Andreas Wolff

Network-centric Middleware for Service Oriented Architectures across Heterogeneous Embedded Systems

The increasing ubiquity of mobile embedded systems has been matched by the evolution of a variety of wireless network standards and technologies. The major constraints of wireless embedded systems are limitations of communication bandwidth, processing capabilities, and battery power. Remote wireless embedded systems often act as sensors, which provide data to a certain community. The exchange of data is nowadays usually realized through Web services. For standard Web services an XML parser is required, which demands a lot of processing power whereas embedded devices with their resource constraints provide only limited processing capabilities. Another drawback of standard Web services is the huge communication overhead due to the use of human readable XML. On the other hand using standard Web service technology enables service developers to easily integrate new embedded devices to their network or to the Internet. This paper reports on the on-going work of the key European research project MORE, which brings the embedded system world together with the Web services world. MORE introduces a new connector concept easing the handling of heterogeneous communication channels and muSOA for reducing the message size and processing demands for parsing XML messages.

Performance analysis of mesh routing protocols for UAV swarming applications

Unmanned Aerial Vehicles (UAVs) are an emerging technology offering new opportunities for innovative applications and efficient overall process management in the areas of public security, cellular networks and surveying. A key factor for the optimizations yielded by this technology is an advanced mesh network design for fast and reliable information sharing between UAVs. In this paper, we analyze the performance of four available mesh routing protocol implementations (open80211s, BATMAN, BATMAN Advanced and OLSR) in the context of swarming applications for UAVs. The protocols are analyzed by means of goodput in one static and one mobile scenario using the same embedded hardware platform installed at UAVs in current research projects. Our results show that layer-2 protocols suit better for mobile applications in comparison to layer-3. On the other hand, they often cause routing flippings, which are unwanted route changes, in static scenarios imposing a small performance decrease. Hence, given the aforementioned routing protocols, we recommend to currently use open80211s or batman-advanced to establish a reliable multi-hop mesh network for swarming applications.

Protocol Design and Delay Analysis for a MUAV-Based Aerial Sensor Swarm

In dynamic networks such as highly mobile swarms of micro unmanned aerial vehicles (MUAV) the round trip time (RTT) is a crucial key figure for the reliability and viability of the system. For this purpose the deployed protocols and underlying network architecture have to be designed in awareness of the available and allocatable throughput. In this paper we are investigating a network and protocol design of a novel system approach for aerial remote sensing by means of MUAV swarms. Since effectiveness and performance ability of the navigation, control and guidance algorithms depend not only on sense and avoid or collision avoidance capabilities of each MUAV, we are particularly analyzing the delays and RTT between each node in the considered agent-based avionic mesh network. As the proposed protocol design has got a significant impact on the processing delay of embedded systems, a major focus is given to the RTT analysis of a multi-hop WLAN mesh network. Next to the self-interference caused in the experimental testbed, external co-channel and common channel interference are considered to emulate a realistic environment.

Performance Analysis of Highly Available Ad Hoc Surveillance Networks Based on Dropped Units

In emergency situations like an attack on a building, public networks are not sufficiently reliable. Hence field forces, who want to share multimedia mission information, have to build up their own multi-hop wireless network. To build up these ad hoc networks, battery driven wireless nodes with meshing capability, called dropped units, present a new technology option for homeland security applications. To ensure sufficient coverage, an excessive usage of dropped units seems to be an obvious solution, but leads to undesired effects, such as interference. To solve this dilemma, rules to find an optimal number of dropped units are presented in this paper. To derive guidelines for this, realistic homeland security scenarios were analyzed. A dedicated simulation environment provided a reliable prediction of the required number of dropped units. We analyzed the communication link of moving or movable wireless clients like surveillance and first responders devices. The simulation results lead to the conclusion that with a limited number of dropped units a practically 100% network availability can be achieved.

A system design framework for scalability analysis of geographic routing algorithms in large-scale mesh networks

It is important to evaluate the performance of large communication networks prior to their deployment, in particular if economic interests are involved. In wireless multi-hop mesh networks, a communication message is transferred from a source to a destination via multiple nodes. Typically the message can be transferred via multiple routes in a mesh network, because several nodes are in communication range. During the design phase of such a communication network, specific characteristics need to be considered in order to avoid boundaries like bottlenecks and dead-end problems of the deployed system. Hence these kinds of problems must be avoided prior to the network deployment. In this paper, we present a system design framework for the OMNeT++ simulation environment, which is able to identify potential bottlenecks and maximum loads of multi-hop networks. The process is presented via a realistic use case scenario for an Energy Management Application, in which geographic routing algorithms are used to identify the shortest route to a destination. The results of the performance evaluation enabled us to support the communication design process with information about reliability, data rate and routing schemes.

MobileEmerGIS: a Wireless-enabled Technology Platform To Support Emergency Response Management

In this paper, the architecture and middleware services of the wireless-enabled emergency response management system MobileEmerGIS are introduced. MobileEmerGIS allows to gather and to distribute multimedia information to and from mobile nodes (either human users or sensors/actors). Thereby, field forces and eye witnesses can transfer data collected in the field, such as pictures from emergency sites, immediately to centrally located decision makers as well as other field forces (using PDA platforms and even their own private mobile phones). While the data collected by the system provides many opportunities for optimizing the coordination of the rescue activities, at the same there is a risk that on-site rescue forces are paralyzed by being overwhelmed with the exponentially growing amount of information. Therefore one of the key capabilities of MobileEmerGIS is that centrally stored data is made available to mobile rescue forces in a highly focused way, for example by applying geopositioning and role-dependent filters. By referencing real-life scenarios (such as a large-scale fire emergency), concrete application services enabled by MobileEmerGIS are presented.

Analysis of IP-based Real-time Multimedia Group Communication in Heterogenous Wireless Networks

Multi-media group communication services are a new kind of community-driven services. One key concern is that access link qualities of different group members may vary significantly especially in case group members are connected via heterogenous wireless networks. As a consequence, the efficiency of a group communication is negatively impacted. In this paper we provide a simulation analysis on how the throughput of a peer-to-peer group communication can be reduced in case only a few members of a group are effected by a link quality degradation. We first classify group communication systems regarding their overall architecture and their floor control mechanisms. Afterwards a dedicated communication protocol (the Push-to-X protocol) - developed to efficiently support group communication on embedded system platforms - is introduced. After presenting the simulation environment, we present a basic group communication traffic model derived from empirical observations. Finally simulation results are discussed, showing a throughput reduction of up to 50% in case only one participant of a group communication is affected by an additional link delay.

Performance analysis of 802.11 DCF parameters which support QoS in emergency scenarios

In case of an emergency, first responders need to establish a communication network at the area of incidence. To support upcoming multimedia services for emergency personnel, a communication technology which provides high data rates is required. At presence the most common wireless communication technology is IEEE 802.11, which provides sufficient data rates for emergency communication. Due to the popularity of IEEE 802.11 networks, existing networks at the incident scene cause harsh conditions for wireless communications. For first respon-ders, a reliable communication infrastructure is crucial. Hence the applied communication technology must provide Quality of Service (QoS). Existing QoS mechanisms, like IEEE 802.11e are designed for multimedia applications. In this paper, we present a new set of parameters for the DCF to support QoS in emergency scenarios. We propose six novel priority classes for heterogeneous types of emergency communication. All types of emergency communication are prioritized over legacy IEEE 802.11 communication. The performance of the new parameter set is analyzed through an analytical model based on Markov chains and is validated through simulations. The results lead to the conclusion that our proposed parameter set provides a reliable communication network for emergency scenarios, even with existing interference networks.

ROUTE-O-MATIC: A comprehensive framework for reactive mesh routing protocols

This paper outlines the implementation design and performance evaluation of ROUTE-O-MATIC, a comprehensive framework for reactive mesh routing on Linux. This framework provides a set of necessary but not natively supported services and interfaces for the development of mesh routing protocols on Linux. In wireless mesh networks, especially in mobile scenarios, reactive ad-hoc routing protocols are used to discover routes on demand between mesh nodes. Those protocols depend on special features for which the current network subsystem of Linux is not designed. These include among others an interface to notify about a required route discovery to a certain host, and a buffer to temporarily store all data while the route discovery is performed. The ROUTE-O-MATIC framework adds these features to the Linux kernel. Additionally, the framework offers a Link Layer Feedback support, which reports every transmission failure in order to accelerate the detection of broken links. The frameworks performance is investigated on an x86 and ARM based embedded hardware. The results show that the time overhead of ROUTE-O-MATIC is significantly low.

Impact of Multilevel Hierarchies on Performance of Wireless Peer-to-Peer Group Communication

Multimedia group communication services in mobile networks as a replacement for traditional trunked radio communication have attracted a lot of attention recently. This paper presents results from a newly developed simulation model to study the transfer of multimedia streams in group communication systems within different setups. As the real time ability of such a system has been identified as the key success feature, the focus of the presented studies is placed on the variation of the users’ upload data rate, the resulting overlay network structure and it’s impact on the overall transfer delay.