Название: Intelligent Secure Trustable Things
Автор: Michael Karner, Johannes Peltola, Michael Jerne, Lukas Kulas
Издательство: Springer
Серия: Studies in Computational Intelligence
Год: 2024
Страниц: 446
Язык: английский
Формат: pdf (true)
Размер: 16.4 MB

This book provides an overview about results of the InSecTT project. Artificial Intelligence of Things (AIoT) is the natural evolution for both Artificial Intelligence (AI) and Internet of Things (IoT) because they are mutually beneficial. AI increases the value of the IoT through machine learning by transforming the data into useful information, while the IoT increases the value of AI through connectivity and data exchange. Therefore, InSecTT—Intelligent Secure Trustable Things, a pan-European effort with over 50 key partners from 12 countries (EU and Turkey), provides intelligent, secure and trustworthy systems for industrial applications to provide comprehensive cost-efficient solutions of intelligent, end-to-end secure, trustworthy connectivity and interoperability to bring the Internet of Things and Artificial Intelligence together. InSecTT creates trust in AI-based intelligent systems and solutions as a major part of the AIoT. InSecTT fosters cooperation between big industrial players from various domains, a number of highly innovative SMEs distributed all over Europe and cutting-edge research organizations and universities. The project features a big variety of industry-driven use cases embedded into various application domains where Europe is in a leading position, i.e., smart infrastructure, building, manufacturing, automotive, aeronautics, railway, urban public transport, maritime as well as health. The demonstration of InSecTT solutions in well-known real-world environments like airports, trains, ports and the health sector shows their applicability on both high and broad level, going from citizens to European stakeholders.

The Internet of Things (IoT) is a revolutionary change for many sectors: Fitness trackers measure our movements, smart fire extinguishers monitor their own readiness for action, and cars turned out to become fully connected vehicles. The availability of the collected data goes hand in hand with the development of Artificial Intelligence (AI) and Machine Learning (ML) algorithms to process them. Despite numerous benefits, the vulnerability of these devices in terms of security remains an issue. Hacks of webcams, printers, children’s toys, and even vacuum cleaners as well as Distributed Denial-of-service (DDoS) attacks reduce confidence in this technology. Users are also challenged to understand and trust their increasingly complex and smart devices, sometimes resulting in mistrust, usage hesitation and even rejection.

These developments mostly cover processing of data in centralized Cloud locations and hence cannot be used for applications where milliseconds matter or for safety–critical applications. By moving AI to the Edge, i.e., processing data locally on a hardware device, real-time applications for self-driving cars, robots and many other areas in industry can be enabled.

The development of AI-based systems in many IoT domains comes across security challenges that are especially important given that these systems are usually performing critical tasks with the help of sensitive data. Since almost a decade, the machine learning (ML) security community (mainly structured on Adversarial Machine Learning and Privacy-Preserving Machine Learning) works unceasingly with a threefold objective: (1) turn the spotlights on attacks that target every step of the machine learning pipeline with an impressive diversity of attack vectors, (2) propose defense schemes to improve the robustness of the models or the systems, (3) build sound evaluation methodologies to properly assess the intrinsic robustness of models or the real impact of protections.

However, most of these works are focused on demonstrating (or defending against) attacks that exploit the inputs and the outputs of a white or black-box target model seen as a pure algorithmic abstraction. Obviously, these studies are compulsory since they enable to reveal theoretical flaws, but the attack surface needs to encompass attack vectors related to the physical implementation of the models on specific hardware platforms. Interestingly, one can draw a parallel with cryptography-based systems for which international standardization and certification are well established. For example, if we consider the actual standard for symmetric encryption (AES–Advanced Encryption Standard), this algorithm is known to be secure as no cryptanalysis-based attack has been proven (unless the brute-force strategy). However, several physical attacks (typically, side-channel and fault injection analysis) have been successfully demonstrated on many platforms (ASIC, FPGA, microcontrollers...) to recover the secret key.

The first part of the book provides an introduction into the main topics of the InSecTT project: How to bring Internet of Things and Artificial Intelligence together to form the Artificial Intelligence of Things, a reference architecture for such kind of systems and how to develop trustworthy, ethical AI systems. In the second part, we show the development of essential technologies for creating trustworthy AIoT systems. The third part of the book is composed of a broad variety of examples on how to design, develop and validate trustworthy AIoT systems for industrial applications (including automotive, avionics, smart infrastructure, health care, manufacturing and railway).

Contents: