Wolfram Hardt

A methodology for rapid analysis and optimization of embedded systems

The paper describes an environment for the rapid analysis, synthesis and optimization of embedded systems. Since the implementation of these systems is rather complicated, we propose a methodology which automates the entire design flow. Flexibility is achieved by allowing manual intervention which is realized via a modular implementation of algorithms which are being provided. The applicability of the proposed approach is shown in terms of an example (UNIX command grep).

Rapid prototyping of real-time control laws for complex mechatronic systems: a case study

Rapid prototyping of complex systems embedded in even more complex environments raises the need for a layered design approach. Our example is a mechatronic design taken from the automotive industry and illustrates the rapid-prototyping procedure of real-time-critical control laws. The approach is based on an object-oriented structuring allowing not only central control units but also distributed control units as needed by todays designs. The implementation of control laws is a hardware-in-the-loop simulation, refined in steps and reducing the simulation part at every one of these. On the lower level, common platforms, such as FPGAs, microcontrollers or specialized platforms, can be instantiated.

Wake-Up-Receiver Concepts - Capabilities and Limitations

A promising idea for optimising the power consumption of mobile communication devices represents the usage of an additional ultra-low-power receiver unit, which is able to control the main transceiver in order to reduce the standby power consumption of the overall system. Such a Wake-Up-Receiver (WuRx) unit senses the medium and switches on the communication interfaces in case of an external request. Otherwise, all system components for the network communication are completely switched off. Especially in the domain of resource-limited and embedded devices, WuRx technologies enable novel communication paradigms. But on the application layer, not all scenarios allow the efficient usage of such WuRx technologies. Dependent on environmental parameters, technological limitations and conceptual requirements, different strategies are necessary to ensure energy-efficient system operation. In this article, we present a critical analysis of the capabilities and the conceptual limitations of WuRx approaches. Therefore, we identify critical parameters for WuRx concepts, which limit the efficiency in real world scenarios. Our goal is to classify sufficient fields of application. Furthermore we evaluate the influences of these parameters on the system behaviour. In addition, we introduce heterogeneous energy harvesting approaches as an efficient way for the system optimisation. The proposed technologies are capable to be integrated into small-sized wireless sensor platforms and prolong the system uptime significantly. The presented simulation results are focusing on actual smart metering scenarios and wireless sensor networks. Based on these measurements, we were able to apply further optimisation steps within the system configuration on the application layer. In this context, we focus on application-specific key issues, like the trade-off between measurement quality and quantity, the usage of data buffering approaches and QoS capabilities.

SimANet – A Large Scalable, Distributed Simulation Framework for Ambient Networks

In this paper, we present a new simulation platform for complex, radio standard spanning mobile Ad Hoc networks. SimANet - Simulation Platform for Ambient Networks - allows the coexistence of multiple radio modules with different communication technologies and protocol stacks within one node, which can be used concurrently. By the usage of efficient data structures like Randomised Skip Quadtrees, SimANet allows the analysis and evaluation of large scale, heterogeneous network topologies in both static and dynamic simulation scenarios based on different movement models. The software design enables a modular extension with additional models for power consumption, communication complexity or barrier simulation. Furthermore, an integrated MPI library provides the possibility to run distributed test cycles on parallel computing systems. Thereby, special sliding time window algorithms avoid the typical disadvantage of a slow network interconnection structure and allow a dynamic load balancing on the available hardware resources during the runtime. With the main focus on the evaluation of abstract multi-interface, multi-standard communication concepts, we compare the functionality and the complexity of SimANet with well-known simulation tools like ns2-MIRACLE, ns2-NW-Node, OMNet++ and TeNS. Simulation results for different application scenarios estimate features like versatility, practicality or usability in large scale network topologies with up to 105 nodes.

An XML format based integration infrastructure for IP based design

IP based design comprises two main steps. First, a suitable IP has to be found. Second, the IP has to undergo further processing before it can be included in existing designs. Since IP providers use different formats and design tools require different parts of the IP for their services it is hard to find an IP format that satisfies and is exchangeable between all parties. This paper presents such a unified format. It can be used to search for, deliver and exchange IPs between the designer and external IP service providers. A tool chain based on the format was implemented and evaluated. The results prove its applicability.

Prototyping of Tightly Coupled Hardware/Software-Systems

Verification and test issues raise the need for rapid prototyping of complex systems and especially hardware/software-systems. We tackle this problem by integration of hardware/software-codesign and prototyping. First we define the concept of the entire system architecture. This concept directs the hardware/software-partitioning process. Our prototyping environment reflects the architecture concept as well. In this overview the architecture concept and all important design tasks (hardware/software-partitioning, speed-up estimation before HW-synthesis, and prototyping of the entire hardware/software-system) are presented and compared to several approaches from literature. Thus a substantial overview over the prototyping problem is given. The latter part of this presentation illustrates our approach by a case study and presents the results. Our automated design process generates a tightly coupled hardware/software-system with very good performance characteristics. The case study focus on the prototyping of a ciphering algorithm. The reported approach leads to a reasonable overall system speed-up of 10 percent. Similar results have been found for further examples as well.

Energy efficient handling of big data in embedded, wireless sensor networks

The development of wireless sensor networks has reached a point where each individual node of a network may store and deliver a massive amount of (sensor-based) information at once or over time. In the future, massively connected, highly dynamic wireless sensor networks such as vehicle-2-vehicle communication scenarios may hold an even greater information potential. This is mostly due to the increase in node complexity. Consequently, data volumes will become a problem for traditional data aggregation strategies traffic-wise as well as with regard to energy efficiency. Therefore, in this paper we suggest to call such scenarios big data scenarios as they pose similar questions and problems as traditional big data scenarios. Although the latter focus mostly on business intelligence problems. We then propose an aggregation strategy tied to technological prerequisites which enables the efficient use of energy and the handling of large data volumes. Furthermore, we demonstrate the energy conservation potential based on experiments with actual sensor platforms.

Speed-up estimation for HW/SW-systems

HW/SW-codesign has been applied to a wide range of applications. Several partitioning methods have been suggested. Thus the designer selects modules for HW or SW-implementation for the best possible performance within a set of performance and design constraints. This paper describes an estimation method to approximate a priori the entire system performance. The estimation method has been integrated into the codesign tool COD and first results could be generated. The estimated speed-up has been determined for a ciphering algorithm and has been compared to the speed-up of the entire HW/SW-system. The estimation speed-up matches the final speedup.

Data aggregation and data fusion techniques in WSN/SANET topologies - a critical discussion

WSN and SANET topologies generate huge amount of heterogeneous data, which has to be transmitted in a dynamically changing network infrastructure. Especially in the domain of wireless low-power applications, the energy-efficiency and the prioritisation of communication tasks is critical. Several research areas deal with this issue. They optimising the respective hardware components as well as the protocols within the PHY, MAC or network layer. But for an optimised media transport in the topology also the data management and the task scheduling on the application layer is essential. Here, the key challenge is to minimise the data amount without decreasing the information quality. Related research work in the field of data aggregation and data fusion offers interesting techniques for an efficient data handling. In this paper, we discuss usual ways for data aggregation, including the adapted communication process. We critically analyse the benefits in theory and compare these conceptual advantages with measured real-world results. The evaluation was done in two steps. The first one is based on simulation scenarios of typical WSN/SANET applications. In a second step, we implement a demonstrator platform for a respective real-world environment. The test bed configuration is similar to the simulation scenario and provides comparable data. Based on the results and the respective analysis, we propose feasible methods for optimising data aggregation techniques. We clarify, that these improvements are essential for an efficient usage in resource-limited, embedded sensor network environments.

Performance analysis of radio standard spanning communication in mobile Ad Hoc networks

In this paper, the feasibility and necessity of a new concept for radio standard spanning communication in mobile Ad Hoc networks (MANET) has been analysed. Past research approaches limited their solutions to the usage in a homogeneous topology on basis of a unique radio standard. The proposed concept offers the possibility to connect standardised radio modules on a hardware near layer by the usage of a dedicated Interface Block. Such an IFB allows the interoperability between tasks with incompatible protocols. For evaluating the conceptual advantages of this approach, different network topologies have been tested in a dedicated simulation environment. The scenario-based performance analysis includes the behaviour of the nodes both in static and in high-dynamic network topologies. Thereby, the main objective of verification is a qualitative evaluation of the power consumption in each node and in the entire topology. Furthermore, improvements of route paths and the connectivity level have been analysed. The degree of improvements in the analysed scenarios averages 15% and rises up to 35% in a random distributed, heterogeneous network topology with a high node density.