Abstracts - Final

Index (Authors, Titles)


Ana Beatriz Alvarez, José Raimundo de Oliveira:
Hand Gesture Recognition using a Lattice Autoassociative Memory.

Olaf Bergmann, Stefanie Gerdes, Jörg Ott, Petri Ylikoski, Nils Seifert, Caleb Carroll, Nigel Wallbridge:
CHIANTI: Enhancing Robustness of Communication in Disruptive Networks.

Rainer Bollmann, Wolfgang Stade, Gerd Stange, Thomas Zelenka:
Recent progress in magnetic-inductive flow metering devices (MID)

Ludo Cambré, Luc Friant:
Didactical principles in the use of embedded servers.

Johan Dams:
Towards a Reliable, Cost Effective and Easy to Operate Unmanned Aerial Vehicle for Technology Education

Johan Dams:
Optimizing Embedded Software - A Look at the NEON SIMD unit in the ARM Cortex Family of Processors

Andre Doria, Daniel Wagner, Iryna Pavlyshak, Mariana Lopez:
A series of insights: location sharing.

Margus Ernits, Kristiina Hakk, Kalle Tammemäe:
Teaching robotics and embedded systems as a learning chain of different subjects.

Yull Heilordt Henao and Fabiano Fruett:
A Sport Activities Monitoring based on Acceleration and Rotation Microelectromechanical Sensors.

Nils T. Kannengießer, Thomas Ladehoff, Thorsten Knutz, Helmut Dispert:
Implementation of a platform independent client software for the GO Bluebox System.

Gisnara Rodrigues Hoelzle and Fernando Morgado Dias:
Hardware Implementation of an Artificial Neural Network with an Embedded Microprocessor in an FPGA.

Luimula, M., Pieskä, S., Pitkäaho, T., and Tervonen, J.:
Ambient intelligence in mobile field work.

Tiago Meirele, Nuno Ferreira, José Alberto Fonseca, João Nuno Matos:
Development of Vehicular Communications based on WAVE (802.11p).

Smail Menani:
Design of an Embedded System IEC61850 enabled.

Felipe Miranda, Rogério Esteves Salustiano and Carlos A. dos Reis Filho:
The Impact of Address-Assigning Heuristics on Network Routing Paths.

Ghodrat Moghadampour:
Web Services for Mobile Intelligent Devices.

Heikki Palomäki:
Multi-feeder systems using RFID

George Palamas, Iasonas Poulis,  George Papadourakis, Manolis Kavoussanos:
Evolution of an Artificial Neural Network of a Simulated Self-Balancing  Bipedal Robot.

Mirco Pieper:
Renesas Technology Communication Devices.

Pieskä, S., Luimula, M., Pitkäaho, T., and Tervonen, J.:
Smart wheel loader based on RFID and positioning technologies.

Carlos A. dos Reis Filho and Rogério Esteves Salustiano:
ParrotNet: a mechanism for data propagation in highly-dense, low-mobility wireless sensor networks.

Vidina Dácil Oliva Rodríguez, Thorsten Knutz, Helmut Dispert:
Development of an Electrode Response Processing System for the Detection of Nitride Ions in Aqueous Solution.

Rogério Esteves Salustiano and Carlos A. dos Reis Filho:
A Java Package for Synchronous Simulation Processing

Filipe E. S. Santos, Joaquim A. R. Azevedo:
Application of ZigBee and Bluetooth to Urban Ambient Monitoring and Guidance.

Luis Melchert Faber Schmutzler:
Motivation, the key to multiplication: Teaching Programmable Logic Controllers with Free Projects.

Index


 

Abtracts


 

Smart wheel loader based on RFID and positioning technologies.

  • Authors:
    Pieskä, S.1, Luimula, M.1, Pitkäaho, T.2, and Tervonen, J.2
    1 CENTRIA Research and Development, Ylivieska, Finland
    2 Oulu Southern Institute, Oulu University, YLIVIESKA, Finland
  • Keywords:

    Wheeled transporters such as forklift trucks and wheel loaders are key elements of material handling and transport in warehouses and factories. These material handling and transport operations are often integrated to information systems of the company. However, data entry has to be done manually, which increases possibility of errors. Bar code readers and data collecting devices are used to automate data entry of material handling operations but still the operator usually has to leave his transporter to scan barcode labels. In harsh industrial environments barcode labels often become unreadable. RFID and positioning technologies give possibility to build smarter transporters which help the operator and decreases mistakes in data input.
    In our paper we describe experiences how RFID technology and positioning technologies can be combined to create smarter transporters. Indoor applications include a RFID equipped forklift truck in a warehouse of a shopping centre and a laboratory mobile robot where positioning is based on hybrid technology including a floor plan, internal measurements and RFID. In outdoor application we introduce a smart wheel loader (Fig.1) which has been tested in wood industry. That wheel loader is based on a location aware system platform, which has a map about the outdoor warehouse area, a GPS system for localization. RFID technology is used both for pallet identification and positioning inside the covered warehouse buildings. RFID technology can be used also in inventory management of different wood products where a special UHF-type paddle device is needed. Economic aspects of proposed smart wheel loader system are also considered

Fig. 1 Smart wheel loader for wood industry and overview of the user interface.

Index


 

Ambient intelligence in mobile field work.

  • Authors:
    Luimula, M.1, Pieskä,S.1, Pitkäaho, T.2, and Tervonen, J.2:
    1 CENTRIA Research and Development, Ylivieska, Finland
    2 Oulu Southern Institute, Oulu University, Ylivieska, Finland
  • Keywords:

    Introduction

    In this paper we will demonstrate the current state of ambient intelligence in industry by reporting about our industrial pilots related to mobile field work. These pilots presented later in details are a mobile maintenance solution 1) for greasing, 2) for electricity meters, 3) for vehicle services, and 4) using RFID technology in production line.

    Industrial pilots

    In the first pilot RFID readers were used in maintenance operation (greasing) in harsh conditions (Fig. 1). The customer wanted to have a system which is proactive based on routes and timestamps, and which identifies greasing spots and stores maintenance information in the database. As a result the system was tested successfully with three mobile devices integrated with a RFID reader.

    Figure 1. Identifying a greasing spot based on RFID technology in harsh conditions.

    In another pilot, electricity meters were identified with mobile RFID readers. Research objectives in this pilot were in gathering meter data with usable way. As a result electricity meters can be read with minimal data input. In addition, a better end user experiences were achieved by using keyboard lights for buttons which are currently available.

    Future visions

    Pilots reported in full paper show that ambient intelligence can already be used in industry in harsh conditions. We believe that also wireless sensor networks are becoming more general in industry like our first experiences show. For example mobile devices can be a part of wireless sensor networks (Fig. 2).

    Figure 2. Mobile device supported with wireless sensor network radio for controlling shutters and for reading temperature measurements.

Index


 

Hand Gesture Recognition using a Lattice Autoassociative Memory.

  • Authors:
    Ana Beatriz Alvarez, José Raimundo de Oliveira
    Department of Computer Engineering and Industrial Automation (DCA)
    School of Electrical and Computer Engineering (FEEC)
    State University of Campinas (UNICAMP)
    Campinas, Brazil
  • Keywords:
    Hand gesture recognition, Human computer interaction, Associative memory.

    The ability to detect, and recognise people, gestures or objects in the environment is a key pre-requisite for many Ambient intelligence applications. The hand gesture recognition has become popular in recent years for being a natural and suitable interaction device in virtual environments. Many hand gesture recognition methods using visual analysis have been proposed: syntactical analysis, neural networks, the hidden Markov model (HMM). However, the development of new structures able to naturally treat this problem of recognition is a broad object of study. This paper presents a Lattice autoassociative memory for static hand gesture recognition from a subset of American Sign Language (ASL). After capturing images, the segmentation phase occurs at pixels from the hands using the color injection method. Invariants moments are calculated forming a vector of descriptive characteristics of the gesture at a particular instant of time.This characteristic vector is used by the trained autoassociative memory to recognize the gesture performed. The structure of this associative memory is based on computational algebra Lattice. The performance of the gesture recognition method is evaluated in applications of recognize gestures of different hand size, shape and skew angle.

Index


 

ParrotNet: a mechanism for data propagation in highly-dense, low-mobility wireless sensor networks

  • Authors:
    Carlos A. dos Reis Filho and Rogério Esteves Salustiano
    School of Electrical and Computer Engineering,
    State University of Campinas (UNICAMP)
    Campinas, Brazil
  • Keywords:
    WSN, master-slave, low-energy, sensors, network.

    The almost boundless scope of applications of WSN (wireless sensor networks) has motivated significant research investments toward dominating the technology in its various aspects. One of the biggest challenges, so far an unsolved problem, has been the lack of appropriate means of providing energy to sustain the network nodes active for long periods of time, without compromising volume, weight, appearance and environment damage as well. To date, the big majority of physically implemented WSN have been powered with batteries, mostly violating at least one of the rules that dictate what a WSN is expected to be. Fortunately, techniques like energy scavenging and energy induction are rapidly evolving to commercially viable levels, reinforcing the actual practicability of WSN within the near future. It is important to notice, though, that while the scarceness of energy remains a fundamental limitation, the success of a WSN relies on the strategy that is devised to make it work. Hence, WSN are application specific systems. This paper addresses a particular type of WSN that features a high density of nodes, which moves however within a short range, have a well-established lifetime and the range covered by its transmitting radio is very small. In operation, the data acquired by each sensor node must reach the base station, upon request of the data collecting base station. Contrary to what happens with the sensor nodes, the base station operates with unlimited energy, so that every message it sends reaches the entire network. The rate of data collection from each node by the base station is low, just a few times per day.
    The strategy developed for the described scenario resulted in a master-slave structure in which the base station plays the full-power master, while the nodes are powered by a battery that must last not less than the time predicted for the existence of the network. A message sent by the master can be received by all nodes in the network, although it is addressed to individual nodes. The master communicates with the slaves by sending two types of messages: One message, “REPORT DATA”, commands the addressed node to transmit the data it has available at that moment. The other type of message is “PROPAGATE”, which is also addressed to a specific node, ordering it to transmit the data it holds in its buffer at that moment. The sensor nodes have limited processing power, its memory is a single buffer restricted to a few-bytes register and the range of the transmitting radio is limited to nearly one meter radius. The sensor node implements two functions: when it is addressed by the master a “REPORT DATA”, it packages the data it has just measured and broadcasts it. The second function is performed in reaction to being addressed a command “PROPAGATE”. In this case, the node acts like a “parrot”, repeating the message is heard last – broadcasts the contents of its buffer. As every node is full time hearing, whenever a message is received that contains no addressee or it is addressed to another node, it simply stores the received message in its buffer. Due to being highly dense and because the sensor nodes move within a relatively small area, whenever a node broadcasts a message it is received by several neighbors, which will be individually ordered to propagate the received information. As a result, after a number of successive commands from the master, the message reaches the collecting base.
    The proposed paper provides a detailed description of the network and its elements. Simulation results and some auxiliary graphs are also provided to illustrate the effectiveness of the proposed approach as a solution to a targeted universe of systems.

Index


 

CHIANTI: Enhancing Robustness of Communication in Disruptive Networks

  • Authors:
    Olaf Bergmann1, Stefanie Gerdes1, Jörg Ott2, Petri Ylikoski2, Nils Seifert3, Caleb Carroll4, Nigel Wallbridge4:
    1 Uni Bremen TZI, Bremen, Germany
    1 TKK>Comnet, Espoo, Finland>
    3 LYSATIQ>GmbH, Berlin, Germany
    4 NOMAD Digital Ltd., Derby, United Kingdom>
  • Keywords:

    Networked computing environments, including Ambient Intelligence and — to some extent — sensor networks and the Internet of Things, often tend to take ubiquituos Internet access as granted in their specific application scenario. But due to the revenue-driven deployment of wireless infrastructure, in the real world, only urban areas and specific locations such as fair grounds, airports etc. already have a chance to provide a user with seamless connectivity. Even then, physical obstacles like large buildings, tunnels or electrical interference can disturb data communication of mobile applications.
    Thus, despite of the high data rates and increased coverage of modern wireless networks, applications still cannot rely on seamless and ubiquitous connectivity. Specifically, applications running on mobile hosts suffer from changing wireless conditions, handovers and disruptions. Although most Internet protocols and applications in theory are designed to use any underlying network, they are quite limited when the network conditions change while the communication is in progress, and often do not survive complete outages that take more than a few seconds.
    The CHIANTI project has investigated the characteristics of those hostile environments especially for moving vehicles and has developed solutions to overcome common failures of applications in disruptive networks.We present the results of an initial trial on a passenger train and during car rides on a highway and discuss the impact of the application-layer CHIANTI proxy on TCPbased communication of mobile users in moving vehicles. The tests show that automatic detection of network links combined with re-establishment of communication between applicationlayer proxies can make applications survive outages of nearly arbitrary length, hence significantly improving the quality of networked applications.

Index


 

Application of ZigBee and Bluetooth to Urban Ambient Monitoring and Guidance

  • Authors:
    Filipe E. S. Santos, Joaquim A. R. Azevedo
    Exact Sciences and Engineering Center, University of Madeira (UMa)
    Funchal, Madeira, Portugal.
  • Keywords:
    ZigBee, Bluetooth, Sensors, Localization, Low power, Air, Pollution.

    Given the growing interest on the population life quality, it’s important to monitor environmental parameters, like gases pollutants, noise, humidity, light, temperature, barometric pressure, and Allergenic pollens, especially in urban areas where concentration of air pollutants are bigger. It’s also interesting to use the everpresent cell phone, to access the monitoring data locally on the street. With this it’s possible to provide a digital street identification system. So would be possible to inform citizens of the environmental quality on that area like joggers, and also guide passerby like tourists. This paper focuses on implementation of a prototype for a Wireless Sensor Network that implements the purposes above. First it’s studied the operation of WSN based on the ZigBee protocol. Second it was chosen and tested a ZigBee and a Bluetooth communication module from the market. After that the same was made for the environmental sensors, giving especially attention to the small size, low power and reasonably priced. Third a circuit board for one node was developed connecting all the components through a small microcontroller. The control was written in C language, assuring efficient energy management, permitting to be powered from a small solar panel. Finally, was developed the software for the system: Java application that records the data from the ZigBee Network on a MySQL database; PHP application for showing the data trough Internet; And a small Java ME application for the cell phone that presents the data on graphics and shows the street name.

Index


 

Didactical principles in the use of embedded servers.

  • Authors:
    Ludo Cambré, Luc Friant
    Katholieke Hogeschool Kempen, Geel, Belgium.
  • Keywords:

    Ethernet is ubiquitous and it is the most widely deployed network in many offices and industrial buildings. Integrated, single-chip microcontrollers are ideal for applications requiring remote control and monitoring. What knowledge is needed in the design of for example: industrial automation, building automation, home control and commercial control.

Index


 

A Sport Activities Monitoring based on Acceleration and Rotation Microelectromechanical Sensors

  • Authors:
    Yull Heilordt Henao and Fabiano Fruett
    Department of Semiconductors, Instruments and Photonics,
    State University of Campinas (UNICAMP), Campinas, SP, Brazil
  • Keywords:
    accelerometer, gyroscope, sport monitoring, virtual instrumentation, wireless communication, real time feedback.

    Although there are several technological tools to aid the training in sports, most of them are high cost solutions and generally very specific to a particular sport, which hinders the diffusion of such technologies. This paper proposes the development of a low cost non-invasive microcontroller Sport Activities Monitoring System (SAMS) prototype, based on acceleration and rotation microelectromechanical sensors (MEMs) for obtaining biomechanical data during training in athletes, without leave the natural environment of their activities. The sensors signals are wireless transmitted from the SAMS to the computer in order to process the data, by an easy and intuitive program interface developed in LabVIEW®. This software saves and displays real time data information in a graphic form. The test were made in two phases: first we use a stationary bike and then we tested the SAMS in a professional cycle track. The system allows the acceleration acquisition in the range form 1.5 G until 10 G and rotation in the span of 50°/s. The maximum transmission range amounts to 70 m. The SAMS has small size (37x49x20 mm) and lightweight (40 g), making it a versatile alternative to aid athletes and coaches during the training, allowing refinement the technique. The SAMS is also a suitable tool for physical education research area.

Index


 

Development of Vehicular Communications based on WAVE (802.11p)

  • Authors:
    Tiago Meireles1, Nuno Ferreira2, José Alberto Fonseca2, João Nuno Matos3
    1 CCCEE, Universidade da Madeira, Portugal
    2 DETI, Universidade de Aveiro, Portugal
    3 Instituto de Telecomunicações, Universidade de Aveiro, Portugal
  • Keywords:

    The use of communication technology is rising within the automobile world. Vehicle manufacturers, highway concessionaries, governments, academic researchers and Industry cooperate in order to develop vehicle communication systems, offering safety services that can reduce the number of road accidents, as well as providing multimedia entertainment services to vehicle passengers. Such communication systems, when embedded on vehicles, will still take many years before having a significant impact on everyday driver habits, due to the current vehicle renewal rate, which presents a large obstacle to the deployment of safety services with pure vehicle to vehicle communications (V2V) networks. We propose a Wireless Access for Vehicular Environments (WAVE) based mixed V2V and Vehicle to Infrastructure (V2I) communications solution for the transitional period, using a simple add-on system to the vehicle. Safety services, such as collision or hard-braking warning, have delay critical requirements. In [1], we proposed a WAVE based architecture and a medium access protocol (MAC) to disseminate time-critical messages for safety services in highways, where the WAVE Control Channel (CCH) interval is subdivided into an infrastructure period, reserved for Road Side Unit (RSU) coordination and beacon transmission by RSUs, and a slotted period, used for rebroadcasting safety messages.

    In [2] we strengthened the RSU coordination scheme.

    A prototype for deploying highway safety services and tolling was developed and its early version demonstrated in Denver [3]. We intent to continue its development and evaluate the protocol proposed in [1].

    Other future work involves using WAVE for multimedia and collaborative car routing applications [4]

    References

    [1] N. Ferreira, T. Meireles, J. Fonseca, J.N. Matos,J.S. Gomes, WAVE Based Architecture for Safety Services Deployment in Vehicular Network, 8th IFAC International Conference on Fieldbuses & networks in Industrial  & Embedded Systems (FET2009), Korea, May 2009

    [2] N. Ferreira, T. Meireles, J. Fonseca, An RSU coordination scheme for WAVE safety services support, Work in Progress Paper, Work in Progress Paper to be published in the 14th International Conference on Emerging Technologies and Factory Automation, ETFA2009, Mallorca, Spain , September 2009

    [3] –D. Carona, R. Abreu, A. Serrador, J.N. Matos, TBC, to be published in the ITS Stockholm World Congress, September 2009

    [4] – DRIVE-IN: Distributed Routing and Infotainment through VEhicular Inter-Networking, http://drive-in.cmuportugal.org

Index


 

A series of insights: location sharing

  • Authors:
    Andre Doria, Daniel Wagner, Iryna Pavlyshak, Mariana Lopez
    Carnegie Mellon in Pittsburgh
    Universidade da Madeira
  • Keywords:
    Privacy, location aware systems, data sharing, locations.

    People share potentially sensitive information with others on a daily basis. This makes it necessary to find a balance between sharing information and protecting private data. We studied location sharing and weighed different factors that impact people’s decisions on sharing their location. We present our findings from conducting 12 interviews on the topic of location sharing. The most important findings from our research are qualitative findings based on the quantitative analysis of the interviews. Some of our findings are supported by previous work; other findings are not found in existing literature and provide interesting avenues for further investigation. In addition to general insights, we present some design suggestions for future location sharing systems. One of these suggestions is the automation of certain rules and the inclusion of humans to solve ambiguous situations where existing systems fail. While our insights are able to stand for themselves, further exploration and verification of our insights in real-world scenarios would be valuable.


Index


 

Hardware Implementation of an Artificial Neural Network with an Embedded Microprocessor in an FPGA.

  • Authors:
    Gisnara Rodrigues Hoelzle1 e Fernando Morgado Dias1,2
    1 Centro de Ciências Matemáticas – CCM, Universidade da Madeira, Funchal, Madeira, Portugal
    2 Departamento de Matemática e Engenharias, Universidade da Madeira, Funchal, Madeira, Portugal
  • Keywords:
    Neural Network, MicroBlaze, Hardware.

    This article describes the implementation in hardware of an Artificial Neural Network with an embedded Microprocessor in a FPGA. The relevance of implementing a Neural Network in hardware is justified by its flexibility, low power consumption and higher performance. Embedding the processor allows achieving the benefits from hardware and from software in a single platform. The implementation with a microprocessor can be quite easy while a regular hardware implementation can be hard to develop.
    The hardware implementation is based in a Feedforward Neural Network, with a hyperbolic tangent as activation function, with floating point notation of single precision. The device used was an FPGA Virtex II Pro XC2VP30, Xilinx with a MicroBlaze soft core processor. The microprocessor soft core subsystem occupied 1766 slices and it used 89KB of RAM for the application, data and results storage (the base value of on-chip memory is of 64KB). The Matlab was used to validate the implementation by comparing it with the results of the hardware solution while using data from a real system. The results show that the implementation is correct and is faster than the Matlab implementation running in a PC with a processor running at 2,8GHz.

Index


 

The Impact of Address-Assigning Heuristics on Network Routing Paths

  • Authors:
    Felipe Miranda, Rogério Esteves Salustiano and Carlos A. dos Reis Filho
    School of Electrical Engineering – State University of Campinas (UNICAMP), Campinas, Brazil
  • Keywords:
    Internet, Routing, Address-assigning heuristic.

    The use of hierarchical identifiers as a mechanism of network addressing has proven to be inadequate as the Internet expands. Since the size of border gateway routing tables is in direct proportion with the growth of the Internet and with the increasing number of accesses, the latency time increases at a higher rate and processing capability is steadily more demanding. Therefore, a more efficient routing mechanism is unquestionably required for the future of the Internet. Peer-to-peer networks brought about a novel routing mechanism, based upon a cooperative behavior, which has shown remarkable results. On peer-to-peer networks, each node retains only a limited number of identifiers in a structure called DHT (Distributed Hash Table), which requires less memory as well as reduces the searching time. There is a key difference between the current Internet routing mechanism and existing peer-to-peer networks: while in the first the identifiers follow a hierarchical addressing, in the second the addressing is flat. By providing the flat nature of the second an appropriate ruling, the routing mechanism can be significantly improved. This paper addresses this particular problem by showing, through simulations, how address-assigning heuristics affect the number of hops on a routing path. It contributes by providing elements that can be used to establish criteria for choosing the appropriate heuristic.

Index


 

A Java Package for Synchronous Simulation Processing

  • Authors:
    Rogério Esteves Salustiano and Carlos A. dos Reis Filho
    School of Electrical Engineering – State University of Campinas (UNICAMP), Campinas, Brazil>
  • Keywords:
    Java, Parallel processing, simulation.

    It is common practice in the process of designing systems, protocols and circuits to implement a simulation phase aiming at to foresee performance and to identify possible flaws. In principle, the simulation of processes, which feature the occurrence of concurrent events, would require some sort of parallel processing in order to be faithful to the order of the events. However, in machines that execute one instruction at a time, this is an impossible strategy. To circumvent this problem, diverse techniques addressing pseudo-parallel processing in simulators have been proposed. This paper presents the development of a Java package with a base structure that allows the implementation of synchronous processing. Among its comprising classes there is a pair of specific ones, namely SynchGlobal and SynchThread, which provides the capability of pseudo-parallel processing. By combining object-oriented paradigm and multithreading, these two classes implement the barrier synchronization technique, which is the essential substrate for developing simulators that feature time coordination. As an illustration of possible applications, two developed simulators are discussed: a logic circuit simulator and a Wireless Sensor Network (WSN) simulator.

Index


 

Web Services for Mobile Intelligent Devices

  • Authors:
    Ghodrat Moghadampour:
    Vaasa University of Applied Sciences
    Vaasa, Finland
  • Keywords:

    A web service is any service that is available over the Internet, uses a standardized XML messaging system and is not tied to any single operating system or programming language. XML messaging can be implemented through XML Remote Procedure Calls (XML-RPC), Simple Object Access Protocol (SOAP), or alternatively through HTTP GET/POST methods.
    The JSR-172 (J2ME Web Services Specification) was approved and finalized by Java Community Process in 2004. The J2ME Web Services API (WSA) extends the J2ME to support web services. The API provides 1) remote service invocation and 2) XML parsing packages, which are required for building clients of web services.
    WSA is designed to work with J2ME profiles based on either the Connected Device Configuration (CDC) or the Connected Limited Device Configuration (CLDC 1.o or CLDC 1.1). The remote invocation API is based on Java API for XML-Based RPC (JAX-RPC 1.1), with some Remote Method Invocation (RMI) classes included to satisfy JAX-RPC dependencies. The XML-parsing API is based on version 2 of Simple API for XML (SAX2).
    The WSA allows integrating fundamental support for web services invocation and XML parsing into the runtime environment of mobile devices. This means that these functionalities do not need to be embedded in each application for resource-constrained devices like mobile phones and personal digital assistants.
    In this paper first the theoretical background of Web services and their specification for mobile intelligent devices are provided and then the required technologies and main steps for developing web services on mobile intelligent devices are described.

Index


 

Motivation, the key to multiplication: Teaching Programmable Logic Controllers with Free Projects

  • Authors:
    Luis Melchert Faber Schmutzler:
    State University of Campinas (UNICAMP), Campinas, Brazil
  • Keywords:
    PLC, Entrepreneurship, Marketing, Projects, Teaching.>

    This paper presents a teaching strategy that was used in Industrial Automation Technology and Automation and Control Engineering night courses. The discipline objective was to introduce students to the use of PLCs, Programmable Logic Controllers. The work was conducted to enable them to develop projects using this type of component, taking into account Marketing, Entrepreneurship and Environment strategies. A package of basic equipment was given, what else more they wanted to build, it would be accepted, but must come on the project. It should use: 1 CLP, 3 tanks of 45 liters, 4 solenoid valves, 6 level sensors, pump, resistance and temperature sensor. The course was divided into 2 parts, one theoretical, 8 weeks, aiming at the definition and planning of project materials and a practical, 12, in the workshops of the faculty. Two exercise lists were given. At the first, students were taken to define the fundamental characteristics of their projects such as: what the machines would produce, how and for what market. In the second, they should draw the systems of their machines and make lists of materials. In the individual test, a favorable situation was presented. Each student should create a way to exploit it commercially, as well as a CLP automated machine to do so. The climate of competition created by the statement led to the creation of interesting projects and many exceeded expectations. The curve of notes was clearly shifted to the positive. All groups showed machines running at the final evaluation. The quality of various projects as documentation photos shows, exceeded expectations.

Index


 

Towards a Reliable, Cost Effective and Easy to Operate Unmanned Aerial Vehicle for Technology Education

  • Authors:
    Johan Dams:
    Vaasa University of Applied Sciences and Vaasa University
    Vaasa, Finland
  • Keywords:
    Teaching, Wireless Communication, Localization, Unmanned Aerial
    Vehicle.

    With the increasing difficulty for many universities to find students willing to invest in a career in technology, it becomes a necessity to attract those capable students whom might not consider technology in the first place. Often the reasons for not choosing an education in technology are fear of the difficulty of the subject, unclear picture of what technology education really entails, or still the perception of technology being boring. A possible way to eliminate some of these reasons in the minds of prospective students is to develop interesting and meaningful yet exciting projects which clearly demonstrate the various aspects of the field. Ideally, these projects should contain a social and/or business oriented relevance in order to attract funding and support from companies or funds.
    In this paper we present one of those projects currently on the way to implement a cost effective and reliable unmanned aerial vehicle (UAV). One of the goals of the project is ease of use, so that prospective or first year students can operate the system with minimal training, while clearly able to observe three main technological fields in practice: embedded systems engineering, software engineering and telecommunications. The project is currently on the way at the Vaasa University of Applied Sciences, Finland (http://www.puv.fi) in cooperation with, and funded for the most part by, WRD Systems, a young British technology company (http://www.wrdsystems.co.uk). The goal for WRD Systems is to apply the results of this project in areas which can benefit from aerial observations such as crop yield estimation, forest monitoring, wildlife tracking and search and rescue.

Index


 

Multi-feeder systems using RFID

  • Authors:
    Heikki Palomäki:
    Seinäjoki University of Applied Sciences
    Seinäjoki, Finland
  • Keywords:
    RFID, multi-feeder, distributed intelligence.

    At Seinäjoki University of Applied Sciences we are researching and developing a new kind of multi-feeder system. Typically in these systems several equivalent feeders are existing, which are serving many separate and different objects, which can be trucks, humans or animals. The objects can fetch its individual predefined portion bit by bit using various feeders. The objects are identified by RFID tags. In traditional systems the master controller has a required database and it controls slave feeders, which have only a few outputs and RFID reader. In the new system that will bepresented in this paper the database and other intelligence is distributed in feeder units without the master. Using a modified multi-master Modbus, the units are updating its databases continuously dividing all history feed data and events.
    The main target of this research is to develop low-cost, universal multi-feeder units, which are suitable both for very small and very large systems.

Index


 

Design of an Embedded System IEC61850 enabled

  • Authors:
    Smail Menani:
    Vaasa University of Applied Sciences
    Vaasa, Finland
  • Keywords:

    The paper describes the recent development of the 5th generation of the Botnia soccer robot. The main electronic board of the5th generation robot is based on a single Xilinx FPGA with a soft CPU and an ARM microcontroller. Digital accelerator sensor and a gyroscope are added to the system for better navigation and control. The key development will be in the communication system side. The current generation support the IEC61850-Ethernet based interface to enable not only debugging and reporting of IEC61850 events and send them to an OPC server, but to permit to the Botnia robots to communicate with industrial IEDs in power substation such as protection relays. The implementation of the IEC61850 is divided into two main projects: The BROMDI project, which will focus on the description of the Botnia robot elements according to the IEC618510 definitions. And the BROFICP project to describe the impact of the Botnia Robot System on the IEC61850 communication protocol. These projects are at a very early stage to disseminate any results but the success will depends on the interoperability test using either VAMP or ABB relays.

Index


 

Teaching robotics and embedded systems as a learning chain of different subjects

  • Authors:
    Margus Ernits, Kristiina Hakk, Kalle Tammemäe:
    Estonian IT College
    Tallinn, Estonia>
  • Keywords:
    Teaching Embedded systems

    Study programs in Estonian IT College (EITC) cover Information systems development, administration, system analysis and technical communication. Studying embedded systems is a relatively marginal part of the curricula. However, EITC runs Robotics club which is popular amongst students of all curricula.
    Students appreciate learning programming by building robots and gaining practical experience during this process following learning-by-doing or serious-gaming study model.
    Curriculum development is a continuous process of improving the coherence of the curriculum materials to improve the relationship between the subjects. Since robotics is an interdisciplinary field covering different subjects areas, it can be used as glue to create a link between different courses over curriculum.
    Using virtual and game-like environments is a contemporary approach for teaching programming. Combining programming with real hardware provides valuable platform supporting studies of math, physics and other IT-related topics. Separated experiments on base on gaming elements have been provided since establishment of Robotics club in 2002.
    Developed environment and contemporary teaching methods results motivated and skilled students. Special robotics classes for children are provided by several Universities to raise interest in IT and science in Estonia.
    Experience can be further extended by attracting students and lectures of other areas. For example, creating smart products in cooperation with students of the Estonian Academy of Arts was an challenging and motivating process both for students and tutors. Joint efforts of people with different background expanded their horizon of knowledge and provided a good product development experience for those who participated in the project.
    Robotics and embedded systems can be used to make studies more interesting both for students in different fields, for children and tutors. Cooperation of curricula tutors is a key factor to achieve results and link curricula subjects. We expect to have results of coherent teaching to be available in 2010.

Index


 

Evolution of an Artificial Neural Network of a Simulated Self-Balancing Bipedal Robot

  • Authors:
    George Palamas, Iasonas Poulis,  George Papadourakis, Manolis Kavoussanos:
    Technological Educational Institute of Crete
    Heraklion, Crete, Greece
  • Keywords:

    Over the course of the last decade, interest in bipedal humanoid robots has increased exponentially, ranging from models that simply stand upright to others performing series of complex movements that endeavor to duplicate human motion.This paper presents an attempt to achieve a stable bipedal balance by using a simulated humanoid robot consisting of joints and body parts designed to match those of a human as close as possible without compromising its overall functionality. Genetic Algorithms and Artificial Neural Networks are introduced to the model in order to calculate the optimal weights to apply to the bipedal's joints, such that a stable stance without wavering can be achieved. The bipedals' stability is then checked against a pre-existing optimal solution, thus calculating the fitness of every attempt. The above simulations are all carried out using C# in conjunction with XNA, ODE and Matlab.

Index


 

Implementation of a platform independent client software for the GO Bluebox System

  • Authors:
    Nils T. Kannengießer1, Thomas Ladehoff1, Thorsten Knutz2, Helmut Dispert1
    1 Kiel University of Applied Sciences
    2 GO Systemelektronik GmbH
    Kiel, Germany
  • Keywords:
    Data Acquisition, Sensors, Actuators, Java Technology, Java Applets.

    The BlueBox is a dedicated data acquisition system produced and commercialized by the North German SME GO Systemelektronik[1]. GO develops innovative systems for industrial measurement and control applications, starting out with individual single sensor solutions and expanding to the BlueBox, a complex general purpose system. Typical applications for the BlueBox can be found in the area of environmental monitoring, e.g. in the aquaculture industry.  In this paper we discuss a software-based extension and interface designed for the BlueBox. The software, developed in Java, allows control of the BlueBox as well as reading and displaying BlueBox-acquired sensor data.

Index


 

Optimizing Embedded Software - A Look at the NEON SIMD unit in the ARM Cortex Family of Processors

  • Authors:
    Johan Dams:
    Vaasa University of Applied Sciences and Vaasa University
    Vaasa, Finland
  • Keywords:
    Embedded Linux, NEON, ARM, SIMD, Optimization

    The cellphone is quickly becoming the ubiquitous device providing both phone conversations as well as Internet access. There are approximately 4 billion mobile phones in use, compared to about 1 billion PC’s and Laptops. In the foreseeable future, the mobile phone will become the device of choice to access the internet. Some problems that appear when using cellphones for Internet access are the performance requirements, and the expected battery life. With customers demanding mobile phones capable of playing high quality video and audio, as well as secure Internet communication from the mobile platform, manufacturers are trying to get as much performance as possible into the small device, but this often goes at the cost of battery life.
    In order to maximise the available computing power, yet still provide long battery life, one has to perform optimisations on the software level. Power management is one of the methods to maximise battery life by turning of services when the device is idle. The main issue however is how to be able to perform processor intensive tasks, such as video decoding and playback, as efficiently as possible.
    The ARM Cortex family of processors contains a SIMD (Single Instruction, Multiple Data) unit known as NEON, primarily used for accelerating media applications such as MPEG decoding. It offers however a high degree of flexibility, and can be adapted quite easily to other applications besides multimedia acceleration.
    In this paper we will present some performance comparisons between NEON optimised code and code using just ARM instructions. We will provide a look at how to optimize existing software, especially looking at the Linux operating system.

Index


 

Recent progress in magnetic-inductive flow metering devices (MID)

  • Authors:
    Rainer Bollmann2, Wolfgang Stade2, Gerd Stange1, Thomas Zelenka2:
    1 Kiel University of Applied Sciences, Kiel, Germany
    2 F&E-Zentrum Fachhochschule Kiel GmbH, Kiel, Germany>
  • Keywords:
    Magnetic-inductive flow sensor, Permanent magnet, Continuous flow

    Flow sensors for the measurement of volume or mass flow rates of gaseous or liquid media are good examples of ubiquitous sensors. They are to be found not only in almost any industrial process but rather in any process at all, where material and/or energy flow rates play a role. In most of the application fields aqueous media have to be measured.
    For volume flow measurement  the magnetic-inductive principle has distinct advantages, since it  allows for smooth inner walls without any obstacles or dead space. It doesn’t depend on the temperature, pressure, viscosity  of the flow medium. Interestingly it doesn’t even depend on the flow profile. It however requires a minimum conductivity of about 5 μS/cm. Fortunately the majority of the aqueous fluids meet this requirement so that  the magnetic-inductive principle has a big market potential.
    Only recently it has been shown that contrary to the existing systems which  apply time dependent fields in order to cope with the error signals caused by electrochemical fluctuations of the solid-fluid interface double layer potentials, now systems working with a time constant magnetic field come into reach. Their additional advantages are manifold: Application of modern rare earth permanent magnets allows for fields two orders of magnitude above that of existing devices with a correspondingly higher resolution, for continuous in contrast to discrete measurement and for zero energy. All these arguments in combination with an easy construction make permanent magnet driven flow sensors ideal candidates in ambient intelligence systems.

Index


 

Development of an Electrode Response Processing System for the Detection of Nitride Ions in Aqueous Solution

  • Authors:
    Vidina Dácil Oliva Rodríguez1,2, Thorsten Knutz3, Helmut Dispert1:
    1 Kiel University of Applied Sciences, Kiel, Germany
    2 Universidad de Las Palmas de Gran Canaria, Las Palmas, Spain>
    3 GO Systemelektronik GmbH, Kiel, Germany
  • Keywords:

    Nitrids are significantly present in nature. For example they can be  in  subterranean waters. Since the concentration over certain levels decrease the quality of the water and can necatively affect the ecosystem, a cost-effective method to monitor this parameter is  needed.  To fulfill this requirement, a complete system is developed in this study for identyfying ion nitrids present in a NaNO3 solution by means of an the Ag/AgCl  four-electrode cell response.  The cell would be connected to a potentiostat.

    The method suggested is based on Impedance Spectroscopy, which is a widespread tool to measure the impedance from the electrode response and extracts its main properties and parameters.  The major advantage seen in this technique is the possibility to calculate an equivalent circuit that allows to simplify the interpretation of the data collected.   The adjusted equivalent circuit of the interface electrode-electrolyte for this especific case is then simulated with the software PSPICE. As a result of this simulations a maximum length  sequence (MLS) is defined as a stimulus. The MLS is a pseudo-random noise signal with similar properties as true noise that improves the signal-to-noise ratio in a noisy environment.

    Except from adapting the signals to the potentiostat, all the mandatory operations-stimulus generation and processing- will be performed by a microcontroller . Analog circuits implement the adaptation. The processing procedure is modeled and validated using Matlab.

    Findinds show that a stabilisation of the electrode response is necessary in order to identify the ions. Future work should focus on the sensor behaviour.

Index


 

Renesas Technology Communication Devices

  • Authors:
    Mirco Pieper:
    Renesas Technology GmbH
    Ratingen, Germany
  • Keywords:
    Embedded Systems, Modem design basics

    Over the last years the request for communication devices is growing very much and in parallel microcontroller and microprocessor devices became more and more powerful. Related to several applications which require wired or wireless communication various modulation techniques and a wide range of high integrated devices are in use. The range of modulation techniques varies form high data rate to extreme robust communications. Depending on the modulation technique used in a modem device, the computational effort can be very huge. Therefore, embedded transmitter and receiver require optimized designs with respect to i.e. computational effort, speed, and memory usage.
    An Overview of general modem design concepts, design basics, optimization methods, and related microcontroller devices for embedded communication devices will be given. Although the level of integration is very high, a few parts of the communication chain are often realized by external devices such as e.g. anti aliasing filter, automatic gain control (AGC), and power amplifier. Therefore, an attractive AGC implementation based on general purpose operational amplifier will also be presented.
    Findings show that a stabilisation of the electrode response is necessary in order to identify the ions. Future work should focus on the sensor behaviour.

Index