Internet of Things (CPS/Industry 4.0)
The Internet of Things (IoT) envisions a world where virtually everything is connected, including billions of mobile devices and everyday objects. Examples include cars, smart homes, wearable devices such as smart glasses and watches or, following the trend of Industry 4.0, industrial automation systems. Sensors embedded into these devices sense the state of users, objects, and ultimately the world, to enable context-aware applications that react autonomously to the current situation without manual intervention by the user.
Networked IoT devices featuring embedded sensors can deliver huge amounts of sensor data to ultimately capture the state of the world. This sensor information is valuable for many applications, including traffic monitoring, urban planning, environmental monitoring, or the efficient management of resources like electricity. However, installing a dedicated, fixed sensor network spanning a large area is costly. Therefore, we investigate public sensing concepts that use the crowd of already existing mobile devices like smartphones or wearable devices to implement a huge mobile sensor network with great coverage in urban areas. This setup is augmented by cheap wireless sensor devices based, for instance, on the BLE standard. Mobile devices are used as gateways to the Internet.
We develop concepts to face four essential challenges of public sensing:
- saving energy of battery-operated IoT devices involved in sensing
- delivering a certain quality of sensor data through a crowd of unreliable IoT devices of private users
- managing a huge crowd of IoT devices, e.g., to track their locations
- protecting the privacy of users in the IoT, which might collect privacy-sensitive user data
Applications:
We investigate how to automatically build detailed 2D and 3D indoor models from crowd-sensed data collected by mobile users.
Additonally, we research concepts to implement cyber-physical systems (CPS), which monitor the state of physical objects through sensors and control their physical state through a set of actuators. Physical objects to be controlled include machines in an Industry 4.0 factory, a fleet of cars, smart home devices, or energy consumers and producers in the smart grid. In particular, we are interested in CPS consisting of a set of geographically distributed networked sensors, actuators, and controllers implementing a networked control system. Such systems are typically highly sensitive to the delay and loss of communicated sensor data. Therefore, operating them over the Internet, which was not designed to support such time-sensitive applications, is a great challenge. To this end, we investigate networking concepts optimizing the quality of control of networked control systems operating on top of an unreliable IP-based networking infrastructure like the Internet.
Projects:
- PriLoc Projekt
- Com'N'Sense Projekt
- SimTech: Mobile Simulations
- Integrierte Reglerentwurfsverfahren und Kommunikationsdienste für digital vernetzte Regelungssystem
Cloud computing offers scalable infrastructures that are well-suited to handle the huge amount of data stemming from the large number of IoT devices. In our research, we tackle essential challenges for utilizing cloud computing for the IoT:
- Quality of service: Designing a cloud computing infrastructure offering predictable quality of service with respect to availability, latency. etc. at minimum cost, i.e., without heavy over-provisioning.
- Fog / Edge Cloud Computing: reducing the latency of geo-distributed IoT services by distributing functions and data between remote cloud data centers and edge cloud (“fog”) resources close to the devices.
- Mobile cloud computing: increasing the energy-efficiency of battery-operated, mobile IoT devices by dynamically offloading computations from the resource-constrained device to the cloud or fog.
Projects
- SimTech: Mobile Simulations
- ARAMiS
Most IoT devices embed sensors that can feed simulations enabling novel applications. In particular, we are interested in simulations utilizing mobile devices, allowing for instance to execute simulations “in the field”. A typical example could be structural engineer who wants to simulate the current state of a building, bridge, or similar based on collected sensor values. Using a mobile augmented-reality headsets like Microsoft’s HoloLens, the engineer can visualize simulation results in real-time, and evaluate planned changes to the architecture on-site.
Typically, simulations are compute-intensive requiring both, long time to calculate on a mobile device and lots of limited battery resources. To solve these problems, we investigate how to enable distributed mobile simulations. The basic idea is to offload parts of the simulation to a powerful remote cloud computing infrastructure or a nearby edge cloud infrastructure (“fog”) to increase efficiency. Using optimized simulation methods and a careful consideration of the trade-off between communication overhead, computation overhead, and required quality of simulation results, we can achieve a significant reduction of energy consumption and execution time to bring even complex simulations to mobile devices.
Projects
- SimTech: Mobile Simulations
Mobile Systems / Ubiquitous Computing
With the advent of smartphones and tablets, mobile devices became ubiquitous in our everyday life. Moreover, the development of wearable devices such as smart glasses, smart watches, and fitness trackers, as well as the trend towards including substantial computing capabilities into mobile objects like cars will further contribute to increase the population of mobile devices, which already today includes billions of devices.
In our research, we investigate concepts to support and enable mobile and ubiquitous computing systems and applications.
Many mobile devices feature sensors to capture the state of the environment surrounding the mobile user or object (e.g., car). This sensor information is valuable for many applications including, for instance, traffic monitoring, urban planning, environmental monitoring, or the efficient management of resources like electricity. Utilizing the crowd of mobile devices for “public” sensing, we can implement a huge sensor network with great coverage without the cost of installing a dedicated, fixed sensor network.
In our research, we investigate concepts to enable the public sensing paradigm:
- saving energy of battery-operated mobile devices involved in sensing
- delivering a certain quality of sensor data through a crowd of unreliable mobile devices of private users
- managing a huge crowd of mobile devices, e.g., to track their locations
- protecting the location privacy of mobile users
Besides these generic challenges, we also target concepts to support a specific public sensing applications for automatically building detailed 2D and 3D indoor models from crowd-sensed data such as movement trajectories, photos, or 3D point clouds collected by mobile users.
Projects
- SimTech: Mobile Simulations
- ARAMiS
- Com'N'Sense Projekt
Mobile cloud computing aims for supporting the implementation of mobile systems by utilizing the virtually unlimited resources of cloud computing. In our research, we target two specific goals of mobile cloud computing:
-
Increasing the energy-efficiency of battery-operated mobile devices by dynamically offloading computations from the resource-constrained device to the remote cloud or a nearby edge cloud computing infrastructure (“fog”).
- Utilizing non-trusted third-party cloud infrastructures to manage privacy-sensitive user information. In particular, we investigate concepts to protect the location privacy of users storing their location data in the cloud.
Projects
- SimTech: Mobile Simulations
- ARAMiS
Novel mobile devices including, for instance, augmented-reality headsets such as Microsoft’s HoloLens, allow for visualizing simulation results in an intuitive manner as an overlay of the physical world. This not only enables novel gaming and entertainment applications, but also serious engineering applications. For instance, a structural engineer could simulate the current state of a building, bridge, etc. and evaluate planned changes to the architecture on-site using his mobile headset.
Typically, simulations are compute-intensive requiring both, long time to calculate on a mobile device and lots of limited battery resources. To solve these problems, we investigate how to enable distributed mobile simulations. The basic idea is to offload parts of the simulation to a powerful remote cloud computing infrastructure or a nearby edge cloud infrastructure (“fog”) to increase efficiency. Using optimized simulation methods and a careful consideration of the trade-off between communication overhead, computation overhead, and required quality of simulation results, we can achieve a significant reduction of energy consumption and execution time to bring even complex simulations to mobile devices.
Projects
- SimTech: Mobile Simulations
Software-defined Environments
Software-defined environments promise to increase the flexibility and efficiency of networking, computing, and storage infrastructures. The conceptual and technical cornerstones of software-defined environments are:
- Logically centralized control: based on a holistic global view onto networking, compute, and storage resources, these resources can be managed more easily and efficiently. Note that although control is logically centralized, it might still be physically distributed to increase scalability and availability.
- Virtualization: Based on virtualization technologies, functionality becomes independent of dedicated physical hardware, and can be distributed flexibly to utilize and share available physical resources efficiently.
- Open standard interfaces to hardware: open standard interfaces to hardware enable the logically centralized controller to control hardware resources. Moreover, exposing functionality through standard interfaces allows for adapting and utilizing available hardware resources that previously appeared to be “black boxes”.
In our research, we develop concepts to improve the performance, efficiency, consistency, and fault-tolerance of distributed software-defined environments.
The trend towards software-defined environments started with the advent of software-defined networks (SDN). SDN separates the network control plane, which is responsible for controlling the network elements (switches), from the data plane, which is responsible for forwarding packets. Network control is outsourced from switches to a network controller, whose logic can be easily adapted by exchanging software components implementing the control logic.
In our research, we investigate how to utilize SDN to improve the performance of communication middleware systems (e.g., publish/subscribe systems) and networked applications (e.g., networked control systems as used, for instance, in automation). These systems were previously implemented on the application layer of the OSI model. With the availability of SDN, functionality can be pushed into the network to tailor the network to the application requirements to substantially improve performance like latency and throughput. Moreover, network and application can be optimized together as an holistic system.
Moreover, we investigate concepts to physically distribute the SDN control plane to increase its performance and availability, as well as concepts to ensure the consistency of the (distributed) SDN control plane and data plane.
Projects
- Software-defined Networking
Time-sensitive Networks (TSN) consider network delay as the primary quality of service metric of communication targeting, for instance, applications from industrial automation, telerobotics, the smart grid, etc. In our research on TSN, we investigate how to optimally manage network resources under given delay constraints for both local area networks according to the IEEE 802 TSN standards, as well as wide-area IP networks, both originally not providing sufficient quality of service guarantees for time-sensitive applications.
For IEEE 802 networks, we investigate, how to achieve deterministic bounds on network delay through the logically centralized configuration of scheduling complementing the IEEE 802.1Qbv standard for scheduled (time-triggered) traffic.
For wide-area networks, we investigate novel network abstractions and SDN-based resource management mechanisms for cyber-physical systems implementing a networked control system of geo-distributed sensor, actuators, and controllers.
Projects
- GSaME - Project E5: Novel Communication Architecture for the Smart Real-Time Factory
- Integrierte Reglerentwurfsverfahren und Kommunikationsdienste für digital vernetzte Regelungssysteme
Virtualization is an essential part of software-defined environments to become independent of dedicated hardware and utilize physical resources efficiently. However, if applications and services with stringent quality of service demands are executed on shared hardware, we face the problem of managing available resources such that quality of service guarantees are always fulfilled. To this end, we investigate suitable resource management concepts for virtualized systems, in particular, targeting time-sensitive systems and distributed systems with strict availability requirements such as cloud services.
Projects
- ARAMiS
Geo-distributed cloud & edge computing
The growing ubiquity of mobile and sensor devices paves the way towards the Internet of Things (IoT) by enabling large-scale applications in the areas of smart homes, smart cities, environmental monitoring, healthcare and social networks etc. These IoT applications on the one hand produce large and fluctuating volumes of data and on the other hand requires high-speed, highly available and resource efficient data processing ensuring low response times for the user queries. For example, semi-autonomous car applications assist drivers in reaching their destinations safely by leveraging online analysis of traffic situations and driving patterns. Likewise, surveillance applications can assist police officers in identifying persons with suspicious activities in the nearby vicinity by online processing of camera streams.
Aggregating and processing data at a centralized cloud platform is not adequate to meet the requirements of these IoT applications. Consequently, there is a paradigm shift pushing the horizon of utility computing model offered by cloud platforms towards the edge of the network – leveraging resources in the network both for compute and storage – enabling application logic to execute on geographically distributed resources throughout the network including routers, edge compute clusters and backend data centres. To this end, large-scale cloud providers such as Microsoft and Google are deploying data centers and edge clusters globally to provide their users low latency access to the computational resources and the cloud services.
In our research, we address many challenges related to processing and management of data in such geo-distributed settings to support latency and availability requirements of IoT applications while ensuring efficient utilization and orchestration of heterogeneous compute and storage resources at the different levels of network hierarchy.
Modern IoT applications need to be able to react to situations occurring in the surrounding world. Thus, a growing number of sensor streams need to be processed in order to detect situations which the application or user is interested in, e.g., the traffic situation in a smart city or the detection of a person in a video surveillance application. To detect situations from sensor streams, Complex Event Processing (CEP) is a well-established paradigm building the bridge between sensors and consumers, i.e., applications or users that are interested in situations. In CEP, domain experts specify the events to be detected following the well-known operator graph model. Each operator defines how to detect event patterns on its incoming event streams and specifies the events to be produced whenever an event pattern was detected. The operators form a network and allow for the stepwise analysis from low level sensor data up to events of interest delivered to the consumers attached to the CEP system. In our research, we tackle essential challenges for utilizing CEP for the IoT applications.
- With the increasing number of data sources and increasing volume at which data is produced parallelization of event detection is becoming of tremendous importance to limit the time events need to be buffered before they actually can be processed by CEP operators. To this end, we are investigating methods for scalable and dynamic adaptation of the parallelization degree of CEP operators so that operator network can meet latency requirements at minimal cost and can ensure the consistency of produced complex event streams w.r.t. the sequential processing.
- Fault tolerance is of critical importance to many IoT applications involving CEP systems. In particular, the event streams provided to consumers of CEP system should be indistinguishable from an execution in which the hosts of some operators fail or unavailable during a temporary partitioning of the network. The CEP operators are usually stateful making it more challenging to provide efficient recovery from failures and to ensure consistency of produced event streams (i.e, no false positives, false negatives and duplicates). To this end, we investigate models and mechanisms to ensure fault-tolerant and correct processing of CEP system in the face of diverse failures.
- To support mobile applications and event sources, we investigate methods for the placement and migration of CEP operators according to the consumer’s location to minimize latency and network utilization. Further research includes on-demand adaptation algorithms to dynamically assign cloud and edge resource to operators in CEP system to minimize processing overhead.
- Respecting privacy is of paramount importance to ensure user acceptance of IoT applications, especially data captured from sensors is often highly privacy sensitive. To this end, we investigate different access control methods for protecting private information in CEP, while ensuring minimum degradation in quality of data to implement IoT applications.
Projects
- Parallel complex event processing to meet probabilistic latency bounds
- Privacy in Stream Processing
- CEP in the Large
Today, myriads of ubiquitous sensors integrated in many modern devices are collecting an innumerable amount of raw observations, e.g., GPS positions. Furthermore, to benefit from this enormous amount of data, there is a recent trend of utilizing machine learning algorithms to derive knowledge from the gathered observations. For example, a smartphone can learn the best time to leave home dependent on the weather conditions and travel time to work.
Today every smart phone, every application and every company, i.e., every connected entity, is gathering and analysing such observations. Clearly, greater benefit can arise if such a knowledge is not only used locally but also accessible to others. This raises the need for the data management mechanisms that allows for the distributed maintenance of the observations and knowledge gathered by different entities. In particular, we investigate effective and efficient methods for indexing, updating and retrieving large and changing quantities of data in a large-scale distributed system.
Projects
- Adaptable Pervasive Flow Ensembles
