InSecTT: Intelligent Secure Trustable Things

For the InSecTT project, Capgemini is working on the wireless platooning communication enhanced by 5G. During this project a system level simulation was created which can simulate the 5G network in presence of mobility scenarios such as platooning. V2X can take advantage of the 5G networks which are expected to overcome the limitations of mobile communication infrastructures supporting multiple network slices. This system has been integrated with an orchestration and management platform, named ONAP, where the objective is to prepare the network by allocation resources for this type of scenarios which require high reliability and low latencies.

Within the InSecTT project, Aeroporti di Puglia works on the application of bio-metric and sensing technologies at airport passenger terminals, in order to provide a faster and effective recognition system, to detect dangerous situations and to track anomalies on the move.

Moreover, Aeroporti di Puglia also works on the application of an electromagnetic sensing network suitable for the monitoring of drainage channels crossing the airport grounds.

ACCIONA works within the InSecTT project on a use case for improving monitoring and management of large civil infrastructure construction projects.

The use case is based on the application of IoT systems to enable instant tracking of workers, machinery, and other construction assets. Advanced processing techniques are then applied to the data collected on site by these systems, to generate relevant information for different project stakeholders.

The demonstration of the use case is carried out in real tunneling construction projects of ACCIONA, including one tunnel built with conventional methods (drill & blast), and another tunnel built with a Tunnel Boring Machine (TBM).

The demonstration integrates three different scenarios, aimed respectively to the automated monitoring of the production, the advanced management of workers safety, and the optimization of machinery maintenance.

One of the main challenges faced by large civil infrastructure construction projects (e.g., construction of highways, railways, tunnels, etc.) is to provide stakeholders with up-to-date information about the actual progress of construction tasks. This is needed to measure productivity and for early detection of deviations from original project planning.

The complexity of such challenge derives from the large areas where these projects are executed, the large number and variety of machinery and workers involved, and the difficulty of deploying monitoring technologies in these environments.

Within the InSecTT project, ACCIONA is addressing this challenge for a specific scenario of automated productivity measurement in the construction of tunnels through conventional methods (drill & blast). This is achieved through advanced processing of data collected by IoT-based workers and machinery tracking systems, and electricity consumption monitoring. Thus, it is possible to detect and automatically measure the different tasks within each tunnel excavation cycle, and to quantify the resources used for each task.

Throughout the last two years, Virtual Vehicle conducted a series of workshops to explore how to develop AI applications that are perceived as trustworthy from the user´s perspective. These workshops led to a model for how to integrate humans with systems to achieve trustworthiness.

Early on, we analyzed existing recommendations in EU Ethics Guidelines and ISO specifications, as well as the proposed legislation in the EU AI Act. These recommendations were discussed in a sequence of workshops to identify trustworthiness risks in individual InSecTT use cases and to find approaches to mitigate them in development. The resulting human systems integration model for trustworthy AI is intended to enable the potential of AI applications toward accepted use and uptake within the consortium and beyond.

​Within project InSecTT, a vehicle communication systems for automated vehicles was developed, called vehicleCAPTAIN. It is intended to release core software as free and open source - stay tuned!

​Vemco is a Polish partner of InsecTT mainly focused at producing Access Control and Time Attendance systems. In InSecTT Vemco works on Physical Security Information Management (PSIM) platform responsible for integrating muliple security and location based systems at once. The platform aims to help companies monitor their facilities from a single interface. On the graphic below we present exemplary integrations reached within InSecTT project. Each integration allows to increase situational awareness, showing what is happening simultaneously in multiple locations. PSIM uses GeoJSON file formats for areas / buildings plans, which makes it easier to imitate the real environment of the monitored facility. The integrations happen in Poland, Turkey, Finland, and Italy.

In InSecTT, Vemco is mainly focused on building Physical Security Management Platform (PSIM). PSIM allows to integrate multiple systems used by single company in one platform. Integrated systems may be developed for security and safety purposes, localization-based, detection, etc. The important functionality of PSIM is allowing operator / administrative user to observe improper on-facility behaviors / devices functioning, by receiving alarms and reacting to them.

On the graphic below a common situation is presented in which integrated system is used in a PSIM interface. In the beginning integrated system detects an anomaly which is then transported via MQTT / AMQP protocol to RabbitMQ message broker. From there, PSIM subscribing service receives detection message and informs the operator about potential threat. The operator is able to raise an alarm and observe the whole situation from the point of view of other integrated systems during detection time.

ISS RFID sp. z o.o. within the InSecTT project works on Real-Time Inventory Management System that can increase both boat and crew safety. Necessary equipment like safety vests, fire extinguishers and other is equipped with RF tags that report that they are on board and ready to be used in case of emergency.

By using RF tags it can monitor in real-time if assets are still on board or if they are valid to be used. By using RF tags combined with security solutions, system will also alert unauthorized use of equipment.

When safety equipment like vest will have tags installed it can also inform when someone is overboard.

Gdansk Tech within the InSecTT project works on Multimodal Positioning System (MPS), which enables indoor localization -  the system consists of ESPAR antennas and Bluetooth Low Energy transmitters. Currently, MPS is installed in A&E (Accident and Emergency) zone in Copernicus Medical Facility located in Gdańsk, where medical equipment e.g. wheelchairs are being tracked.

Westermo is proud to be one of the partners in the InSecTT project – Intelligent Secure Trustable Things. Together with RISE, MDU, and TietoEvry, Westermo is collaborating to improve cyber security for industrial data communication systems using Artificial Intelligence.

One of the ongoing activities within the project is to collect a data set with network intrusions and human misconfigurations. AI is constantly learning and to improve there is a need for AI to learn new information. The data set will train distributed or federated AI in misconfigurations and cyber-attacks. The ambition is to use AI to take steps forward in the domain of cyber security for industrial data communication networks.

As a partner in the InSecTT project, Westermo invest in designing, implementing and quality assuring a new container feature. Several Westermo products run the Westermo Operating System (WeOS), which is a critical part of many switches and routers in industrial applications. The new container feature enables many different types of applications to run in the devices.

The next step for the new container feature is to implement Artificial Intelligence (AI) to improve cyber security. By implementing AI the mission is to detect anomalies such as intrusions and misconfigurations. By connecting AI with the alarm system in WeOS, there are many promising possibilities in the future when creating secure industrial communication systems.

To read about Westermo's contribution to the InSecTT project, please refer to Secure Industrial Communications System at: https://www.insectt.eu/use-cases/

Whitepaper: Development of Ethical and Trustworthy AI Systems

The whitepaper written by Peter Moertl & Nikolai Ebinger (Virtual Vehicle) provides processes and guidelines to develop AI that is trustworthy from a user´s perspective. Building on current guidelines and processes, the whitepaper proposes a human-system-integration (HSI) framework to define the required organizational structures. The whitepaper is available here.

In this episode, Anamarija talks with Prof. Lukasz Kulas from Gdansk University about smart ideas for using secure connected things in real life. For example, about retrofitting ships and harbors, or localizing medical devices in a hospital. Lukasz has also organized Open Innovation and Student Contests, and talks about how creativity can lead to cool innovations. These are good examples for bringing students, scholars and industry together to spark new ideas (and have fun).

Go to https://podcasts.apple.com/at/podcast/project-insectt/id1605747720

MLOps is a set of practices that aims to deploy and maintain machine learning models in production reliably and efficiently (source: https://en.wikipedia.org/wiki/MLOps).

In InSecTT project Wapice is researching MLOps frameworks for managing intelligent algorithms throughout their lifetime. Key research areas are to find solutions that

By improving methodology how to effectively and easily deploy and host algorithms in different IoT ecosystem components we can respond to the demand of rapid prototyping of AI and machine learning concepts.

​As vehicle platforms modernize, OEMs are integrating the capability to remotely update the vehicle’s electronic components software. Although OTA updates have a lot of benefits, they also widen the attack surface of a vehicle. OEMs must ensure the security of this process as more and more regulatory bodies are mandating vehicular cybersecurity. The OTA solution developed by VORTEX follows the Uptane standard, which is the current state of the art for vehicular updates. It implements the full E2E stack, including secondary ECUs. In InSecTT, VORTEX works on extending the trust-chain to other elements of the ecosystem and hardening the OTA update process as a whole.

#OTA #OEM #automotive #vehicle

In a world pushing towards connected everything, it is more and more critical to ensure that IoT devices are secure. The InSecTT project's partners are working in solutions to create secure, safe, reliable, and trustworthy things. Trustworthiness means you can be at ease while using domotics (the control of domestic appliances by electronically controlled systems) in your house, or while checking your car's status on the app. It also means a push to Industry 4.0 with process automation and increased efficiency, and even more quality control with less defects. The InSecTT project aims at providing solutions for these environments.

#safety #security #iot #industry4.0 #vortex #insectt

Innovation is one of the cornerstones of research collaborations!

The innovative hybrid--remote and in-person--research collaboration between Gdansk University of Technology (GUT) and RISE Research Institutes of Sweden within EU Project InSecTT has led successfully to fostering intra- and inter-project research collaborations. Augmenting the jamming mitigation technology for reconfigurable antennas--piloted at a number of safety-critical industrial use cases--with an intelligence-driven optimization technique for online decision-making, which led to unprecedented desired results; kicking start the development of the ML-assisted variant of the existing localization technology to make it smarter and more accurate, and concertizing more steps of the data analysis pipeline for TUCANA autonomous boat in the Polish Maritime industry, are among the early achievements of their collaborative works.

The collaboration is led by the head of the digital technology center at GUT, Prof. Łukasz Kulas, and driven by Dr. Mahshid Helali Moghadam from RISE and Mateusz Rzymowski and Mateusz Groth from GUT.

​On the InSecTT project, Vortex is researching towards safe platooning applications. Platoons are a "train" of (usually trucks) that work as a single unit, where all the elements share a portion of the route, and keep themselves at the same speed and inter-truck distance. This technology promises fuel saves up to 20%, as well as reducing the fatigue on follower trucks since the driver only has to react to alerts. In the project, Vortex is developing a framework that monitors maneuver execution such as the trucks have to change lanes or overtake another vehicle, that these maneuvers are executed correctly.

#safety #V2X

A group of researchers from University College Cork is surveying public opinion on the ethics of #smart #wearables. Their survey gathers YOUR awareness, attitudes, and opinions about #smart #wearables and #ai. They would appreciate your participation at https://www.surveymonkey.com/r/7HJQVPC.

​Ken Brown of UCC gave on October 5th 2022 the keynote talk on AI and IoT for supporting humans, as part of the event "Augmenting humans with AI-driven knowledge" at VTT, Espoo, Finland.

October is Cyber Security Month 2022. This year's themes are phishing and ransomeware. UCC is one of many participants, offering talks on cyber security in UCC, technology assisted abuse and much more.

#CyberSecMonth

#ThinkB4UClick

The InSecTT project has started mid-2020. After a virtual-only review last year, the InSecTT consortium is happy (look at the faces!) to meet again in person in Brussels, Belgium at the JU to do the review of the 2nd year results. After two days fully packed with presentations and demonstrations we can report a successful Y2 Review. A big thanks for the good feedback and advise received by the project officer @Francisco Ignacio and the reviewers @Birgitte Lønvig and @Stamatis Karnouskos

In InSecTT project, researchers of UPM investigate post-quantum cryptographic schemes for IoT edge devices, along with AI Deep learning based processing for LiDAR point-clouds, Ultra Wide Band technology for precise distance measurements and wireless sensor networks based on custom modular hardware platform for run-time train integrity. All these works are integrated in a railway domain, targeting autonomous train performing.

​Do you trust AI? How to make things "trustworthy"?

Anamarija interviews Peter Mörtl from VIF about what "trust" really is, how to make things trustworthy and what the research project InSecTT contributes through its "Trustworthyness Framework".

As InSecTT is also about developing AI methods and technologies for future applications, the question needs to be asked: can we trust AI?

Tune in using the Podcasting app of your choice, or go to https://podcasts.apple.com/at/podcast/project-insectt/id1605747720

​Guest in episode #7 of the InSecTT podcast is Michael Karner, the coordinator of project InSecTT. Michael talks about the challenges of coordinating such a large project as InSecTT, especially during the Covid-19 pandemic. What is the "secret sauce" from managing a successful project?

Get this and all previous podcasts on https://podcasts.apple.com/at/podcast/project-insectt/id1605747720 or your favorite podcast catcher

As part of InSecTT, TU Delft developed radio-frequency (RF) information harvesting, a channel sensing technique that takes advantage of the energy in the wireless medium to detect channel activity at essentially no energy cost. RF information harvesting is essential for event-driven wireless sensing applications using battery-less devices that harvest tiny amounts of energy from impromptu events, such as operating a switch, and then transmit the event notification to a one-hop gateway. As multiple such devices may concurrently detect events, coordinating access to the channel is key. RF information harvesting allows devices to break the symmetry between concurrently transmitting devices based on the harvested energy from the ongoing transmissions. To demonstrate the benefits of RF information harvesting, TU Delft integrated it in a tailor-made ultra-low power hardware MAC protocol called Radio Frequency-Distance Packet Queuing (RF-DiPaQ) and built a hardware/software prototype of RF-DiPaQ to study its performance at scale. Comparing RF-DiPaQ against staple contention-based MAC protocols, It can be seen that it outperforms pure Aloha and 1-CSMA by factors of 3.55 and 1.21 respectively in throughput, while it saturates at more than double the offered load compared to 1-CSMA. As traffic increases, the energy saving of RF-DiPaQ against CSMA protocols increases, consuming 36% less energy than np-CSMA at typical offered loads.

​TU Delft has developed a wireless batteryless avionics sensor, called Hermes, that simultaneously enables piezoelectric energy harvesting as well as sensing, making it self-powered and batteryless. Using a set of piezoelectric films which flutter due to incoming wind the sensor can accurately determine the wind speed(U) and Angle of Attack(𝛼) of the incoming airflow. After conducting wind tunnel testing to characterize the generated voltage signal of these piezoelectric films, an AI algorithm for sensing and modelling the sensor dynamics has been designed and evaluated. The estimation error of 𝑈 is below 0.2 m/s and error of 𝛼 is within 1.2∘. The sensed data is communicated wirelessly in an WAIC.

Experiments were conducted inside an actual fuselage of Boeing 737 to measure its performance under realistic wireless signal attenuation and propagation conditions. The wireless sensor represents a class of new energy harvesting wireless sensors which will be fitted in aircraft of the future to improve reliability of the different systems and ensure safer flights. The sensor has been patented and the work involving the design and working of the sensor has been published in Robotics and Automation Letters 2021 as well as presented at the International Conference on Intelligent Robots and Systems (IROS) 2021.

TU Delft has developed a wearable IoT device, Heart Watch, that enables smart monitoring of body temperature and ECG of a person 24x7. The system generates synthetic ECG signals from clinical data in real-time using a dynamical systems model in conjunction with a training-free, real-time machine learning algorithm for learning the ECG generation parameters. The parameters of the trained system are then transmitted in each cycle of the ECG wave to reconstruct the original signal using the same model at the medical practitioners' location. The advantage of such a system is that only a set of parameters are transmitted to a central server instead of raw ECG data. In addition, systems can be trained to detect and classify disease conditions as one of the parameters, which can be transmitted along with signal parameters. TU Delft’s systems currently achieve an average processing time for clinical data of one second in 0.68 seconds on a microcontroller, with an RMSE error of 0.0038, for 17 parameters per ECG cycle. This system is also easy to implement, requires minimal storage (only one ECG cycle at any given time) and does not depend on offline training.​

TU Delft as part of the InSecTT project has developed a privacy-aware system called LOCI that can perform joint localization, occupancy detection, and tracking. LOCI comprise of a fusion of two sensors Passive Infra-Red (PIR) sensor employed as a depth sensor and a thermopile sensor (Melexis 32×24 array). Such a combination of sensors also enables it to localize multiple people. While this system is extensible to other applications such as user movement tracking, fall detection, it does not extract personal information that high resolution cameras do. Hence, the system is GDPR compliant. One LOCI sensor unit is sufficient to cover the area of 8m x 8m. AI algorithms estimate the position and occupancy, and machine learning techniques are employed to make this solution work seamlessly in most indoor locations.

InSecTT Project Coordinator Michael Karner (Virtual Vehicle) will present the InSecTT project and give a talk about "Bringing Internet of Things and Artificial Intelligence together – But is it Trustworthy?" at the 59th Design Automation Conference (DAC 2022). Join us on 11-July 11:00-12:30 in session "Embedded Systems and Software".

​In this episode, Anamarija talks with ​Johannes Peltola from VTT about developing AI building blocks in InSecTT. Is data the new oil? What applications are there being researched in InSecTT? And how hard is it to lead the work package #2 in such a large project?

​InSecTT Project Coordinator Michael Karner (Virtual Vehicle) will present the InSecTT project and give a talk about "Bringing Internet of Things and Artificial Intelligence together – But is it Trustworthy?" at IoT Week 2022 taking place in Dublin, Ireland. Join us at 23-June 11:00 in session "Identity, trust and privacy in an intelligent, smart IoT World. Challenges and outcomes - Session 2: AI and ML technologies as enablers for a more secure IoT", organised by projects ERATOSTHENES and ARCADIAN-IoT.

In the scope of Silicon Austria Labs work on interference tracking and prediction in wireless sensor networks, their researchers have implemented methods to track active Bluetooth Low Energy (BLE) connections over time and predict future collisions with the own network. By listening to only one BLE channel the algorithm is able to reconstruct the connection parameters of all active BLE connections with low-cost HW, which is definitely beyond state of the art. The targeted applications are:

If you are interested on this topic, example measurements of sniffed BLE connections are published as open-source dataset in:

Julian Karoliny, Thomas Blazek, Hans-Peter Bernhard, & Andreas Springer. (2022). InSecTT BLE Channel Sniff Dataset.  https://doi.org/10.5281/zenodo.6523365

Silicon Austria Labs is working on a distributed channel monitoring solution that is capable to measure interference directly as a real part of an industrial sensor network. The work is being developed as part of the Use-Case 3: Wireless Security Testing Environment for smart IoT in the #InSecTT project. This low cost solution (requires minimum and low cost hardware) enables to evaluate interference caused by other networks/devices, find unusual (intended) interference behaviour due to malicious devices, track inference, perform predictions on future channel access and apply counter measures for own network.

#H2020 #InSecTT #wirelesssystems #SAL

As part of the Building Block BB3.4 in the InSecTT project, Silicon Austria Labs is working together with Gdansk University of Technology in developing an algorithm that finds the best network structure by minimizing latency and maximizing SNR. This solution is based on QoS criteria and considers a switched beam ESPAR antenna developed by researchers at GUT. The algorithm provides a hierarchical structure of the network sorted in layers of relay nodes, where each relay node is capable to operate either in omnidirectional or directional mode.

The results indicate that by utilizing a directional mode with just as low as two switching beams can reduce the number of layers by 65% and still maintain the same QoS criteria. This solution can be especially beneficial in delay-critical applications.

See more about this topic on our recent publication “Relay-Aided Wireless Sensor Network Discovery Algorithm for Dense Industrial IoT Utilizing ESPAR Antennas” published in IEEE Internet of Things Journal.

#InSecTT #SAL #GUT

Silicon Austria Labs (SAL) is working on Multi Hypothesis Interference Tracking algorithm that is capable to detect and track periodic interference in a wireless channel. The main goal is to find different sources of interference and distinguish them by their channel access behavior. If the channel access is not random, which is for example the case for many low-power and synchronized WSNs, it will show a certain pattern, e.g., periodic channel access. This allows to detect and synchronize to these patterns which are then used to identify the source of interference, estimate the transmission frequency, predict future channel access, avoid collisions with own network.

​RISE Research Institutes of Sweden is leading a use case on "Driver Monitoring and Distraction Detection using Artificial Intelligence" in the InSecTT project. Researchers from RISE are working together with EU partners from industry, research and academia to provide a solution to improve driver distraction detection processes in order to reduce risks and accidents. Technical challenges include: (a) labelling driving distraction data efficiently without compromising privacy, (b) predicting distractions in real-time depending on edge computing.

ICSSIM framework (Developed by “#RISE” in #InSecTT Project) is to simulate customized virtual Industrial Control Systems (ICS) security testbeds, which facilitates attack/threat investigation. Through ICSSIM, realistic details and high-fidelity ICS testbeds are produced that are extendable, versatile, reproducible, low-cost, and comprehensive. (Link: https://github.com/AlirezaDehlaghi/ICSSIM)

The Docker container technology is used in ICSSIM, which enables realistic network simulation and isolates the ICS components on private kernels for a private operating system. As well as reducing the time spent developing ICS components, ICSSIM also enables physical process modeling using software and hardware.

​​​Interested in knowing how #Artificial_Intelligence could contribute to Industrial Control Systems (#ICS) safety and security? Check out our paper " Digital Twin-based Intrusion Detection for Industrial Control Systems," recently published in the 2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events.

This paper presents “#RISE” Contributions to the “#InSecTT” Project, building up a digital twin-based security framework for industrial control systems and extending its capabilities for simulation of attacks and defense mechanisms.

​Within the InSecTT project Philips Research Eindhoven works on the improvement of hospital asset management solutions using AI and IoT. The concept demonstrator (under development) will demonstrate an end-to-end implementation suitable to be deployed in a healthcare environment.

An integrated dashboard calls the RESTful API of various subsystems (length-of-stay prediction, vital signs monitors, …) to fetch associated patient data in a JSON format. A floorplan on the dashboard can show the location and status of each asset (e.g. bed) based on the FHIR operational status of the location resource. A smart IoT device is used to update the status and location in an automated way using proximity technologies. Technical challenges include:

See also: https://www.philips.co.uk/healthcare/consulting/clinical-operational-excellence/performanceflow.

​Philips Research Eindhoven is leading a use-case on “Location awareness for improved outcomes and efficient care delivery in healthcare” in the InSecTT project. Experts from Philips Research together with EU partners from both universities and industry are working together on providing new and improved solutions for tracking of casualties in Mass Casualty Incidents (MCI) and valuable or critical assets on a hospital campus.

Technical challenges include:

​In this episode, Anamarija talks with Markus Pistauer, CEO of CISC Semiconductors. Markus explains his vision for trustworthy, intelligent IoT, and how this will shape our future. We also hear some thoughts on how collaboration in such a large project works.

​Within the InSecTT project, Philips Research Eindhoven has developed a logistics IoT device with GPS for outdoor localization, a tricolor LED for indicating triage status and cellular (LTE-M) communication for having bi-directional connection to a cloud server anytime and anywhere.

An NFC tag (or QR code) on the device allows for linking the battery operated IoT device to a patient, a caregiver or a hospital asset. The programmable IoT device supports TCP, UDP or HTTPS protocol using the standardized geoJSON format (RFC 7946) to report the device’s location. In the absence of a GPS signal the device will report its location based on the nearest cell location. The device is being used for concept evaluation in health care use-cases. Technical challenges include:

​Philips Research Eindhoven is leading a use-case on “Smart and adaptive connected solutions across health continuum” in the InSecTT project. Experts from Philips Research together with EU partners from both universities and industry are working together on providing an AI and IoT driven solution to improve a hospital workflow in order to reduce length-of-stay (LOS) and healthcare costs. Technical challenges include:

​In this Podcast, Peter (substituting for Anamarija) interviews Ramiro Robles from ISEP.

What is a "high level architecture"? How is it used in such a large project like InSecTT? And why is trustworthy, intelligent IoT important for airplanes? Ramiro has the answers....

see https://podcasts.apple.com/at/podcast/ramiro-robles-from-isep-portugal/id1605747720?i=1000558443683

Interested in #explainable-AI techniques? Read the #whitepaper from NXP Netherlands on Ethical framework and European directives: THE MORALS OF ALGORITHMS

​The InSecTT consortium is meeting again! Nearly 40 partners are on-site and more than 60 people are joining online the hybrid sessions from all  Workpackages, Use Cases and other domains.

The meeting takes place in the impressive city Istanbul hosted by our partner Marmara University – VeNIT LAB. Great to be here, to meet all partners in person or online and fostering the good collaboration, looking for new opportunities and solutions and have open discussion.

Just for the sake of completeness – we are smiling under our masks!

In InSecTT, NXP Semiconductors Austriahas set a focus on Ultra-Wideband (UWB) related innovations and deployment. The integration of a state-of-the-art UWB system comes with a variety  of challenges, one of them being system scalability. In smart health care, hundreds of patients’ ad lab devices need to be localized in a highly-secure and reliable manner to allow seamless access and easier logistics inside a hospital. Therefore, NXP Austria is developing a UWB systems simulator that can be used to evaluate any use case focusing on system reliability, stability, performance and scalability. Hence, the work in InSecTT will significantly drive a simple and reliable use case evaluation without the need to build up a real hardware demonstrator in the first concept evaluation stage.

As a partner for InSecTT project, Nurd is also focusing on simulating attack scenario cases over the network. A predetermined set of attack cases was decided for the project in order to be implemented. These attack scenarios mainly consist of Denial of Service, Man in the Middle, Probe, Auth and Brute Force attacks etc. Implementing these attack scenarios is really important because these simulation cases will be used to check the competence of our deep learning based anomaly detection module. These cases will be also used for labelled data for our training purposes. Since we intend to use supervised learning for our anomaly detection cases, we need labelled data. At this part of the project, we are currently trying to create our own data set.

​We are proud to announce that our third InSecTT Podcast is now online! In this podcast, Anamarija interviews Michael Jerne from NXP. Michael talks about the complexities and benefits of managing a large technical work package in InSecTT.

We expect to publish a new interview every 4 weeks. All podcasts will be hosted on https://www.spreaker.com/show/project-insect and of course on https://podcasts.apple.com/us/podcast/insectt-what-about-automotive-security. Please rate it and provide feedback!

NURD focuses R&D work in InSecTT on authentication, authorization and anomaly detection over OPC UA networks. For authentication process, our implementation only accepts approved and valid X.509 certificates when connecting an OPC UA server. Using this, every incoming connection will be checked whether it is coming from valid sources or not. Any client that does not have authorization on server will be rejected by the certificate management system. System also encrypts the network after authentication, just like TLS, in order to protect private and sensitive data between client and server applications over the network. It is also important to select the correct type of encryption algorithm and mode of operation in order not to compromise the system.

​Inside the InSecTT project, Nurd will focus on anomaly detection over OPC UA protocol. This anomaly detection routine will be implemented using deep learning and trained with supervised learning. Since deep learning offers a lot of potential for cases like network traffic monitoring, we have decided to use it for detecting any anomalous activity. However, in order to use supervised learning, we have decided on several attack simulation cases which will be implemented to check the competence of the system. These attack scenarios consist of DoS, MitM, Auth Attacks etc. The network packets of these attack simulations will be labelled and according to the captured data, we will train the deep learning model.

​Within the scope of the InSecTT Project (www.insectt.eu), the PhD candidate Erislandy Mozo is defending his doctoral thesis next 7th of April at the EPS-MU. The doctoral thesis named “Evaluation and enhancement of IEEE802.11 amendments for the connected train”, is focused on the research of 802.11p/bd protocol performance to support Train-to-Train (T2T) and Train-to-Infrastructure (T2I) communication links. In addition, the T2T channel modelling based on measurement is carried out in order to provide more realistic simulation results. The T2T and T2I communications enable traffic-safety-related services such as Trustable Warning Systems (TWS), virtual coupling, and Communication Based Train Control (CBTC) systems. The main results of his thesis stamm from the activities carried out in the Use Case 5.7 Intelligent Transportation for Smart Cities and Use Case 5.8 Intelligent Automation Services for Smart Transportation.

​

​https://www.mondragon.edu/en/phd/delivered-theses

#Railway #WirelessSystems #H2020 #PhDThesis

PS-MU is working on vehicular communications in the scope of the InSecTT Project. Different V2X systems are being integrated in a single platform to increase the reliability of the wireless communications. This system shares traffic-safety information among the different actors of the roads, such as vehicles, infrastructures, pedestrians etc. In addition, manages multiple radio technologies: ITS-G5 and C-V2X. Klas is collaborating in the task providing the integration platform TRX R6. A measurement campaign is planned for April to test the performance of the device .

Follow InSecTT project and find more at: www.insectt.eu

#IoT #automotive #wirelesssystems #V2X #TRX6 #MU

​VeNIT Lab (Marmara University) in InSecTT Project is doing R&D on unsupervised anomaly detection algorithms to provide safe operations and detection of anomalies on IoT systems in real-time. Timely evaluation and partitioning of the data are important to address network-related issues correctly. Resilience to outliers, feedback, and filtering methods are investigated to provide error-prone anomaly detection for network performance monitoring towards reliable communication in various domains such as manufacturing and smart environmental monitoring.

Follow InSecTT project and find more at: www.insectt.eu & venit.org

#AI #IoT #manufacturing #anomaly #reliability

​In InSecTT project (H2020-ECSEL), VeNIT Lab (Marmara University) is contributing to increasing traffic safety with object detection and V2X communication solutions. VeNIT Lab is developing an end-to-end system that detects and tracks 2D objects in traffic including vehicles and vulnerable road users (VRU) with the streaming data of camera feeds in real-time. Combining AI-based detection with V2X communication, it is possible to provide detected object information and create situational awareness of the surroundings to nearby vehicles when their sensors may be insufficient to provide that information.

Follow InSecTT project and find more at: www.insectt.eu & venit.org

#ECSELJU #H2020 #ITS #AI #artificialintelligence #safety #trafficsafety #transportation #objectdetection #trafficmanagement #traffic #mobility #GPS #map #machinelearning #venitlab @VeNIT Lab

​The ever-growing human population and the mobility requirements cause a dramatic increase in the volume of traffic in transportation. There is a need for a total perception of the environment to create more trustable, organized use of transport networks with intelligent systems. Nowadays, the 3D detection of objects from traditional camera inputs is one of the most popular and low-cost ways of perceiving an environment.

In InSecTT project, VeNIT Lab (at Marmara University) is building innovative solutions on the detection of objects from surveillance camera inputs at intersections for Intelligent Transportation Systems (ITS). With the cameras, detailed information of the objects (their distance, orientation, dimensions, etc.) is extracted and transferred to nearby vehicles for creating situational awareness of the surroundings and increasing safety. Such actions are taken to prevent traffic collisions and to monitor traffic at intersections.

Follow InSecTT project and find more at: www.insectt.eu & www.venit.org

#ECSELJU #H2020 #vulnerableroadusers #IoT #AI #aiot #V2X #ITS #objectdetection #transportation #safety #traffic #mobility #connectedvehicles #venitlab @VeNIT Lab

​VeNIT Lab (at Marmara University) is doing R&D on dynamically generating up-to-date regional road infrastructure maps in InSecTT project by analyzing large-scale multi-trajectory GPS data and developing AI-assisted methods/procedures.

Digital road infrastructure maps are utilized in Intelligent Transportation Systems (ITS) for reliable navigation and optimal route estimation. Currently available digital maps are updated manually by a long and laborious process and thus they may fail to portray the recent or temporary changes in real-life road networks such as closures or accidents. To address this problem, GPS data collected from recent traffic activity can be used for inferring commonly used routes and generating up-to-date digital maps. Generated digital maps can be local alternatives to map services when these services are not available. In addition, a map generation system based on GPS data can also be used for uncovering the maps of unmapped territories and commonly traveled unofficial paths such as routes within forests, rural areas, and construction sites.

Follow InSecTT project and find more at: www.insectt.eu & venit.org

#ECSELJU #H2020 #ITS #AI #trafficsafety #transportation #traffic #mobility #GPS #map #machinelearning #venitlab @VeNIT Lab

​Jamming is a renowned threat to the wireless community and a crucial issue in real-time and mission/security-critical applications. Real-time jamming detection is extremely important to counteract and prevent jamming attacks in IoT devices. Additionally, the solution needs to use no extensive computational resources and should be deployable in IoT devices to detect and safely prevent faulty when the wireless communication is interrupted.

VeNIT Lab (at Marmara University) in InSecTT (Intelligent Secure Trustable Things) project is building innovative AI-based security attack and jamming detection mechanisms to increase safety and security in manufacturing operations. Collected data related to the network traffic and wireless link quality is used to find related parameters and anomalies correlated to security and jamming attacks. With the security attack/jamming test setup in VeNIT Lab, several security attack/jamming types with various scenarios have been experimented, analyzed, and assessed. All these are used to derive innovative solutions to understand-detect-prevent security/jamming attacks.

Follow InSecTT project and find more at: www.insectt.eu & venit.org

#ECSELJU #H2020 #iot #AI #aiot #bigdata #monitoring #reliability #smartfactories #manufacturing #cybersecurity #cyberphysicalsystems #wirelesscommunication #industry40 #anomalydetection #edgecomputing #machinelearning #venitlab @VeNIT Lab

​VeNIT Lab (Marmara University) is doing R&D and building innovative solutions for real-time object detection and tracking with lightweight architectures and commercially available IoT edge devices to analyze the applicability of low-cost object detection solutions for the industry. AI-based detection is being adapted to the environment over time by collecting feedback data and enhancing the models on the cloud for further training and analysis purposes while edge devices with limited sources focus on detection and spreading of the detected objects for awareness. Interoperable algorithms offer an innovative approach in terms of traffic conditions, enhanced situation awareness, road safety and traffic efficiency.

Follow InSecTT project and find more at: www.insectt.eu & venit.org

#ECSELJU #H2020 #ITS #iot #AI #aiot #monitoring #trafficsafety #objectdetection #transportation #traffic #mobility #anomalydetection #edgecomputing #machinelearning #venitlab @VeNIT Lab

​In InSecTT project, VeNIT Lab (at Marmara University) is working on the visualization of network traffic, service quality, and service availability with the data gathered from IoT devices in addition to building a real-time data collection platform. Real-time graphs, charts, and alerts are generated to increase user perception and indicate network-related issues. It helps to maintain reliable communication between entities and to find root causes for problems in the network. Access to real-time and historical data enables comprehensive analysis and perception on different states of the network over time and exploits seasonality, trend, and anomalies that occurred in the network.

Follow InSecTT project and find more at: www.insectt.eu & venit.org

#ECSELJU #H2020 #iot #AI #aiot #bigdata #monitoring #visualization #reliability #cybersecurity #cyberphysicalsystems #wirelesscommunication #industry40 #anomalydetection #machinelearning #venitlab @VeNIT Lab

​V2X is based on wireless communication used to increase road safety & efficiency and driving comfort through the cooperation of the entities in the traffic. As a key partner of InSecTT, VeNIT Lab (Marmara University) is developing solutions to increase road safety & traffic efficiency by taking advantage of Connected Cars/V2X Digital Twin Platform and the Connected Cars Service Platform (CCSP). VeNIT Lab’s end-to-end complete solutions provide enhanced situation awareness for the drivers and provide intelligent services for the stakeholders in C-ITS. Key components are below:

Follow InSecTT project and find more at: www.insectt.eu & www.venit.org

#ECSELJU #H2020 #testing #digitaltwin #its #V2X #cybersecurity #cyberphysicalsystems #connectedvehicles #connectedcars #connectedcar #venitlab @VeNIT Lab

​Real-time data monitoring requires continuous streaming of data and processing in real-time. Furthermore, the monitoring system must provide the right visualization and analysis tools that lead to increased and enhanced user perception and indicate anomalies to ensure the safety and security of the operations as well as business continuity.

VeNIT Lab (Marmara University) is an academic partner in InSecTT (Intelligent Secure Trustable Things) Project (H2020-ECSEL). VeNIT Lab is building a quality monitoring tool and an AIoT service platform that allows users to transfer, store and inspect the link quality and network performance data collected from the IoT devices in the operating environment. The data is monitored and analyzed for the reliability and availability of the wireless links and IoT services. The network performance analysis and statistics are also visualised along with the alerts and warnings.

VeNIT lab is developing AI-based (machine learning) algorithms and methods to detect anomalies in the collected data to identify malicious attacks and jamming. This tool is already contributing to various application domains including transport, automation, manufacturing, cybersecurity, and healthcare. The lightweight features of ML algorithms enable the use of them in IoT and edge devices as well.

Follow InSecTT project and find more at: www.insectt.eu & www.venit.org

#ECSELJU #H2020 #iot #AI #aiot #bigdata #monitoring #reliability #smartfactories #manufacturing #cybersecurity #cyberphysicalsystems #wirelesscommunication #industry40 #anomalydetection #edgecomputing #machinelearning #venitlab @VeNIT Lab

Mäladalen University (MDU)  researchers have proposed an overall approach for a fault- and threat tolerant platooning for materials transportation in production environments using automated guided vehicles (AGV). Our functional use cases include the platoon control for collision avoidance, data acquisition and processing by considering range, and connectivity with fog and cloud levels. To perform the safety and security analyses, the Hazard and Operability (HAZOP) and Threat and Operability (THROP) techniques are used. Based on the results obtained from them, the safety and security requirements are derived for the identification and prevention/mitigation of potential platooning hazards, threats and vulnerabilities. The assurance cases are constructed to show the acceptable safety and security of materials transportation using AGV platooning. We leveraged a simulation-based digital twin for performing the verification and validation as well as fine tuning of the platooning strategy. Simulation data is gathered from digital twin to monitor platoon operations, identify unexpected or incorrect behaviour, evaluate the potential implications, trigger control actions to resolve them, and continuously update assurance cases. The applicability of the AGV platooning is demonstrated in the context of a quarry site.

A journal article detailing the work is published in the Journal of Systems Architecture, December 2021

DOI : https://doi.org/10.1016/j.sysarc.2021.102309

​​We are happy to announce that our 2nd podcast is online ... use your favorite podcast catcher or go directly to https://podcasts.apple.com/at/podcast/project-insectt/id1605747720

​In this podcast, Anamarija interviews Andreas Springer, head of the Institute for Communications Engineering and RF-Systems at JKU Linz. Today he will talk about JKU's involvement in InSecTT, and what a trustworthiness indicator has to do with ultra-wide band (UWB) radios. This work is part of a joined European research project InSecTT: Intelligent Secure Trustable Things, https://www.insectt.eu/. The project has received funding from the ECSEL Joint Undertaking (JU) under grant agreement No 876038. The JU receives support from the European Union’s Horizon 2020 research and innovation programme and Austria, Sweden, Spain, Italy, France, Portugal, Ireland, Finland, Slovenia, Poland, Netherlands, Turkey. The podcast reflects only the interviewee's view and the Commission is not responsible for any use that may be made of the information it contains.

​Smart environment monitoring systems are composed of physical elements such as IoT & edge devices and non-physical elements such as data, connectivity, and software applications. These systems allow users to track, monitor, and control the operation environment & processes for safety, security, efficiency, utilization, and emergency cases. Anomalies and emergency events that would adversely affect are detected in real-time, and indeed, can be estimated earlier based on the learned patterns.

With the thriving information depending on the volume and the amount of data obtained from the environment, VeNIT Lab (at Marmara University) is transforming data into value by developing monitoring platforms and tools for various application domains. Through these tools with the implementation scope from “data collection” to the “data visualization” aspects, real-time data for link quality, network performance, and many more are collected and visualized while alerts and warnings are generated based on connectivity issues and anomalies at the control center for the operator.

Follow InSecTT project and find more at: www.insectt.eu & www.venit.org

#ECSELJU #H2020 #iot #AI #aiot #bigdata #monitoring #reliability #smartfactories #manufacturing #network #cybersecurity #cyberphysicalsystems #venitlab @VeNIT Lab

​Digital Twins take part in digital processes as virtual copies of cyber-physical systems for testing and demonstration. Test data from a digital twin is collected and analyzed to observe systems behavior. It adds repeatability and decreases the cost of tests meanwhile digitalizing the test process.

As a partner of InSecTT, VeNIT Lab (at Marmara University) is building a Connected Cars/V2X Digital Twin Platform and a framework for testing V2X communication and ITS applications. Digital Twin provides a large-scale testbed suitable for creating multiple virtual vehicles and running V2X applications on each of them. It enables the testing of software and applications developed to ensure the security of vehicle systems and wireless communication between vehicles. This plays an important role in testing and realizing the applicability of cybersecurity solutions for Cooperative Intelligent Transportation Systems (C-ITS).

Follow InSecTT project and find more at: www.insectt.eu & www.venit.org

#V2X #ITS #testing #digitaltwin #ECSELJU #H2020 #cybersecurity #cyberphysicalsystems #connectedvehicles #connectedcars #connectedcar #venitlab @VeNIT Lab

​Connectivity is one of the main requirements for IoT systems for collecting data and monitoring operations. Within InSecTT project, VeNIT Lab (at Marmara University) is building innovative solutions on reliable and secure communications. VeNIT Lab is working on link quality and network performance parameters to monitor and analyze the availability and reliability of wireless links. Data from different layers of the protocol stack, active and passive performance metrics, and application/hardware-specific information is collected from IoT devices and transferred to the control center for detecting and identifying anomalies in real-time with AI-based solutions.

Follow InSecTT project and find more at: www.insectt.eu & www.venit.org

#ECSELJU #H2020 #IoT #AI #aiot #network #wireless #cybersecurity #cyberphysicalsystems #anomalydetection #data #venitlab @VeNIT Lab

VeNIT Lab at Marmara University is an academic partner in InSecTT (Intelligent Secure Trustable Things) Project (H2020-ECSEL). VeNIT Lab is doing research and building innovative solutions in the fields of artificial intelligence, IoT, V2X communications/connected cars, cybersecurity, and reliable&secure communications. VeNIT Lab is contributing to 5 different use cases with 10 scenarios to bring AI and IoT together through close collaboration with partners from the academy and industry.

VeNIT Lab's innovative R&D solutions:

Follow InSecTT project and find more at: www.insectt.eu & www.venit.org

We are proud to announce that our first InSecTT Podcast is now online:

InSecTT: What about Automotive Security?

In this podcast, Anamarija interviews Stefan Marksteiner from AVL. Stefan will explain "What to do if you don’t want your car to be hacked".

We expect to publish a new interview every 4 weeks. All podcasts will be hosted on https://www.spreaker.com/show/project-insect. Please rate it and provide feedback!

The use of energy harvesting devices (EHDs), such as solar cells and thermoelectric generators, offers a promising opportunity for powering self-sufficient wireless devices. This becomes more important due to the miniaturization of electronic components and the resulting reduction in power consumption of integrated circuits. As a result, wiring and maintenance costs can be reduced or possibly omitted at all. However, the available output power of EHDs is highly dependent on the mounting location as well as on environmental conditions and may vary greatly over time. Therefore, it is meaningful to evaluate the EHDs at the location of use over a certain period of time in order to characterize them in real world scenarios.

LCM has developed a mobile and wireless measurement system for the characterization of EHDs. The measurement system enables the acquisition of different EHDs’ characteristics for specific applications under various environmental conditions. This simplifies the selection of a suitable EHD and the design process of an energy management system for a particular application. Read more: https://doi.org/10.5162/ettc2020/1.2

Wireless embedded devices are key elements of Internet-of-Things (IoT) and industrial IoT (IIoT) applications. The complexity of these devices as well as the number of connected devices to networks increase steadily. The high intricacy of the overall system makes it error-prone and vulnerable to attacks and leads to the need to test individual parts or even the whole system.

LCM develops a flexible, modular structured, and distributed testbed is necessary to perform different tests to ensure correct operation in various operation and attack scenarios. It is based on the Robot Operating System (ROS) as communication framework to ensure modularity and expandability. The testbed integrates RF-jamming and measurement devices to evaluate remote attack scenarios and interference issues. An energy harvesting emulation cell is used to evaluate different real-world energy harvesting scenarios.

The concept of the testbed is presented in: https://doi.org/10.1109/WFCS46889.2021.9483593

​Dependable wireless communication gains its momentum as it is envisioned as a backbone of many interesting applications in various industrial domains. However, to successfully deploy such bring those there seems still some lack of confidence for applying wireless solutions to industrial processes in a wider area. These demanding research challenges will be addressed in a special session co-organized by the European research project InSecTT (Intelligent Secure Trustable Things – https://www.insectt.eu/) that will cover MIKON 2022 conference topics that may be used to bring intelligent, secure, and reliable wireless systems closer to reality.

The submissions may address the following specific topics: AI/ML for wireless transmission, including explainable AI/ML; dependable wireless communication, real-time monitoring, and response; real-time critical communication; verification and validation of reliable wireless communication.

More on MIKON 2022 can be found on https://www.linkedin.com/showcase/microwave-radar-week-mrw2022/ and https://www.linkedin.com/feed/update/urn:li:activity:6872421345695727616/

#MRW2022 #EuMA #microwaves #mikon #conferences2022

The liaison will allow a close collaboration between teams from both projects on creating intelligent, secure and trustworthy systems. We are for example planning to conduct joint technical workshops and panel discussions to discuss concepts where we can benefit from each others expertise. We are also planning to explore future opportunities, such as joint activities in standardization events and organizing joint summer school and training events.

InSecTT (https://insectt.eu) started in June 2020 and receives funding from the ECSEL Joint Undertaking (JU) under grant agreement No 876038. The JU receives support from the European Union’s Horizon 2020 research and innovation programme and Austria, Sweden, Spain, Italy, France, Portugal, Ireland, Finland, Slovenia, Poland, Netherlands and Turkey. It is coordinated by Michael Karner (Michael.karner@v2c2.at) from Virtual Vehicle.

DAIS (https://dais-project.eu) started in May 2021 and receives funding from the ECSEL Joint Undertaking (JU) under grant agreement No 101007273. The JU receives support from the European Union’s Horizon 2020 research and innovation program and Sweden, Spain, Portugal, Belgium, Germany, Slovenia, Czech Republic, Netherlands, Denmark, Norway, and Turkey. It is coordinated by Ali Balador (Ali.balador@ri.se) from RISE.

@KTH continues to add capabilities to its AD-EYE testbed. OPEN-KTH is an open Lidar dataset of the KTH campus. Watch a video about it here: https://www.adeye.se/open-kth

For an overview and advancements of the KTH AD-EYE testbed, see here: https://youtu.be/LNjM94oovSM

#automateddriving

#opendata

#intelligenttransportation

We are very pleased to announce the newly started liaison of projects InSecTT and ERATOSTHENES !

This liaison will allow a close collaboration of teams from both projects on trustworthiness in (A)IoT.

We plan joined technical workshops to discuss requirements and concepts, using each other project's expertise for mutual benefits. In addition, discussions have started to explore future opportunities for joined events, like workshops on conferences and presentations at exploitation events.

Project ERATOSTHENES (eratosthenes-project.eu) started in October 2021 and is funded under H2020-EU.3.7.4.

It is coordinated by Konstantinos Loupos (konstantinos.loupos@inlecomsystems.com) from INLECOM.

Project InSecTT (https://www.insectt.eu/) started in June 2020 and receives funding by ECSEL Joint Undertaking (JU). The JU receives support from the European Union’s Horizon 2020 research and innovation programme and Austria, Sweden, Spain, Italy, France, Portugal, Ireland, Finland, Slovenia, Poland, Netherlands, Turkey. It is coordinated by Michael Karner (Michael.Karner@v2c2.at) from Virtual Vehicle

A new course will give @KTH students the basics of one of the hottest technology subjects: smart Cyber-Physical Systems (CPS). The course is a result of collaboration between three departments at KTH and with Ericsson Research and emphasis risk and opportunities of future CPS, systems thinking and artificial intelligence as deployed in, and with, CPS.

"To be able to invest in new technology, we need people at expert- and management level with knowledge in the area," says @Martin Törngren at KTH, who has been involved in developing the course.

Read more about the new course: https://www.kth.se/indek/nyheter/framtida-ledare-far-koll-pa-cyberfysiska-system-1.1116385

During the fall 2021, KTH also developed a new course on CPS safety and cyber-security.  Safety and security are increasingly important for the design of complex technological systems, as they are becoming more intelligent, always connected and influencing the societal infrastructure at all levels.

"Current education tends to emphasize either safety or cyber-security. This course, on the other hand, gives an overview of both topics and their relationships".

The first instance of the course was given for PhD students and industry. The plan is to make it available in a modified form later for engineering education and distance education.

See more information here: https://www.kth.se/student/kurser/kurs/MF240V?l=en

#cyberphysicalsystems

#cybersecurity

#safety

#artificialintelligence

​Did you miss out on the 9th edition of the Scandinavian Conference on Systems and Software Safety (SCSSS)? Don't despair! You can read some presentations afterwards, and here is a recap of the popular event:

• @Simon Burton, @Fraunhofer IKS, addressed the safety assurance of automated driving, emphasizing how to deal with machine learning.

• @Sven E. Hammarberg, @BVR Academy & Investigations, talked about the Boeing 737 Max accidents.

• @Fredrik Törner, @Volvo Cars, presented the evolving safety standards landscape for automated driving.

Find the complete program and presentations here: http://safety.addalot.se/2021/program

Thanks to all 130 participants!

The event took place in Gothenburg on November 23–24, and was organized by the two Swedish competence networks, ICES https://www.linkedin.com/company/ices-theindustrynetwork/  at @KTH and @Safer at @Chalmers.

How to assure safety and security in edge computing applications? And how to work with machine learning for these systems while ensuring trustworthiness?

Join a hybrid workshop on December 17.

The workshop is arranged by KTH Royal Institute of Technology and is part of the 6th ACM/IEEE Symposium on Edge Computing (SEC), 2021.

Read more and register: https://www.tecosa.center.kth.se/.../sec-2021.../

#edgecomputing #machinelearning #InSecTT #H2020

Get insights on InSecTT project at the virtual booth during the #EFECS from 23 - 25 November. If you want to know how we work on our vision of bringing AI+IoT together - visit us!

#ECSELJU #H2020 #internetofthings #IoT #AI #artificialintelligence

An example use case of #InSecTT is Intelligent Automation Services for Smart Transportation, with an objective to improve the reliability, safety and security of the systems used in #railway.

#Klas #TRX R6 is the onboard compute gateway underpinning these R&D efforts.

The TRX R6 provides developers with an open architecture, with the flexibility to connect to multiple systems onboard. With #5G available on the TRX R6, researchers are assessing how #IoT and #reliableAI can make use of the next generation of mobile connectivity to deliver on the #InSecTT smart transportation objectives.

#H2020 #railway #smartinfrastructure #internetofthings #IoT

Successful #IoT implementations in #railway, will rely on data transmissions that are secure and come with zero data loss. Assuring #reliableAI can be evolved based on trusted data from remote train systems. #Klas is collaborating with #InSecTT researchers to develop intelligent routing based on the next generation of #5G mobile connectivity.

By leveraging, Klas compute gateways, developers are implementing #AI based virtual machines that ensure reliable data exchanges, required for the efficient implementation of smart, trustable, safe and secure systems for rail automation.

#H2020 #railway #smartinfrastructure #internetofthings #IoT

The InSecTT consortium meeting started - with several partners joining in person and also online. A great pleasure to meeting so many partners personally but also seeing and hearing you online: we have the best conditions for fruitful discussions! Thanks a lot to our partner Indra for hosting the meeting!

#H2020 #ECSELJU #meeting #AI #IOT

​​Communications delay is one of the most critical metrics for many applications today. Real-time applications demand low and stable delay to work disrupt-free. For example, video monitoring and analysis are a big part of today’s industrial operations. For safe and efficient use of video, transferring the content reliably and with low latency is essential, which requires continuous measurement that the latency requirements are met.

Measuring one-way delays for network traffic necessitates accurate clock synchronization between measurement points. This is sometimes hard to obtain and maintain for different reasons. #Kaitotek, a partner in #InSecTT, has developed an algorithm in the project that detects clock synchronization issues and, in real-time, corrects the measured delay values for application traffic accordingly. The solution is being integrated into their #Qosium, passive QoS/QoE measurement product.

#ECSELJU #H2020 #qos #qoe #criticalcommunications #networkdelay

​Traditionally, network quality has been measured periodically with artificial test traffic. However, when increasingly mission-critical applications and services are being used over networks, this measurement does not provide accurate and reliable enough results to determine how well the network performs. #Kaitotek, a partner in #InSecTT, has revolutionized how network performance and quality, like one-way delays, jitters, and loss, is measured and monitored. Instead of burdening networks with test traffic, Kaitotek’s #Qosium solution passively measures network quality for real active applications. The measurement can be deployed even to embedded devices, making it possible to obtain even truly end-to-end connection quality monitoring, regardless of the underlying network technologies.

The real-time measurement data provides essential data for control systems and AI mechanisms to adjust networks and applications to cope with the prevailing network conditions. The use cases include, for example, application adaptation, dynamic routing and network resource management, and security monitoring from traffic characteristics.

#ECSELJU #H2020 #smartinfrastructure #networkperformance #networkmonitoring #criticalcommunications #qos #qoe

Communications networks are the key to the modern industrial environments. Networks serve mission-critical applications that directly reflect business efficiency and operation safety.

#Kaitotek creates rich and real-time visibility for industry about their network quality and how applications experience it. The basis is the continuous network quality measurement that feeds the situation awareness solution with real-time measurement data about network Quality of Service (QoS). This data is processed for KPIs and visualized to quickly see how the network performs in terms of applications used, time, and location.

Kaitotek pilots the solution, #Qosium Storage, in the development phase in the Port of Oulu. The measurement is installed on tens of cargo handling vehicles using a private LTE network. A port operator is provided with a solution to detect changes in the port network quality before they harm/disrupt the cargo operations. The heatmap visualization shows where problems are located and which vehicles they affect. Historical data helps analyze how the network performance has evolved and taking SLA reports.

#ECSELJU #H2020 #smartinfrastructure #maritime #qos #qoe #criticalcommunications #networkmonitoring #privatenetwork

​Within the InSecTT project JKU is working on the detection of physical tamper attacks in OFDM-based communication systems with deep learning approaches using physical layer information. We focus on scenarios with static transmitters and receivers while the attacker changes antenna orientation or position of transmitter and/or receiver. Main challenge is to distinguish this attack from environmental changes like the movement of people during regular operation. Deep learning approaches are capable of solving this issue.

We have published first results in this paper(https://ieeexplore.ieee.org/document/9403404) and is working on further improving this method.

This work is performed within the #InSecTT Project.

Within the InSecTT project JKU is working on enhancing the well-known Distance Bounding protocols by adding passive verifiers that perform TDOA measurements. Distance bounding protocols ensure that a legitimate node, e.g. a wireless car key, is located within a certain distance to a proving entity, e.g. the car. This can improve the security not only for wireless car access but also in industrial environments such as factories. By using our approach, an attacker that could trick the regular distance bounding protocol, would not be able to access restricted factory areas or get access to the car because the TDOA measurements would uncover the attack.

JKU has published a paper on this topic and is working on further developing the TDOA-Enhanced Distance Bounding approach.

This work is done in cooperation with #InSecTT Project. #JKU #Research

InSecTT project will be present @ ICCV 2021 one of the most important venues worldwide for computer vision research. University of Modena and Reggio Emilia and Technische Universität München will present a joint work about the use of purely synthetic data for training efficient and effective people detection, tracking and ReID systems. Check out ICCV in October and take a look at the paper “MOTSynth: How Can Synthetic Data Help Pedestrian Detection and Tracking?” https://arxiv.org/abs/2108.09518.

Thanks and congrats to all the authors who made it possible Matteo Fabbri, Guillem Braso, Gianluca Maugeri, Orcun Cetintas, Riccardo Gasparini, Aljosa Osep, Simone Calderara, Laura Leal-Taixe, Rita Cucchiara.

Why is it that human progress often seems to be tied to polluting our environment, damaging fauna and flora? Does it have to be that way? There is a rethinking going on, joining forces to combine human well-being with environmental sustainability. #greentechvally in the south of Graz is a hotspot for innovative energy and environmental technology. So it was logical to join forces and discuss what the #insectt vision working on AI + IOT = AIOT can do for our world. I had to pleasure to present ideas from InSecTT at the Green Tech Innovators Club 12.10.2021. For more information see Das war der Green Tech Innovators Club Oktober 2021... #sustainability #iot #ai

ISEP is working on the lower layers of the emerging technology for communications on board aircraft called wireless avionics intra-communications or WAICs. The main objective is the design and improvement of the lower layers, particularly to increase the trust in this technology from the different stakeholders and general users of the aviation industry and in the end to enable the concept of “fly–by-wireless”.

The concept of “fly-by-wireless” will allow future aircraft to be completely modernized by a reliable wireless transmission infrastructure with self-healing, automatic troubleshooting and reconfiguration that will replace many of the low or medium critical cabling subsystems of an aircraft, making it lighter, more flexible and with several advantages such as increased payload, weight reduction, higher fuel efficiency, etc.

ISEP is one of the contributing partners to the use case of wireless platooning applications. The core of our contribution is on the physical and medium access control technologies based on multiple antenna systems and artificial intelligence to improve the communications between the elements of a platoon and the roadside infrastructure. The platoon of autonomous or semi-autonomous vehicles aims to be connected to cloud and edge infrastructure. This infrastructure will help, using artificial intelligence and machine learning, with improved decision-making, obstacle/risk detection, route planning and emission reduction of a set of platoons or vehicles in a smart city.

One of the objectives is to improve the experience of drivers, passengers, stakeholders, and also pedestrian users. The main idea is to use a combination of V2V and V2I technologies reducing interference and allocating resources that allow vehicles to communicate reliably and under the minimum latency bounds necessary to reduce potential safety issues. ISEP is developing improved V2V and V2I multiple antenna and propagation models that recreate the diversity found in the platoon scenarios, for example inside tunnels, with reflections and waveguide effects of tall buildings, with obstacles or scatterers and with ground reflections that will make system level simulations more accurate.

ISEP is the leader of the task of the reference architecture and reference implementation of the InSecTT project. The reference architecture is a set of guidelines and collection of best practices to accommodate, design and implement the necessary infrastructure to support dependability, security, trust and privacy in IoT use cases in different industrial domains. The reference architecture is an evolution of the predecessor architecture of the DEWI and SCOTT projects. This improved architecture targets the impact analysis of the new technology developments such as cloud, edge computing, 5G, multiple interface nodes and mainly artificial intelligence and machine learning on IoT architectures.

The reference architecture is also a collection of expertise and examples of multiple use cases of three different projects towards achieving dependable, secure, trustable and safe AI-IoT applications for industrial connectivity. The reference architecture consists of multiple views or models that allow designers to analyze the system from multiple perspectives that fit the new requirements of multiple stakeholders. The reference architecture allows us to look at use-cases from a high-level perspective, identifying functionalities, entities, interfaces, communication and processing hardware standards, potential vulnerabilities associated with each interface or functionality, and in many cases, it provides a detailed forensic analysis of building blocks and indicators of stress in different parts of the architecture.

The Insectt architecture is based on the concept of the Bubble, a concept developed in the project DEWI that has been adapted to the new technologies. A bubble is a set of communicating industrial objects and industrial infrastructure encapsulated under a single physical or virtualized cloud gateway infrastructure to ensure security and confined, secured and private communication. The services inside the bubble are enabled by the unique instance of the bubble gateway, guaranteeing the secure access from external sources and protection from malicious attacks, being compatible with modern technologies such as Edge, cloud computing, as well as block chain, and 5G direct cloud connectivity technologies.

The reference architecture also allows us to introduce the concept of security and trustworthiness metrics calculated with different models or approaches across different layers and entities of the architecture. This metrics approach is relevant for future certification and standardization of technologies related to AI and IoT products and services. This has a high impact on the future consumer market and economy of the EU zone. The objective of the task is also to introduce the concept of trust by design in AI-IoT systems.

A crucial aspect in the design, validation and testing of modern IoT use cases is the ability to recreate most of the physical phenomena and networking protocols of the system in software/hardware platforms using realistic models. These software platforms are also called system level simulators. CISTER/ISEP is building a detailed system level simulator for two IoT use cases: wireless platooning and wireless avionics intra-communications.

The objective is to evaluate the detailed performance of the different network elements with an implementation of different reliability and security issues. These issues include detailed fading distributions, interference, and also cybersecurity attacks and their corresponding countermeasures using artificial intelligence and machine learning algorithms.

The main attacks and issues that are being considered lie in the physical layer such as jamming interference, eavesdropping, man in the middle, node capture, node tapering, denial of service, etc. Upper layer issues and attacks are also considered in a basic form: spoofing, man-in-the-middle, tampering, repudiation, denial of service, etc.

TU Delft startup Zero Energy Development (ZED, http://zed-iot.com) is a combination of an innovation lab, service provider, and technical and scientific consultancy, that delivers batteryless (and low power) Internet of Things (IoT) solutions that can harvest energy from their surroundings, sense the environment at the same time and wirelessly transmit the sensed information, providing insightful analytics. ZED develops new, sustainable and scalable energy harvesting (EH-)IoT systems, based on the needs and requirements of clients.  Three of the five core members of ZED are from InSecTT project: associate professor Dr. R.R. Venkatesha Prasad, Postdoc Ir. S. Narayana, Ir. N.H. Hokke, and PhD student Ir. S. Sharma, and is guided by Prof. Dr. John Schmitz (co-founder of NXP). ZED wants to be the springboard that can roll out these innovations from the university labs to the factory floors. By developing high TRL prototypes ZED wants to validate the team’s EH innovations not only from a technical but also from a business perspective. #innovation #startup #iot #ai