Research in the Computer Networks Group

The working group deals with the manageability of wireless networking and mobile systems. We work algorithmically on the basis of a graph view of such systems, analytically and simulatively on the basis of channel models and models of the networking of wireless systems, as well as empirically on the basis of prototype implementation on real hardware (WSN and AUV platforms, as well as single-board PCs), field test, and laboratory experiment.

The following research topics and projects currently being worked on are derived from the research fields mentioned in this context:

Connectivity, percolation and inherent local structural properties in wireless networked systems of any size

We consider systems of any size (formally with an infinite number of nodes and from a practical point of view thousands of nodes), which are connected by means of mobile communications, described by technical channel models. The resulting graphs are random. An obvious question is about connectivity in such graphs (weak connected component). In systems of any size, however, connectivity is a too conservative requirement, since a network node can always be isolated. Here, on the other hand, percolation, as a weakened form of the connectivity, is a suitable system variable. In infinite graphs this means to find an infinitely large connected component; in finite graphs comparable to a very large connected component, which includes almost all network nodes.

We develop and investigate methods for structuring such random graphs, which preserve percolation and promote certain local structural properties of the resulting graph. Here we do not consider deterministic guarantees, but examine to what extent such properties can be generated at a high rate. The properties themselves are subject of further research about local communication and topology control.

The research project is funded in the context of the DFG project "Reactive construction of and reactive routing in Euclidean and topological network spanners in wireless ad-hoc and sensor networks".

Local topology control under graph structure assumptions in mobile networked systems and their computer-aided verification

The starting point of our research in this section are graphs with assumed local structural properties, as these can be observed or generated in wirelessly networked systems at high rates. Under these conditions, we design local methods, i.e. methods in which a network-wide goal can be achieved on the basis of local decisions by the network nodes based on their neighborhood information. Any networks that follow these structural assumptions are structured in such local way. Here, graph connectivity, Spanner property, limited degree and planarity are possible examples of such properties to be generated.

Another aim is to formalize the resulting graphs of such local method using logical rules. Using such rules, we describe the results of the algorithms as so-called axiomatically defined graph classes. Many questions as to whether our algorithms generate graphs with certain properties can be traced back to containedness relations of graph classes. By means of software tools from formal verification, and questions formulated or programmed in such systems about containedness relations, we want to automatically derive guarantees of certain graph properties from such logically described graph classes.

Cooperative distributed regulation and topology control for wirelessly networked mobile autonomous distributed miniature robot systems

In this project we consider cooperative control of mobile nodes (control of AUV or UAV teams and vehicle fleets respectively; summarized under the term agents in the following) that interact over a wireless network. The network is set up by the nodes, i.e. no external communication infrastructure is assumed. Due to hardware constraints, in many cases only distributed approaches can be implemented. In such distributed cooperative control schemes, where each agent is equipped with a local control unit, the overall system performance is achieved by information exchange over a wireless communication network. Due to the broadcast property of the wireless network, agents can send samples in one transmission to their immediate neighbours. The subset of neighbours which further process the received samples in their local control is described by the interaction topology. This is a subgraph on top of the wireless communication graph. It is evident that a change in the topology as well as non-ideal communication will affect the performance and robustness of the interconnected system.

Cooperative control of multi-agent systems, as well as topology control in communication networks, have been studied extensively, but separately. Research on cooperative control of multi-agent systems usually assumes the interaction topology to be given. Research on topology control usually aims at improving the network in the sense of communication properties. Since the interaction topology in a network of mobile agents has a significant influence on the achievable cooperative control performance, in this project we combine a control scheme for dynamic control of the movement of the agents with the adaptation of the interaction topology to the time-varying conditions of the communication network. Both simulatively and analytically, we prove that such combined procedures significantly improve the achievable control performance (for example, specifically in terms of convergence rates).

The work in this project is funded as part of the DFG project “Cooperative regulation and topology control in wireless networked mobile systems (AUVs and UAVs)” in the priority program SPP1914 “Cyber-physical Networking”. The project is in cooperation with the Institute of Control Systems, TU-Hamburg-Harburg, the Center for Research in Electric Autonomous Transport (CREAT), Department of Electrical Engineering and Computer Science, University of Central Florida, and the Complex Systems Control Lab, School of Electrical and Computer Engineering, College of Engineering, University of Georgia.

Reliable communication and ultra low latency for controlling autonomous mobile robot platforms in urban logistics and industry

This project focuses on 5G mobile communication for urban logistics and material transport in the factory environment using Automatic Guided Vehicles (AGVs). The reliability of 5G mobile communication in the factory environment is a particular topic to be investigated. At a specific location, methods for 5G measurements in hardware and software are implemented in real operation. The aim is, among other things, to derive quantities on communication channels in the context of modern self-organizing factories. Here, the question is examined to what extent ultra reliable low latency communication (URLLC) defined in the 5G standard can be achieved in the case studies considered and where the limits are. For this purpose, models are to be derived. Such models are crucial for planning future smart infrastructures such as smart cities and smart factories.

Furthermore, questions about system integration are also considered, including the following: To what extent can technologies that are not part of the 5G standard be integrated? What fallback mechanisms are there in the event that a wireless 5G connection deteriorates? To what extent can 5G be integrated into communication without infrastructure or to what extent can it be integrated into 5G?

This project is a co-supervision with the Koblenz University of Applied Sciences as part of a doctoral student funded by industry.

Data centric networks in the context of vehicular ad-hoc networks

With the success of the Internet, communication behavior has changed significantly over the years: Communication participants are interested in content and less in having to know the physical location of the content. In addition to communication behavior, the type of access to networks has also changed: Participants are increasingly mobile (e.g. smartphones and laptops, mobile devices in the Internet of Things, networked vehicles). As a result, participants frequently change communication connections (network change, change of communication technologies (WLAN, cellular network)), which involves addressing and routing information from and to end points in a network, maintaining end-to-end connections, as well the identification of participants is becoming increasingly difficult.

In recent years, work has been carried out on technologies and concepts for the “Internet of the future”, which moves away from host-based end-to-end communication towards data-centric communication. The principle: Instead of end nodes, data is addressed directly. This is summarized under the term “Information-Centric Networking” (ICN).

The continuous networking of everyday objects is also finding its way into the automotive industry. Vehicles are increasingly being equipped with sensors, actuators and communication units (cellular radio, WLAN, etc.). They form a mobile ad-hoc network, also known as a vehicular ad-hoc network (VANet), and are therefore able to exchange information with one another or with an infrastructure. However, vehicles in this network have one common denominator: they are highly mobile, which makes it difficult to exchange information using classic communication models.

This project investigates the extent to which concepts from data-centric networks can be transferred to the domain "Vehicular Ad-hoc Networks". Concepts such as addressing, caching and forwarding are examined, new strategies are developed in this context and empirically evaluated using computer simulation and prototype implementation.

The research is carried out as part of an external doctorate with the central research department of Robert Bosch GmbH in Renningen.

Software defined networking for the management of mobile networked IoT systems

In this project we consider Software-Defined Networking (SDN) in the context of IoT with the aim of developing methods with the following properties. (1) Network management and scalability of wireless networks should be designed in such a way that the entire network can be reprogrammed, configured and managed very easily. (2) Mobility of wireless sensor networks: As nodes move, the network topology changes. As a result, time is required to calculate new paths (so-called routing convergence). Another goal of this project is to show how using the SDN concept will improve the routing convergence and path quality mentioned above. (3) Limited communication bandwidth, limited processing / storage capacities and limited available energy (from battery-operated nodes): The SDN approach is designed in such a way that only a small number of control messages is required and solutions with limited compute resources can handle different requirements different applications (e.g. latency and throughput). In addition, in the case of battery-operated nodes, the SDN organization is designed to be very energy-efficient.

Specific challenges in the project are: How can each node be reached with its IP address? How to implement OpenFlow tables in nodes while reducing the need for OpenFlow control messages as much as possible? How to deal with mobility and changing network topology in order to achieve a largely transparent mobility / dynamic network change? How can the required number of control messages be kept as small as possible in order to improve throughput and latency, improve energy consumption and extend battery life?

Investigation of wireless networking from and by flying robots on the basis of empirical field test and laboratory studies

Another research field worked on in our working group are empirically investigated research questions based on prototypes of cooperating flying robots. More complex algorithms can be empirically investigated on a prototype of several flying quadro copters with an embedded Linux board. Furthermore, a smaller swarm with approx. 30 micro-UAVs (CrazyFlies including extensions) is available in the laboratory.

We are currently dealing with questions about coordinated placement of robots such that the positions taken create the best possible network structure for the robot team while maximizing the covered area. We are concerned with the design of algorithms based on mass-spring systems and grid-based structures and their evaluation in real test setups. Signal strength measurements (RSSI values) are used as the objective of mass-spring systems and grid structures.

In addition, we deal with self-organized development of support networks. This includes, on the one hand, the provision of a wireless network infrastructure through suitable positioning of the flying robots themselves, as well as the dropping of relay nodes in suitable positions. For the latter, prototypical hardware nodes are available. Suitable network formation protocols for very low-performance microcontrollers are to be implemented on these nodes.