Information Technology Reference
In-Depth Information
in FFT and other signal processing com-
putation which enabled us to emulate our
Matlab-based benchmark results without
sacrificing the security of the approach
(Banerjee, Venkatasubramanian, & Gupta,
2009).
•
Non-Determinism Tolerant
: The use of
physiological signals based features in-
troduces a level of non-determinism into
the key agreement process which is diffi-
cult to overcome. It is possible (although
the probability is very low) with PSKA,
that the physiological features between
two sensors are not significantly similar
to enable successful unlocking of the key.
In such cases the whole process has to be
repeated. Using PSKA however the key
agreed can be arbitrarily long. Therefore, it
needs not to be executed on a regular basis,
thus amortizing the cost of the compensa-
tory mechanism.
which response actions have to be taken for the
criticality to be controlled (Gupta, Mukherjee, &
Venkatasubramanian, 2006).
CAAC is an adaptive access control approach
designed to facilitate the control of all the active
criticalities within the system. It uses an Action
General Model (AGM) based on the stochastic cri-
sis planning technique developed in (Mukherjee,
Venkatasubramanian, & Gupta, 2006). The results
of AGM execution (list of response actions for
different combinations of criticalities such that
the window-of-opportunity of all the criticalities
is satisfied) are provided to the CAAC, before it
is deployed. CAAC monitors the environment it
is deployed in, at regular intervals, and depending
upon the state the system is in (the current state of
the system is a result of all responses and criticality
which occurred since the system observed its first
criticality), it identifies the best response actions
(authorizations) that need to be taken from that
state to reach the normal state (
e.g.,
data access or
actuations to be performed or even reduced privi-
leges). It then identifies the set of subjects (
e.g.,
caregivers) that can be either statically specified
or dynamically computed based on subject context
and are best suited to execute the response actions,
and provides them with credentials to execute the
actions. We call these new privileges for chosen
subjects as alternate privileges. This change in
credentials of subjects is temporary and reverts
back when the criticalities in the system change,
are controlled or expire.
CAAC enables authorized actuation in a PHM-
CPS, and can be used in both autonomic and
active modes, as it only facilitates the execution
of response actions. CAAC, combines stochastic
crisis model with access control primitives, and
enable authorizations of subjects of the system for
managing criticalities even before they explicitly
ask for it, making emergency management more
user-centric. CAAC inherently also brings out new
“allied” properties to security policy specification:
Example: Securing Information
Access from PHMS
For securing access to information in a PHM-
CPS (or any smart-infrastructure), we have
developed a novel access control model called
Criticality Aware Access Control (CAAC) (Gupta,
Mukherjee, & Venkatasubramanian, 2006) (Ven-
katasubramanian, Mukherjee, & Gupta, CAAC
- An Adaptive and Proactive Access Control Ap-
proach for Emergencies for Smart Infrastructures,
2011). CAAC is a CYPSec access control model
which has the ability to provide the right set of
privileges for the right set of subjects, at the right
time for the right duration to facilitate emergency
(criticality) response. Criticalities are situations
which require urgent response actions in order to
control and maintain the stability of the system.
Each criticality has a timing duration associated
with it known as window-of-opportunity, within
Search WWH ::
Custom Search