Showing papers by "Stefan Pickl published in 2023"
TL;DR: In this article , the authors present a structured data set of physical attacks against railways and analyze the data regarding the used means, the railway system's target component, the attacker type, and the geographical distribution of attacks.
Abstract: Recent attacks encouraged public interest in physical security for railways. Knowing about and learning from previous attacks is necessary to secure against them. This paper presents a structured data set of physical attacks against railways. We analyze the data regarding the used means, the railway system's target component, the attacker type, and the geographical distribution of attacks. The results indicate a growing heterogeneity of observed attacks in the recent decade compared to the previous decades and centuries, making protecting railways more complex.
01 Jan 2023
TL;DR: In this paper , the authors proposed a hybrid risk analysis framework based on a hyper-forest of multi-trees, which combines the elements of space, time, objects at risk, goals, and actors into a single structure.
Abstract: Changes within the worldwide security environment proceed to challenge
our ability to comprehend and react to the constantly changing hybrid
threats that are becoming more diverse, emanating from a wide range of
actors who are enabled by technology. Actors can wield an array of means and
ways to further their security interests at the expense of a target and are
able to do so without being identified.Developing proper situational
awareness is a first and crucial step on the road to achieving better
protection against hybrid threats. Here we propose a novel framework for
hybrid risk analysis that enables the better understanding of operations of
the adversary before their taking place.The idea of the framework is based
on the model of hybrid operations, which combine the elements of space,
time, objects at risk, goals, and actors into a single structure - a
hyper-forest of multi-trees.Taking into account that hybrid operations are
carried out according to certain scenarios characterized by repeatability of
tools in relation to certain goals, we propose using case-based reasoning
approach based on calculating the dynamic similarity of the information
structure of ongoing attack to retrospective sequences of hybrid attacks for
which the goals, tools, and methods are known. Retrospective data is stored
in the case base.The proposed framework combines several models and methods,
the main of which are the multi-tree model of hybrid attack representation,
spatially distributed model of hybrid attack distribution, and the method
for hybrid risk analysis. The method for hybrid risk analysis is based on
two additional models such as vulnerability model and consequences
assessment model that are developed for each type of object at risk.The
suggested framework for hybrid risk analysis offers a better comprehension
of adversary operations prior to them occurring and aids in formulating an
appropriate reaction to the changing scenario.