Uppsala VINN Excellence Center for
Wireless Sensor Networks
WISENET

WISENET Nordic Darkness School November 8-9, 2010

Lectures

WSN 2.0

Per Gunningberg, Department of Information Technology, Uppsala University
Abstract
It is time to push wireless sensor networks out of the research labs into real use in industry and public organizations. The top main barrier for industry to start to use WSN is reliability. It is followed by cost, security, standards and ease of use. To get a wider acceptance, researcher need to create the next generation WSN - here called WSN 2.0 - where these problems are put forward.

The Uppsala Wisenet center is working closely together with industry on these issues. In this talk I will outline the research challenges ahead for a more widespread usage of WSNs.

Transmission strategies for massive wireless networks in stores

Nils Hulth, Pricer AB
Abstract
Electronic shelf labels (ESL) are used to display prices on the shelf edge, thus saving labour and increasing response time for retailers. An ESL installation can consist of up to 100.000 battery powered nodes in the same store, where each label has to run 8-10 years on one set of batteries. Label communication is two-way with a star network of ceiling mounted tranceivers transmitting data to the labels and receiving acknowledgement signals. The presentation will give details in of the transmission and retransmission strategy and how that influences speed, response time and battery life.

Discussion question: How can the protocol be made optimal and tunable to different stores in order to maximize speed, response time and battery life?

Atmospheric sensing in tornados with unmanned aircraft

Tim Brown, University of Colorado Boulder, USA
Abstract
This workshop describes the networking and communication challenges with airborne sensors in small (10kg) unmanned aircraft. In particular we discuss the roles of flight dynamics, interference, limited spectrum, and flight safety requirements. Mobility is a key feature of these networks and can be exploited to improve network performance. Experiments with atmospheric sensing in a variety of environments will be described including a recent "tornado chasing" campaign to understand the origins of tornados.

Discussion question: What is different and what is common between airbornes sensor networks and other sensor networks? How do these differences affect the node and network design?

Denial of service challenges in wireless networks

Tim Brown, University of Colorado Boulder, USA
Abstract
Wireless networks are well known to be susceptible to jamming. However, as communication and networking protocols increase in functionality and complexity, wireless networks are susceptible to "intelligent jamming". Intelligent jamming uses protocol behavior against a victim network and enables attackers to disrupt communication with much less effort. We discuss these issues from both a red team (attacker) and blue team (defender) perspective. Specific protocols such as 802.11, 802.16, TCP, DNS, and others will be examined.

Discussion Question: Are wireless sensor networks more or less susceptble to denial of service attacks and how can they be better protected?

Natural disaster science – Potential of sensor development in the management of water-related disasters

Sven Halldin, Department of Earth Sciences, Uppsala University
Reading material
A Science Plan for Integrated Research on Disaster Risk – Addressing the challenge of natural and human-induced environmental hazards from International Council for Science.

Internet of Things

Heikki Seppä, VTT, Finland
Abstract
I will shortly describe the historical background of RFID and, of course, basic principle and technology in general. RFID is based on reflection where the reader creates either magnetic field or electromagnetic waves and a tag phase or amplitude modulates the field. The modulation is detected by the reader. The semipassive tag in powered by the reader whereas a semipassive tag consists of a battery, or other energy source. In both cases the basic principle is the same and the reader even does have know whether a tag is supported by a battery or not. Semipassive tags are used in data loggers because RFID reader is not present. RFID reader as a wireless communication technology either in the proximity mode (10 cm) or in a long distance mode (maximum 10 m in the case of passive tag) In paying and ticketing etc. applications 13.56 MHz is used whereas in logistics they work in the UHF band. Soon mobile phones are equipped RFID reader (acting as a tag as well. This technology is called NFC (Near Field communication). All the major mobile phone companies are announced that they will soon have several NFC phones. Nokia already had two phones 2005 and next year all the Nokia's smart phones will be equipped with the NFC reader. Samsung has have these phones already in the internal markets of South Korea and Apple will also do the same soon. This technology enables ticketing and paying (via the phones directly without an operator - phone will act as cash, bank card, visa card, vip card, etc. Mobile phone manufactures, banks and operators are already standardizes this technology. NFC enables us to use a phone as a physical browser (as a mouse in computers) as a service interface (access to information, control, ..) and between an user and his environment including things and items. Now the about 5 billions tags are made but it will grow up to 100 billions soon. The main applications thus far have been ticketing, access control, and logistics a general. Clothing industry does not use RFID only in the packages but also in a single cloth and it is utilized not only in the production and logistic chains but also retailers e.g., for inventory purposes. Next big step will be RFID based sensors e.g., humidity sensors and also well being and medical sensors: temperature, EKG, EEG, etc. data loggers. Finally, as my title is hinting, the big thing it that RFID will create Internet of Things - think about it.

Personal and Pervasive Sensing Systems

Prabal Dutta, Department of Electrical Engineering and Computer Science, University of Michigan, USA
Abstract
This tutorial will discuss emerging approaches to personal and pervasive sensing. As a case study, we will focus on HiJack, a system for stealing power and bandwidth from the mobile phone's audio interface. HiJack enables a new tier small and cheap phone-centric sensor peripherals that support plug-and-play operation including, for example, EKG. More broadly, we will explore how the mobile phone might become the portal for pervasive sensors that are physically-embedded and perpetually-powered, and the scaling laws that will enable this emerging computing class.

Discussion question:
How small should researcher attempt to make wireless sensor nodes given the increasingly complex energy harvesting/storage, energy/power density, radio propagation, timer stability, software development, discovery and networking issues, and integration challenges? Is a killer app needed to justify the research?

Programming and Debugging Sensor Networks

Kay Römer, Institute of Computer Engineering, University of Lübeck, Germany
Abstract
Developing bug-free application software for sensor networks is a fundamental challenge. As the programmer must not only implement the functional requirements of the application, but also has to take into account the resource constraints of typical platforms, programming is a complex multi-objective optimization problem in the mind of the developer. Sensor networks are also massively distributed systems, adding another level of complexity to software development. Finally, sensor networks are exposed to an unpredictable environment that has a strong impact on the function of a sensor network that needs to be considered.

After outlining the fundamental challenges, this talk first addresses the programming side. We present typical abstractions that hide part of the complexity from the programmer and therefore simplify the programming task. We focus in particular on so-called macroprogramming techniques that abstract from the complexity of distributed programming. While these techniques help to avoid bugs, it is common that the resulting applications fail when deployed in the field. In the second part of this talk we therefore discuss debugging techniques to find and eliminate the causes of failures in deployed sensor networks. In particular, we present techniques to increase the visibility of the system state while minimizing the consumption of resources.

Discussion question:
Identify key differences between traditional application software and sensor network applications. For each difference, discuss whether it makes programming easier or more difficult. Also discuss the implications of these differences on programming and debugging.

Reading material that may be helpful:
Luca Mottola, Gian Pietro Picco: Programming Wireless Sensor Networks: Fundamental Concepts and State of the Art. (To appear) in ACM Computing Surveys.

Jan Beutel, Kay Römer, Matthias Ringwald, Matthias Woehrle: Deployment Techniques for Wireless Sensor Networks. In: G. Ferrari (Ed.): Sensor Networks: Where Theory Meets Practice. Springer, Heidelberg, November 2009.