Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020231334 - MODELLING AND BLACK-BOX SECURITY TESTING OF CYBER-PHYSICAL SYSTEMS

Publication Number WO/2020/231334
Publication Date 19.11.2020
International Application No. PCT/SG2020/050271
International Filing Date 08.05.2020
IPC
G05B 17/00 2006.1
GPHYSICS
05CONTROLLING; REGULATING
BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
17Systems involving the use of models or simulators of said systems
G05B 13/04 2006.1
GPHYSICS
05CONTROLLING; REGULATING
BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
13Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
02electric
04involving the use of models or simulators
G06N 20/00 2019.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
G06F 21/50 2013.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
21Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
50Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
CPC
G05B 17/02
GPHYSICS
05CONTROLLING; REGULATING
BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
17Systems involving the use of models or simulators of said systems
02electric
G06F 21/577
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
21Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
50Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
57Certifying or maintaining trusted computer platforms, e.g. secure boots or power-downs, version controls, system software checks, secure updates or assessing vulnerabilities
577Assessing vulnerabilities and evaluating computer system security
G06N 20/10
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
10using kernel methods, e.g. support vector machines [SVM]
Applicants
  • SINGAPORE UNIVERSITY OF TECHNOLOGY AND DESIGN [SG]/[SG]
Inventors
  • CASTELLANOS, John Henry
  • ZHOU, Jianying
Agents
  • DAVIES COLLISON CAVE ASIA PTE. LTD.
Priority Data
10201904214T10.05.2019SG
Publication Language English (en)
Filing Language English (EN)
Designated States
Title
(EN) MODELLING AND BLACK-BOX SECURITY TESTING OF CYBER-PHYSICAL SYSTEMS
(FR) MODÉLISATION ET TEST DE SÉCURITÉ DE BOÎTE NOIRE DE SYSTÈMES CYBER-PHYSIQUES
Abstract
(EN) A method of training a model of a cyber-physical system (CPS) includes: obtaining raw data relating to plant inputs, plant outputs, and external messages received by controllers of the CPS; automatically identifying, based on the inputs, a set of locations of a hybrid system, and respective destination locations to which respective members of the set of locations can transition; learning, for each location of the set of locations, based on the inputs and outputs, parameters of a continuous-time model of events of the hybrid system; and training, for each location of the set of locations, a discrete-time classifier; wherein features of the discrete-time classifier include, or are based on, the outputs and external messages; and wherein class labels of the discrete-time classifier include, or are based on, the respective destination locations.
(FR) La présente invention porte sur un procédé de formation d'un modèle d'un système cyber-physique (CPS) consistant : à obtenir des données brutes relatives à des entrées d'installation, à des sorties d'installation et à des messages externes reçus par des contrôleurs du CPS ; à identifier automatiquement, sur la base des entrées, un ensemble d'emplacements d'un système hybride, et des emplacements de destination respectifs où des éléments respectifs de l'ensemble d'emplacements peuvent effectuer une transition ; à apprendre, pour chaque emplacement de l'ensemble d'emplacements et sur la base des entrées et des sorties, des paramètres d'un modèle en temps continu d'événements du système hybride ; et à former, pour chaque emplacement de l'ensemble d'emplacements, un classificateur à temps discret, des caractéristiques du classificateur à temps discret comprenant ou étant basé sur les sorties et les messages externes, et des étiquettes de classes du classificateur à temps discret comprenant ou étant basées sur les emplacements de destination respectifs.
Related patent documents
Latest bibliographic data on file with the International Bureau