GeoSWIFT: Open Geospatial Sensing Services for Sensor Web Vincent Tao, Steve Liang, Arie Croitoru, Zia Moin Haider, and Chris Wang Geospatial Information and Communication Technology (GeoICT) Lab Department of Earth and Atmospheric Science, York University 4700 Keele Street, Toronto, ON, Canada M3J 1P3 E-mail: {tao, liang, arie, ziamoin, chunwang}@yorku.ca URL: www.geoict.net 1. INTRODUCTION With the presence of cheaper, miniature and smart sensors; abundant fast and ubiquitous computing devices; wireless and mobile communication networks; and autonomous and intelligent software agents, the Sensor Web has become a clear technological trend in geospatial data collection, fusion and distribution. The Sensor Web is a Web-centric, open, interconnected, intelligent and dynamic network of sensors that presents a new vision for how we deploy sensors, collect data, and fuse and distribute information. The Sensor Web is an evolving concept with many different research efforts working to define the possibilities. The Sensor Web offers full-dimensional, full-scale and full-phase sensing and monitoring of Earth at all levels: global, regional and local. The Sensor Web is a revolutionary concept toward achieving collaborative, coherent, consistent and consolidated sensor data collection, fusion and distribution. Such sensors include flood gauges, air-pollution monitors, stress gauges on bridges, mobile heart monitors, Webcams, and satellite-borne earth imaging devices. The Web is considered a "central computer" that connects enormous computing resources. The Sensor Web can similarly be thought of as a global sensor that connects all sensors or sensor databases. 2. CHARACTERISTICS OF SENSOR WEB The Sensor Web is an evolving concept with many different research efforts working to define the possibilities. Examples of pioneering work include sensor pods (Delin and Jackson 2000), Smart Dust (Warneke, Atwood et al. 2001) and integrated earth sensing (Teillet, Gauthier et al. 2002). Inspired by significant advances in sensors, communication, computing and positioning technologies, the Sensor Web is being developed by connecting heterogeneous sensors or many proprietary sensor networks. To Copyright © 2004 CRC Press, LLC 267 succeed, the Sensor Web must be interoperable, intelligent, dynamic, and scalable. Interoperable The Sensor Web is achieved by connecting the distributed, dynamic and heterogeneous in-situ and remote sensors to an open, interconnected network - the Web. The Sensor Web is a universe of network-accessible sensors, sensory data and information. Just like building a Web system, developing the Sensor Web requires thorough and careful design of system hierarchy, registry, Internet Protocol domain services and applications. Implementation of interoperability requires commonly accepted standards and specifications, and the Open GIS Consortium http://www.opengis.org has pioneered much of this activity. Intelligent Such intelligence comes from sensor connectivity, just like human intelligence comes from connected neurons in the brain. Sensor networks forage for information the way ants forage for food. By linking existing databases or previously sensed data with the Sensor Web, we will dramatically increase the efficiency and performance of sensing, monitoring and change detection. Dynamic Thanks to wireless communication and real-time positioning technologies, sensors no longer need to be fixed at a certain location. By removing physical constraints, sensors can be mobile, placed anywhere or even be seeded for large-area monitoring. As sensors are position aware, their location and movement can be tracked continuously via wireless networks. Scalable The Sensor Web is intended to offer massive sensing by interconnecting a large array of sensors. Therefore, the Sensor Web must not be inherently restricted to contain a specific number of nodes or operate within a predefined area (Sohrabi, Merrill et al. 2002). Mobile Location is an essential component of the Sensor Web. When and where sensor data is observed is of equal value to the sensor data itself. There are wide options to integrate positioning technologies (e.g., GPS, A-GPS, Internet GPS, radio-frequency identification, real-time locating system, cellular network positioning, etc.) with sensor networks. There are important Copyright © 2004 CRC Press, LLC 268 GeoSensor Networks research topics in location-based routing of sensors as well as optimized sensor network topology and configuration. 3. OPEN SENSOR WEB SERVICES There are many technical issues involved in building the Sensor Web framework. From a system architecture viewpoint, the Sensor Web involves the following layers: • Sensor layer-sensor design, materials, miniaturization, energy consumption, etc. • Communication layer-networking, protocol, topology, etc. • Information layer-agents, management, fusion, distribution, etc. Depending on the properties of sensors, geographic coverage, network access capabilities and, more importantly, domain applications, the physical architecture (i.e., the first three layers) can be very different. The information layer serves as a backbone and shares a commonality. This layer is a gateway to integrate and fuse observations from spatially referenced sensors. It connects widely distributed in-situ sensors and remote sensors over wired or wireless networks. Interoperability becomes a key to enable the information layer's integration capability. 4. WEB SERVICE ARCHITECTURES We use a new technology framework, Web Services, to design the architecture of our open geospatial sensing service for Sensor Web: GeoSWIFT. Web Services represent the convergence between the service- oriented architecture (SOA) and the Web (W3C, 2002). SOA has evolved over the last 10 years to support high performance, scalability, reliability, and availability. However, traditional SOA are tightly coupled with specific protocols. Each of the protocols is constrained by dependencies on vendor implementations, platforms, languages, or data encoding schemes that severely limit interoperability. The Web Services architecture takes the advantageous features of the SOA and combines it with the Web. The Web supports universal communication using loosely coupled connections. Web protocols are completely vendor-, platform-, and language-independent (Systinet Corp., 2002). Web Services support Web-based access, easy integration, and service reusability. Web Services satisfies our requirements to build a geospatial infrastructure for Sensor Webs with openness, interoperability, and extensibility. Web Services now are developing as standards in W3C group and will become future standards for the Web. Copyright © 2004 CRC Press, LLC GeoServNet SensorWeb 269 5. OGC WEB SERVICE OGC Web Services are an evolutionary, standards-based framework that will enable seamless integration of a variety of online geoprocessing and location services. OGC Web Services will allow distributed geoprocessing systems to communicate with each other using technologies such as XML and HTTP. This means that systems capable of working with XML and HTTP will be able to both advertise and use OGC Web Services. OGC has taken a leadership role in developing and building a unique and revolutionary open platform for exploiting sensors. OGC's goal is to make all types of sensors, instruments and imaging devices as well as repositories of sensor data discoverable, accessible and, where applicable, controllable via the Web. OGC Web Services will allow future applications to be assembled from multiple, network-enabled geoprocessing and location services. This capability will be possible because rules will be established for these services to advertise the functionality they provide and how to send service requests via open, standard methods. In this manner, OGC Web Services provide a vendor-neutral interoperable framework for web-based discovery, access, Figure 1 : GeoServNet OGC WMS 3D Viewer . Copyright © 2004 CRC Press, LLC 270 GeoSensor Networks integration, analysis, exploitation and visualization of multiple online geodata sources, sensor-derived information, and geoprocessing capabilities. As a member of OGC, GeoICT Lab is actively involved in the development and implementation of OGC specifications. Figure 1 shows a screen capture of the GeoServNet OGC WMS 3D Viewer developed at GeoICT lab for OGC CIPI (Critical Infrastructure Protection Initiative). GeoSWIFT is based on the same technology, GeoServNet, for the components of server and 2D/3D sensor viewer. 6. GEOSWIFT GeoSWIFT is an ongoing project sponsored by GEOIDE, PRECARN and CRESTech. The goal of GeoSWIFT is to build a geo-spatial infrastructure to connect distributed sensor networks for the sharing, access, exploitation, and analysis of sensing information. The development of GeoSWIFT involves both software engineering and communication development work. Various intelligent sensors will be deployed to the fields and several different real- world applications of the system, such as snow depth monitoring, crop yielding prediction, nuclear safety, meteorological services, and real-time GSN Image RDB Web Accessible Sensor Observations GeoSWIFT Sensor Service GeoServNet DEM Streaming Service GeoServNet Vector Streaming Service GeoServNet Image Streaming Service GeoServNet O GC Web Mapping Service GeoSWIFT User GeoServNet Spatial Data Services Raw Observations Sensor Networks Mother Nodes GeoServNet Database GetMap GML/ Observations / Return Map Request Vector Return Vector Request Observations Raw Position Request Data Return Data Matched Position Historical Raw Observations Raw Observations SensorML Update RDB Update RDB Image Vector DEM Attributes GSN 3D Analysis GSN 3D Fly GSN Map Sensor Nodes GeoSWIFT Tracking Service Figure 2: System Architecture o f GeoSWIFT . Copyright © 2004 CRC Press, LLC GeoServNet SensorWeb 271 tracking, will be evaluated with our industrial and government partners, such as MacDonald Dettwiler Inc., Canadian Centre of Remote Sensing (CCRS), and Canadian Nuclear Safety Commission (CNSC). GeoSWIFT is built upon GeoServNet, a GIServices system developed at GeoICT lab at York University. GIService allows users to access, assemble, or even “rent” geoprocessing components that are distributed across a network via standard web browsers (Tao 2001). Figure 2 shows the system architecture of GeoSWIFT system. The octagons in Figure 2 are service providers. GeoSWIFT Sensor Service is the portal of sensors to provide web services interfaces for sensor networks, its observations and the related geo-spatial information as well. Depending on the properties of sensors, geographic coverage, network access capabilities and, more importantly, domain applications, the physical architecture of sensor networks can be very different. GeoSWIFT Sensor Service serves as a wrapper which hides the different communication protocols, data formats and standards of sensor systems behind this layer and provides a standard interface for clients to collect and access sensor observations and manipulate them in different ways. Various clients only need to follow the OGC interfaces which are provided by GeoSWIFT Sensor Server, and do not need to deal with the annoying protocols of various sensor networks (Liang, Tao et al. 2003). Table 1 shows the four interfaces defined by OGC (OGC 2003). The hexagons in Figure 2 are viewers for Sensor Web in the architecture of GeoSWIFT. GeoSWIFT Viewer is built based on GeoServNet Viewer. GeoServNet Viewer is a 2-D/3-D Java Web GIService viewer, and designed for streaming large amount of spatial data via Internet.(Han, Tao et al. 2003) Sensor Web comprises a large number and different type of in-situ and remote sensors which will produce massive and diverse data that can challenge our ability to process and manage the data. How to accommodate various sensor types in a viewer and deliver valuable information to the end user becomes a challenging task as well. For example, a 3D viewer is needed for the visualization a series of deployed balloons of differing buoyancy to determine air movement. The multi-dimension and diversity of sensor platforms and its observations, however, offers the opportunity to take advantage of interactive 3D visualization techniques that can improve the efficiency and accuracy of processing, and provide an unprecedented perspective of sensor observations . 7. CONCLUSIONS The Sensor Web presents endless opportunities for adding a sensory dimension to the Web's globe-encircling virtual nervous system. This has Copyright © 2004 CRC Press, LLC 272 GeoSensor Networks extraordinary significance for science, environmental monitoring, transportation management, public safety, homeland security, defense, disaster management, health and many other domains of activity. The fundamental revolution in the Sensor Web vision lies in its interoperable, intelligent, dynamic, flexible, and scalable connectivity. It is an evolving framework enabled by many emerging technologies. Our understanding of the Sensor Web is in its infancy. GeoSWIFT is just a beginning, and we have already seen many exciting applications of it. Just like the Web 10 years ago, and it is only a matter of time before the Sensor Web joins the fabric of our lives and becomes indistinguishable. Table 1: Interfaces Provided by GeoSWIFT Sensor Server Requests Responses GetCapabilities The responding XML of service’s capabilities conforms OGC Service Information Model Schema (OGC 2003), provides detailed information for a client to access the service. The provided information includes Service Type, Service Instance, Content Type, and Content Instance. GetObservations The responding XML of GetObservation request is encoded conforming to GML and O&M schema. It contains values, units, and locations of the requesting sensor observations. DescribePlatform The XML response describes the sensor platform, and conforms to SensorML schema. An example of a sensor platform can be a plane tht carries a camera, several inertial sensors, and meteorological sensors. The plane is the platform. DescribeSensor The XML response contains detailed information of sensor characteristics encoded in SensorML. The sensor characteristics can include lists and definitions of observables supported by the sensor. 8. REFERENCES Delin, K.A. and S.P. Jackson, 2000. Sensor Web for In Situ Exploration of Gaseous Biosignatures. 2000 IEEE Aerospace Conference, Big Sky, MO. Copyright © 2004 CRC Press, LLC GeoServNet SensorWeb 273 Han, H., V. Tao et al., 2003. Progressive Vector Data Transmission. 6th AGILE Conference on Geographic Information Science, Lyon, France, Presses Polytechniques et Universitaires Romandes. Liang, S.H.L., V. Tao et al., 2003. The Design and Prototype of a Distributed Geospatial Infrastructure for Smart Sensor Webs. 6th AGILE Conference on Geographic Information Science, Lyon, France, Presses Polytechniques et Universitaires Romandes. OGC, 2003. OWS 1.2 Service Information Model, OpenGIS Interoperability Program Report. J. Lieberman. OGC, 2003. Sensor Collection Service. T. McCarty. Sohrabi, K., W. Merrill et al., 2002. Scaleable Self-Assembly for Ad Hoc Wireless Sensor Networks. IEEE CAS Workshop on Wireless Communications and Networking, Pasadena, CA. Tao, V. 2001. Online GIServices. Journal of Geospatial Engineering, 3(2): 135-143. Teillet, P.M., R.P. Gauthier et al., 2002. Towards Integrated Earth Sensing: Advanced Technologies for In Situ Sensing in the Context of Earth Observation. The Canadian Journal of Remote Sensing, 28(6): 713-718. Warneke, B., B. Atwood, et al., 2001. Smart Dust Mote Forerunners. Proc. 14th IEEE International Conference in Micro Electro Mechanical Systems (MEMS 2001), Interlaken, Switzerland. Copyright © 2004 CRC Press, LLC 274 GeoSensor Networks . positioning technologies (e.g., GPS, A-GPS, Internet GPS, radio-frequency identification, real-time locating system, cellular network positioning, etc.) with sensor networks. There are important Copyright. coupled connections. Web protocols are completely vendor-, platform-, and language-independent (Systinet Corp., 2002). Web Services support Web-based access, easy integration, and service reusability provide a vendor-neutral interoperable framework for web-based discovery, access, Figure 1 : GeoServNet OGC WMS 3D Viewer . Copyright © 2004 CRC Press, LLC 270 GeoSensor Networks integration,