Mics Newsletter Mar09

NEWSLETTER March 2009 • The measurement technology used in the PermaSense program could pave the way for a future real-time monitoring of potential rock fall in areas of high vulnerability (see the picture taken on the Matterhorn field site) Pages 2&3 • Since its launching in 2007, the Swiss Experiment project has made a name for itself and the community is now understanding its purpose and value. Page 4

• This year, the NCCR-MICS offers for the 6th time 6-10 weeks internships, between June and mid-September, in laboratories to interested undergraduates. Apply now! Pages 5&6 • The move towards Phase 3... Find out about MICS main objectives for the coming years. Pages 9&10

page  | March 2009

MICS NEWSLETTER

The Quest for Uptime in the Mountains After an initial phase of prototype development and experience with deployments on the Jungfraujoch the PermaSense project has developed rapidly, increased its team and is now successfully operating two high alpine field sites based on a second generation technology platform. The goal of the PermaSense project is to provide long-term high-quality sensing in harsh environments, to obtain better quality data more effectively and make measurements that have previously been impossible. The measurement technology used in the PermaSense program could pave the way for a future real-time monitoring of potential rock fall or large scale slope instability in areas of high vulnerability. Therefore, reliability and precision of the data collected are of primary importance for a successful exploitation in geoscience and risk-assessment. At present, a network of miniature, autonomous wireless sensors is recording diverse parameters related to freezing processes as well as rock movement and temperature in the permafrost area of the Hörnligrat ridge of the Matterhorn. A second network exists around the Sphinx observatory and high altitude re-

search station on Jungfraujoch. The data is collected at a base station and relayed to a database server using GSM/GPRS or Wireless LAN. For much of the technology involved, PermaSense is relying on seasoned developments from within the MICS community: The Dozer protocol (Wattenhofer) is driving the low power data gathering based on TinyNodes (Shockfish). The Global Sensor Network (Aberer) is an integral part of the data backend and the Deployment Support Network (Thiele) is heavily used in development and system testing. THE EXPLOITATION PHASE After initial installation of sensors in September 2007 and simple data logging over the following winter, the Matterhorn field site has been equipped with wireless sensors in July and August 2008. It has been operational since, continuously delivering data that will improve the understanding of processes occurring in permafrost regions. More recently the Jungfraujoch field site has also been equipped with wireless sensors in February 2009. In effect the PermaSense project has now moved on from gaining and initial understanding of the requirements but also research culture of the interdisciplinary project partners, technology develop-

page  | March 2009

ment, research, system integration and test to a phase of operations, maintenance, data management and exploitation. This is significantly different from shorter term experiments, be it in or out of the lab, and posing real challenges that are calling for innovative solutions. Apart from pretty “standard” 24/7 uptime of relevant infrastructure, e.g. servers, network trunks and sensor nodes, gathering data over a period of multiple years means a lot of logistics around equipment, data and configurations. A COMPLEX PROCESS A simple exchange of a sensor node not only requires a climbing trip (possibly with an helicopter flight), trained personnel and favorable weather, but also a slurry of actions and configuration changes in documentation data bases and server infrastructure. Remember that the actual goal would be to “just” click on a simple query hidden behind a colorful icon in a web browser to display a set of data captured in the field, ready to be analyzed. This data should automatically include the appropriate correction functions, calibration parameters, origin, type, circumstances, date of the sensor installation and much more that is today primarily kept on paper, emails and whiteboard

MICS NEWSLETTER

sketches. Thus, the meaning of meta-data is extended beyond just auxiliary information about the measurement site and methods to a complete documentation of the instrumentation history to allow a traceable and reproducible conversion of measured “digital” numbers to relevant information. So clearly, the list of todo’s for the PermaSense project is not done yet, now that sensor nodes have been deployed and data is streaming off the mountain in minute intervals. For PermaSense the next steps have already begun that will allow turning data into relevant information and so facilitating geo-science research. Joining forces with Karl Aberer’s team and also with the Swiss Experiment on questions of data management, procedures and the integration of the necessary infrastructure is expected to allow obtaining both a critical mass and the necessary expertise for the tasks ahead. An improved ability to document and process sensor network data streams is urgently needed now: With six months of near-continuous data gathered to date, the thorough geo-science exploitation of results is about to begin. Meanwhile, measurements must continue because environmental research often requires several years of data, and likewise, future

warning systems will have to demonstrate their quality by long-term reliable operation – the quest for uptime in the mountains. Jan Beutel (ETH Zurich) & Stephan Gruber (Uni Zurich) Project webpage: http://www.permasense.ch Online data access: http://tik42x.ee.ethz.ch:22001 PermaSense in a Nutshell Computer scientists, electronics engineers, and geoscientists alike, profit from this interdisciplinary project. Stephan Gruber, Senior Assistant at the Department of Geography at the University of Zurich, and head geoscientist of the PermaSense project has been studying permafrost at high altitudes for years. In order to get a reliable model of the permafrost, authentic data is needed, something the researchers are aiming to get from their new measuring system. The researchers have already come up with some promising first findings, but an exact analysis will only be possible after a few months of continuous collection of data. Results are expected to show as from mid-June.

page  | March 2009

MICS NEWSLETTER

Swiss EXPERIMENT makes a name for itself Michael Lehning is an atmospheric physicist and environmental scientist. He works on snow cover modeling and on snow-atmosphere interaction. He leads the research unit «Snow and Permafrost» at the Swiss Federal Institute for Forest, Snow and Landscape Research (WSL). He is also the Swiss Experiment project’s head, partly funded by the MICS Center... - The SwissEx project was launched in 2007. Where does it stand now? We have just had the annual review recently, and has been good to see how well the launch and progress of SwissEx was received. It is very satisfactory to see that a lot of individual projects now want to be integrated or related to SwissEx and how SwissEx serves the environmental science projects. We receive plenty of demands e.g. for the Sensorscope stations. Finally, I can say that the community is now understanding the purpose and value of SwissEx. - Is WSL contribution to the SwissEx project expected to develop further in the future? WSL acts as bridge between pure theoretical science and the pratical implementation of scientific find-

ings. Monitoring environmental aspects is part of its mission. Thus, we will work hard to have the infrastructure and support through SwissEx continuing even after the current project may end. - How are the numerous partners of the SwissEx getting together? Are there productive results? Everybody is at work, there is no race for notoriety. We operate a database and all the partners collaborate. Every group collects the data it needs. Therefore, it guarantees that all information is being used and analyzed. I also want to

mention that MICS has developed and is developing a lot of the key technology to be used for environmental sciences in general and for SwissEx in particular. - Recently, the SwissEx was presented at the World Economic Forum in Davos. What kind of impact can it have? Yes, SwissEx was presented at both the Microsoft house at WEF and at SLF Davos, but it is always hard to judge the impact this will have. We might get some feedback in the future...



Interview by Florence Luy

© Dominique Meienberg/SNF

page  | March 2009

MICS NEWSLETTER

APPLY NOW FOR THE SUMMER INTERNSHIP PROGRAM! This year, the NCCR-MICS offers for the 6th time 6-10 weeks internships, between June and mid-September, in laboratories at EPF Lausanne, ETH Zurich and the Universities of Bern, Neuchatel, Lugano and Basel to interested undergraduates who would like to gain experience working in a research environment. The internships are designed for university students who are interested in pursuing a PhD after graduation, but would like to experience a uni-

WHAT DO THEY SAY ABOUT THE PROGRAM... • RAPHAELLE LUISIER was finishing her first year of the Master program when she applied for an internship. She was assigned a topic more or less close to her studies in life sciences for ten weeks. “Participants have to write a report at the end of their training”, explains Raphaelle. “It was the opportunity to prepare a document in English, but also to have something that could be useful for the future”. The title of her report? “Design and fabrication of an hydrophilic mi-

versity research environment first hand before making a decision. The salary is around Fr. 1900/month. A special Female Undergraduate Internship is also proposed, so to encourage women to pursue within the academic environment.. The internships are offered on a competitive basis and cover a wide range of research topics in the field of information and communication sciences. While many of the research projects are technical in nature, successful candidates do not necessarily need an engineering background. Last year, out of 23 internships

crochannel for complete on-chip immunoassay”. The young lady is very enthousiastic about her experience: “I really shared the labo’s life, while having a lot of freedom”. Today, she is about to finalize her Master work - on 3D skin models - in

offered, 17 applications were accepted. A report reflecting the participants comments, showed that most students thought they had learned a lot regarding methodology and work organization. On the other hand, most of them would have liked to see the requirements better defined. Interested students are requested to apply now at http:// www.mics.org/micsEducation. php?action=undergrad. They will be notified if their application is successful. Florence Luy

the Queensland University of technology, in Brisbane (Australia). She wants to pursue an academic career. “The internship convinced me of that choice. I advise everybody to have this kind of experience before deciding to pursue with a PhD”, she says. (read next page)

page  | March 2009

• ROBIN SCHEIBLER chose an internship to know better about the academic world while having a paid summer job. He was satisfied with the topic that he was given since he knew the professor and could discuss with him. Actually, Robin is accomplishing his Master work at IBM Zurich. He thinks the students should be better informed about the internships: “To receive just one mail is not enough. More advertisement needs to be done”. He also suggests that participants interact more with each other so that they could exchange about their experience. For Robin Scheibler, the internship

PULSAR GETS MICS SPINFUND FUNDED Under the project name Pulsar, MICS researchers Nicolas Burri, Remo Meier, and Dr. Pascal von Rickenbach have been granted a MICS Spinfund. The project is hosted by Prof. Wattenhofer at ETH Zurich and will commercialize an academic prototype of a new multimedia streaming technique in the months to come. Multimedia streams over the internet have recently become very

MICS NEWSLETTER

enabled him to practice all of what he had learned previously, as well as to discover the PhD students environment. • POLINA MAKEEVA chose this MICS internship because she wanted to gain more experience in the research work. After three years of Bachelor, she still felt a lack of experience in developing a complete project, from design to testing and obtaining experiment results.

to implement a “distributed hash table (DHT) lookup in peer-to-peer networks”. Although she suggests that a talk with the supervisor should take place before the internship to better clarify the work, she says she acquired a useful experience in reviewing a lot of theoretical material.

She noticed that the subject of the project was a little bit different then the one described in the list of available topics, but it was even more interesting for her. She had

Later, the internship helped her for her courses and other projects, as well as for her search for other internships and jobs. “I recommend this kind of experience to everybody”, she concludes. Interviews by FL

popular. Especially events like the inauguration of Barrack Obama provoke a massive user interest. Standard server-based streaming solutions are often not capable of providing a large audience with the requested quality of service.

from other users consuming the same stream. As a consequence using the already existing bandwidth and computation power more users can be served with audio and video streams in better quality than it was possible so far.

Pulsar solves these scalability problems by incorporating peer-topeer technology to increase the streaming quality and stability of popular audio or video content. Users viewing a video or listening to an audio stream do no longer download the whole content from the server of the provider but also

During the time of funding, Pulsar will move from a prototype to a market-ready product and develop business relations with content providers. Initially, the focus will be on the Swiss market to acquire first customer references and thereafter expand the business activities to further European countries and the US.

page  | March 2009

MICS NEWSLETTER

Fostering Interdisciplinary Research within MICS Alcherio Martinoli holds a M.Sc. degree in Electrical Engineering from the ETH Zurich, and a Ph.D. degree in Computer Science from the EPF Lausanne. After a stay of about three and an half years at the California Institute of Technology, he joined again EPFL as an SNF Junior Professor at the School of Computer and Communication Sciences (I&C) in Fall 2003.

sional and academic reputation in distributed, possibly networked and self-organized, intelligent systems able to collaborate to achieve a common mission. His work addresses system engineering both at the hardware and software level, model-based and data-based methods for design and optimization of the system with particular emphasis on distributed control algorithms. Alcherio joined the NCCR-MICS at the beginning of the Phase II, starting in November 2005. Within the Center, he has brought the multi-disciplinary system design expertise typical of robotics and thus contributed to move the center of gravity of MICS towards more experimental research work. COLLABORATION WITH ENAC

During his master studies and even more so during his following academic career, Alcherio Martinoli has been fascinated by robotics and, more generally, intelligent devices able to perceive, compute, decide, and act in the physical world. More specifically, he has built his profes-

Among several parallel projects during his SNF junior professor period, the one sponsored by MICS focusing on distributed odor localization using multi-robot systems has promoted the collaboration with the School of Architecture, Civil, and Environmental Engineering (ENAC), because of the availability of expertise and tools in wind engineering, in particular within the laboratory of Prof. Marc Parlange, also involved in MICS since the Phase II.

Starting from experiments with multi-robot systems within the ENAC wind tunnel and stimulated by exciting activities of other MICS laboratories working in environmental monitoring, Alcherio realized that the scientific knowledge and technology developed in his group could have been applied more and more to environmental, civil, and architectural research problems. TOWARDS NEW ACTIVITIES This initial intellectual curiosity has become a more concrete daily reality for Alcherio since May 2008, when he has been appointed as Associate Professor at ENAC. The moving of his laboratory has been completed in Fall 2008 and while the core activities in distributed robotic systems have been conserved and actually reinforced, new neighboring activities focusing on sensors and actuators networks are emerging. For instance, Alcherio’s laboratory is now involved in the Swiss Experiment project focusing on environmental monitoring in the Alpine Region or, in an original project investigating distributed mechatronic solutions for increasing the stability of long-span bridges solicited by strong wind fields. On the teaching front, he has just started with his collaborators to

page  | March 2009

teach an innovative, introductory course in embedded systems for civil and environmental engineers at the bachelor level and will resume from next year a course in distributed intelligent systems offered to both I&C and ENAC students at master level, thus even further promoting the cross-fertilization of the two schools and respective core research areas. Florence Luy

EPFL RESEARCH DAY 09 EPFL Research Day will take place on April 30 and will be dedicated to robotics at large. Prof. Gaurav Sukhatme (University of Southern California, Los Angeles), an expert in distributed robotics, is playing a similar role as Alcherio Martinoli in the NCCR-MICS for the Center of Embedded Networked Sensing at UCLA. He will be one of the speakers of the Research Day and will emphasize this cross-fertilization trends between distributed robotics and wireless sensor networks, finding exciting deployments in environmental and civil engineering applications. MICS will be represented at the EPFL Research Day by Alcherio Martinoli and other laboratories showcasing activities directly centered on robotics or crucial for the progress of this field.

MICS NEWSLETTER

MICS WORKSHOP AT EPFL This year, the NCCR MICS is organizing its annual event in the form of a Workshop on 9 and 10 June at EPFL. The goal of this workshop is to present some of the main research results of MICS Phase 2 and to provide an outlook to Phase 3 (Nov 2009 - Oct 2013). The keynote speakers will be Prof. Anja Feldmann, TU Berlin, and Peter Landrock, CEO of Cryptomathic. The technical sessions will be comprised of 2-3 presentations each. These presentations will be selected based on 1-page proposals, with a priority given to those that would contribute most to cross-pollination. A 1-page proposal should contain: - Title of the presentation; - Authors of the presented work, affiliation; - Name of the potential speaker (a PhD student, typically); - Abstract of the presentation (up to 20 lines); - If and where this work has already been published; - Explanation of why this presentation could be of interest to researchers in other fields of MICS;

concrete examples of other fields that could benefit from this work are most welcome. The presentation proposals will be reviewed by an ad-hoc committee; non-selected presentations will be presented as posters. All current MICS PhD students are expected to provide proposals. Please send these by 27 March to Jacques Bovay (jacques.bovay@ epfl.ch). Jacques Bovay For further information: www.mics.org

page  | March 2009

MICS NEWSLETTER

nCCR-MICS: THE MOVE TOWARDS PHASE 3 NCCR MICS has realized an excellent academic research program in its first two phases, led to the creation of strong research partnerships among Swiss researchers in different institutions and locations, to an extent beyond what could have been expected, and created important and long-lasting multidisciplinary synergies, by opening up to disciplines beyond computer science and electrical engineering. In particular, the decision to concentrate in Phase 2 on wireless sensing networks for environmental monitoring has proven to be a very wise choice. The challenge for the NCCR-MICS in a 3rd phase will be to maintain the momentum, the breadth of the project portfolio, and preserve a sense of cohesion. As the Centre goes towards its planned end, an important aspect will be to define a coherent blend between the continuation of present activities and the start of new work. OVERALL VISION The main objectives of MICS phase 3 are to consolidate the research work of phases 1 and 2, including the technology transfer to industry,

the establishment of long term research and collaboration structures across institutions, and to ensure the long-term impact of the program beyond its duration as a centre. To that extend we will • further develop the applications, in particular in environmental monitoring • complement the existing technological developments through selected research efforts addressing novel challenges from applications in MICS complementing and completing earlier research • pursue various exploitation and technology transfer strategies • develop strategies to connect to emerging research initiatives and research programs, in such a way that they will draw optimal benefit from MICS. COMMON THEME In MICS phase 3 we will study the whole data life cycle using wireless technology for capturing information from the environment, securely transmitting it through heterogeneous networks, processing the resulting data and provide platforms for personalized use and decision-making. Environmental monitoring for scientific purposes will remain the specific focus to exemplify this approach and a substantial fraction of our effort will directly or indirectly contribute to

this application space. The Swiss Experiment will be used as a platform to perform interdisciplinary research for understanding of complex environmental systems in their entirety using wireless information and communication systems. It has started to develop critical activities in the areas of sensor node development, data technology, education and community building which will be further pursued in phase 3. Different but related technologies will be explored in collaboration with other research programs for achieving mutual benefit. We identified three main research challenges to be addressed in order to deal with the expected growth of wireless sensor network applications: • data management to deal with the massively growing sensor data volumes, • practicability of wireless sensor network technology to enable its wide-spread use, • wireless security to deal with potential threats. In addition, we will further pursue low-power UWB communication as a case for building a technology portfolio of intellectual property in an academic research environment. SPECIAL PROGRAMS The main objectives in technology

page 10 | March 2009

MICS NEWSLETTER

transfer will be to further intensify the successful collaboration with industry, to transfer results through the creation of spin-offs, and to create IP portfolios (in particular in the UWB domain). In education, we will particularly focus on the problem of attracting more young people to the field of information technology while ensuring the continuation of the successful existing programs and impact on the curriculum. This will also be supported by a special action of the promotion of women program having a bus travelling to secondary schools all over Switzerland, in order to promote IT to young people. In public relations, we will in particular take advantage of the high visibility of issues related to environmental concerns. In summary, we are very satisfied that we have converged to a high quality research program with an exciting and relevant application dimension and high degree of collaborative and interdisciplinary research. We are confident that, with the proposed research program and structural measures, we will consolidate the long-term impact of the NCCR MICS centre at the participating institutions, as well as for the IT field in Switzerland. Jacques Bovay

NEW PUBLICATIONS Conference papers: • Gilbert, Seth; Guerraoui, Rachid; Kowalski, Dariusz; Newport, Calvin, Interference-Resilient Information Exchange, IEEE InfoCom 2009; Rio de Janeiro, Brazil, April 19-25.

conferences • 8th ACM/IEEE International Conference on Information Processing in Sensor Networks, San Francisco (USA), 13-16 April. • 7th International Conference on Pervasive Computing, Nara (Japan), 11-14 May. • 10th International Conference on Mobile Data Management: Systems, Services and Middleware, Taipei (Taiwan), 18-21 May. • MICS Workshop, EPF Lausanne, 9-10 June.

Performance Computing, Networking and Communication Systems, Orlando (USA), 13-16 July. • International Conference on Wireless Algorithms Systems and Applications, Boston (USA), 16-18 August 2009. • S-CUBE, 1st Conference on Wireless Sensor Network Systems and Software, Pisa (Italy), 7-9 September. • 9th International Conference on Peer-to-Peer Computing, 8-11 September, Seattle, Washington, USA.

Editor : Florence Luy Mail : [email protected] The National Centres of Competence in Research are a research instrument of the Swiss National Science Foundation

• 6th International Conference on Networked Sensing Systems, Pittsburgh (USA), 17-19 June. • IEEE International Symposium on Information Theory, Seoul (Korea), 28 June-3 July. • International Conference on High

EPFL IC NCCR MICS Station 14 CH-1015 Lausanne Tel +41 (0)21 693 8106 Fax +41 (0)21 693 8140 www.mics.org | [email protected]