PSKA: Usable and Secure Key Agreement
Scheme for Body Area Networks, 2010). PSKA
is a scheme designed to enable automated key
agreement between sensors without any form of
external user involvement. The deployment of the
key establishes the trusted channel between the
sensors in the network. It functions by generating
simple frequency domain (Fast Fourier Transform
(FFT)) based feature vectors from common physi-
ological signals (e.g. Photoplethysmogram and
Electrocardiogram) from the human body, and
using them to hide/lock a session key in a cryp-
tographically secure manner. The hidden session
key is then transmitted to another sensor in the
open, which uses its own copy of the feature vector
(derived from the same physiological signal in a
loosely synchronous manner) to unhide/ unlock
the session key. PSKA has been implemented
on implemented on VHDL and Crossbow mote
platform and analyzed for sustainability using
well known energy scavenging mechanisms. The
results demonstrate that PSKA, despite its seem-
ingly complex operational requirements, is not
only implementable on low-capability systems
but is also sustainable. Details can be found in
(Banerjee, Venkatasubramanian, & Gupta, 2009)
(Venkatasubramanian, Banerjee, & Gupta, Green
and Sustainable Cyber Physical Security Solutions
for Body Area Networks, 2009)
The operation of PSKA inherently depends
upon features from the physical element of the
PHMS cyber-physical system. It takes advantage
Secure Interoperability : The PSKA
CYPSec solution reduces one of the im-
portant hurdle from achieving interopera-
bility by enabling sensors in the PHM-CPS
to establish a secure channel between one
another enabling them to communicate in a
seamlessly manner.
Authentication : As physiological signal
features uniquely represent the host at
that given time, any successful agreement
of keys between two sensors assures the
participating nodes that the other is in the
same network. This eliminates the need for
additional authentication primitives need-
ed by traditional key agreement protocols
such as Diffie-Hellman and its variants.
Minimally Complex: Given the dynamic
nature of the physical process it might be
possible that in some cases the precision
of implementation can be reduced with-
out significantly affecting behavior of the
solution. A case in point is the implemen-
tation of PSKA on FPGAs and Crossbow
Motes (http://www.xbow.com) for which
we had to introduce many approximations
(for floating point and complex arithmetic)
Search WWH ::




Custom Search