TTX5 Demonstrations and Posters

SPINE is a software framework for the design of Body Sensor Network (BSN) applications. SPINE enables dynamic configuration and control of signal processing and sensing functions on TinyOS-based sensor nodes. The framework allows for efficient implementation of real-time signal processing applications via an extensible function library that can be activated dynamically. In addition, SPINE manages node communication to increase reliability. SPINE can greatly decrease development time of BSN applications, increase application interoperability and allow sensor network hardware to be easily repurposed for a variety of applications. We present a demo of a healthcare monitoring application which performs real-time motion classification by activating local processing functions on the nodes in real-time.

Wireless sensor monitoring system application, solutions, H/W platform and software based on TinyOS technology. We provide WSN motes, various application and solutions using this mote that is Telos platform. Application, solution and platform are watch type sensor node for healthcare, WSN node applied NAND Flash memory, network security and air quality monitoring system and so on. And many WSN projects in Korea will be introduced via poster and brochure.

Delivering sensor readings from beyond the confines of a building's envelope to the internet using intranets and other available communication channels has very specific engineering challenges. The driving force for sensors beyond a building's envelope is the need for low cost, reliable, real-time data, for businesses that are solving real world complex problems. The user experience, equipment management, and data management for business users is as much a requirement as the reliable collection of the data. Ease of wireless systems installation, commissioning and low overhead costs are engineering solutions that need to be built into the product early to be attractive to customers that want the data and not the headaches associated with wireless telemetry. The Azonde series of wireless instruments is a multi-parameter stage gauge - measuring water height and temperature. It is designed to reduce total life cycle costs of ownership. Powering options include a) an integrated solar panel/power storage and a unique urban antenna design, b) standard solar power & battery input ranges c) any combination of the previous two plus low cost 'D' cell batteries. Wireless ranges depend on the customer's site and geographical constraints-theoretical ranges in free space for a) 2.5Ghz are 1km/0.6miles/3300ft b) 900Mhz with low cost antenna configurations are 10Kms/6miles c) wherever there is cellular access. The Azonde can be configured for other types of data collection with the SDI-12 interface.

Wireless links display a wide range of variability in packet reception rate over both time and space. The rate of variability, or churn, of the links are fundamental to the overall behavior of collection, dissemination, and routing in low-power wireless networks. In our work, we explore effects of link churn on the upper layers of communication and try to quantify the effects on network topologies, neighbor selection, and overhead in communication.

Sensor centric mote platforms are important for the wireless sensor network service application. This demo. includes small form factor Kmote, Umote and Kmote2 which are general purpose mote and various sensor boards as well. Kmote and Umote are logically same with UCB Telos rev. B platform with very small hardware form factor and support TinyOS 1.x and 2.x. In order to speed up the market of wireless sensor network, it is necessary to have well-defined mote platform and various sensor boards for commercial application. Kmote and Umote are easy to interface with PIR motion sensor board, door magnetic sensor board, electric power consumption sensor board, CMOS camera board. On the gateway side, ARM9 based gateway (KWSG, USG-100) with Kmote and Umote interface supports various network interface simultaneously, such as Power Line Communication, RS-285, GPRS, CDMA, etc.

Sensor Look easily delivers wireless sensor network service to customers. It based on the efforts to reduce computing resource in local area and move it to Internet area. In the Sensor Look, user can handle the wireless sensors like small Internet devices with light-weight gateway device and application software at IDC. The Sensor Look system is consist of wireless sensor devices with TinyOS, light-weight gateway with Linux, and IDC server application. The target of Sensor Look system is providing low cost wireless sensor network application with service framework based on IDC servers.

In this demonstration, we show the capabilities of a IPv6/6LoWPAN running on typical wireless sensor nodes. In addition to IP, nodes communicate using familiar protocols, such as ICMPv6 at the network layer, UDP and TCP at the transport layer and Telnet and HTTP at the application layer. By communicating with standard protocols, it is now possible for nodes to communicate directly with other more traditional computing devices across a variety of links and even over the Internet. Standard protocols also allow easier integration into existing networks, supporting familiar networking concepts no different than any other IP device. Familiar APIs based on BSD sockets lower the barrier to network programming in wireless sensor networks. Supporting the IP stack is a responsive, low-power link-layer and robust routing protocol. The IPv6/6LoWPAN implementation is built entirely in nesC using TinyOS.

Released versions of the TOSSIM simulator include a noise-modeling algorithm but assume that reception power is constant. Experiments, however, show that this assumption is not accurate. In this work we explore the application of concurrent pattern matching (CPM) to model power variations. CPM was first introduced by [1] for the modeling of short-term noise patterns. As in [1], experimental traces form the basis of the simulated trace. It becomes a nontrivial problem to fill in the value of the power if some packets are not received in the experimental trace. We explore a possible algorithm that fills in the power trace and performs the simulation using CPM. To check the validity of this approach, we also collected data from several experiments and compared them to the corresponding TOSSIM simulations. [1] H. Lee, A. Cerpa, and P. Levis. Improving wireless simulation through noise modeling. In IPSN 07: Proceedings of the 6th international conference on Information processing in sensor networks, pages 21?30, New York, NY, USA, 2007. ACM Press.

NXTMOTE
Rasmus Ulslev Pedersen
TinyOS is a small operating for small (wireless) sensors and LEGO MINDSTORMS NXT is a platform for embedded systems experimentation: The combination of NXT and TinyOS is NXTMOTE. We demonstrate nxtmote to nxtmote communication with continuous sensing and sending/receiving. The embedded ARM7 is fully programmable via the LEGO open source files and the freely available hardware description and schematics. This project is available in tinyos-2.x-contrib and downloadable from Sourceforge under the project name nxtmote.