Sensors and Systems
Breaking News
Leica Geosystems Enables a New Era for GIS Data Capture
Rating12345Heerbrugg, Switzerland  – Leica Geosystems, part of Hexagon, announced...
Unreal Engine 4.24 to Ship with Free Quixel Megascans, Unreal Studio Features, and More
Rating12345 Quixel, creator of the world’s largest photogrammetry 2D...
Commercial UAV Expo Americas 2019 Completes 5th Annual Event, Exceeds All Prior Records
Rating12345LAS VEGAS, NV– A record-breaking number of exhibitors and...

September 8th, 2008
Sensors and the Environment

image
  • Rating12345

sensor.gifWe can’t understand or manage what we can’t observe, measure and describe. To advance the human project of environmental learning and management that we began millennia ago using only our minds, our cultures and our naked senses, today we also supply ourselves with data from many kinds of sensor systems.

“It is the task of Geography . . . to present the known world as one and continuous, to describe its nature and position . . . .”  (Geographike Uphegesis, Ptolemy)

We can’t understand or manage what we can’t observe, measure and describe. To advance the human project of environmental learning and management that we began millennia ago using only our minds, our cultures and our naked senses, today we also supply ourselves with data from many kinds of sensor systems.

We have far more environmental sensors and far more data than ever before, but their value overall has, arguably, increased only arithmetically, not exponentially, because they are largely isolated and unpublished, not online in ways that make them effectively discoverable, assessable, accessible and usable by many. To deal with today’s critical environmental challenges, we need to increase knowledge exponentially. Interdisciplinary and international cooperation, integration of research communities, and scientific progress and innovation all will surely benefit from bringing sensors and their data online.

We share digital data and make digital systems connect by means of common schemas and protocols, and today these are developed mainly by standards organizations like the Open Geospatial Consortium (OGC). In 2000, the developers of the SensorML (Sensor Markup Language) specification, initially developed under a NASA Advanced Information Systems Technology (AIST) program, brought that candidate standard into the OGC. The OGC, a geospatial standards organization, was a logical organization in which to continue this project because every sensor has a location, and sensor location is always important – even more so when networked on a web. A temperature reading of 40 degrees Centigrade in July means normal heat in Dubai, but unseasonable heat in Minneapolis, Minnesota.  In addition, many sensor-interested agencies, organizations and businesses were already engaged in the OGC’s well respected standards process.

SensorML became the first focus of the OGC’s Sensor Web Enablement initiative, which now includes a number of sensor-related standards. Most of these have gone through rigorous development and review in OGC testbeds and working groups and they have now been formally adopted by the membership as OGC standards. Together with the OGC’s other geospatial standards, they comprise a revolutionary open framework for exploiting Web-connected sensors and sensor systems of all types: satellite-borne Earth imaging devices, flood gauges, air pollution monitors, stress and motion sensors, chemical/biological/radiological hazard sensors, Webcams, and countless other sensors and sensor systems. The standards also support improved management and discovery of stored sensor data and improved use of Web services for processing data.

swe_Fig1.gif
Figure 1: This figure depicts OGC SWE & IEEE 1451 standards converged in environmental applications. Diverse sensors, some in IEEE 1451 configurations, are discoverable and Web-accessible via OGC Sensor Web Enablement (SWE) interfaces, in diverse architectures and applications, with geospatial context. (Figure OGC)

OGC standards and OGC best practice that relate specifically to sensors include :

1.    OpenGIS® Sensor Model Language (SensorML) Encoding Standard – Used to define the general models and XML encodings for sensors.
2.    OpenGIS Observations & Measurements (O&M) Encoding Standard – Used to define the general models and XML encodings for sensor observations and measurements.
3.    OpenGIS Sensor Observation Service (SOS) Interface Standard – Provides an open application programming interface (API) for managing deployed sensors and retrieving sensor data and specifically “observation” data.
4.    OpenGIS Sensor Planning Service (SPS) Interface Standard – Provides an open API for a service by which a client can 1) determine the feasibility of collecting data from one or more mobile sensors/platforms and 2) submit collection requests to these sensors/platforms.
5.    Web Notification Service (WNS) (OGC Best Practice) – Provides an open API for a service by which a client may conduct asynchronous dialogues (message interchanges) with one or more other services.
6.    Sensor Alert Service (SAS) (OGC Best Practice) can be compared with an event notification system. The sensor node is the object of interest.
7.    Transducer Markup Language (TML) Encoding Standard – Used to define an application and presentation layer communication protocol for exchanging live streaming or archived data to (i.e. control data) and/or sensor data from any sensor system.

The OGC has worked closely with the other standards organizations, particularly IEEE, whose standards efforts relate to networked sensors. As a result of the OGC’s collaboration with other organizations, legacy sensor networks and sensor control systems that use IEEE 1451 can become part of Web-based sensor webs that use OGC standards.

swe_Fig2.gif
Figure 2: In a recent OGC testbed activity, live data from sensors was used to demonstrate how SWE standards can provide integration of diverse data streams to provide a complex real-time picture of the advance of a brush fire. (Figure OGC)

Benefits of Open Sensor Standards
The OGC’s SWE standards enable any kind of environmental sensor system to be deployed on the Web in a way that makes the sensors discoverable and useable through open standard interfaces. Both system developers and system users benefit.

IEEE 1451 is an interface standard for “smart sensors” and actuators. “Smart sensors” are sensors that handle their own acquisition of data and conversion of data into a calibrated result in terms of physical attribute units. The OGC-IEEE collaboration focused on finding a universal way of connecting two basic interface types – transducer interfaces (which typically mirror hardware specifications) and service interfaces (which typically mirror application requirements). Sensor interfaces and application services may need to be bridged at any of many locations in the deployment hierarchy. As a result of the collaboration, many legacy sensor networks and sensor control systems can now be integrated with other Web-based sensor networks that use OGC standards to interface to the Web and, in particular, the “geospatial Web” enabled by other OGC standards. The standards can be used together to increase the utility, reuse and mobilization of new sensors.

Developers of Web-based sensor systems can save time by working from OGC standards documents that are essentially well-vetted “cookbooks” of standard protocols and application program interfaces (APIs) for making sensors discoverable, accessible, and controllable. Working from these standards documents, developers can spend less time on requirements analysis and custom coding, and the standards ensure that their systems can be easily expanded and integrated with other current and future systems. Standards-based software components are more reusable and can easily incorporate open source tools and components. Standards-based products can reach larger markets. Developers can use any of the common service-oriented architecture (SOA) patterns (SOAP/WSDL, EBxml or REST).

The users of SWE-based environmental sensor systems can save time and exploit tremendous new capabilities. For example, through SPS and SOS, water quality sensors in a region can be automatically read at frequent intervals and those readings can be aggregated as map layers. Flood gauges and rainfall gauges can provide live map layers during storms. SWE and associated OGC standards provide for the publishing and discoverability not only of sensors, but of sensor data: Through the OpenGIS Catalogue Service (Web) Interface Standard, live and stored oceanographic data acquired from sensors mounted on multiple agencies’ buoys, ships, satellites and autonomous underwater vehicles can be published in online directories for wide use, with applications automatically aggregating data from various sources into spatial data layers for diverse purposes. When data is collected from a sensor, ISO standards-based metadata can be created automatically at the same time. The XML basis of the schemas that describe sensors and data (and that encode data), along with standard metadata tools, enable information communities or communities of practice to more easily derive and use pared-down common “application schemas” that meet community needs.

Open Data
SWE standards provide key parts of the technical infrastructure necessary for the “open data” or “open science” movement that is gaining international momentum*. The Open Science movement is supported by organizations such as the Creative Commons, the Open Data Commons , the Committee on Data for Science and Technology (CODATA) of the International Council for Science (ICSU), the Open Data Foundation, Scholarly Publishing & Academic Resources Coalition (SPARC), Center for Earth Science Information Network (CIESIN), and the Organization for Economic Cooperation and Development (OECD).

swe_Fig3.gif
Figure 3: This is a common depiction of how growth in the value of a node on a network increases with growth in the size of a network. The value of a sensor, data store, data schema or Web service similarly increases with the number of users that can use it. (Figure from http://en.wikipedia.org/wiki/Network_effect)

 

Intuitively, it seems obvious that the “network effects ” that helped sell the Bell telephone monopoly, the Internet and the Web can bring great benefits to environmental science and management. Empirical evidence is also available:  A Harvard Business School “Working Knowledge” article, (“Open Source Science: A New Model for Innovation”, a Q&A with Harvard researcher Karim R. Lakhani by author Martha Lagace, published November 20, 2006 http://hbswk.hbs.edu/item/5544.html)  reports on research that analyzes how open source norms of transparency, permeable access, and collaboration might work with scientists. Lakhani says, “Innovations happen at the intersection of disciplines. People have talked about that a lot and I think we’re providing some systematic evidence now with this study.” 

Consider a typical environmental data story: A researcher studying stickleback fish in Alaska’s lakes collects water quality data as part of a study. The researcher publishes a paper, submits a final report to the funding organization, gets paid, and puts the research data into a drawer on a DVD, invisible to others and soon forgotten.

If the data were instead kept online and if the metadata and a link to the data were registered in a catalog conforming to the OpenGIS Catalogue Service (Web) Interface Standard, the study results could be more easily verified by interested peers. Longitudinal studies could easily be performed. Other scientists in the same discipline or other disciplines could use the water quality data in their studies. The Alaska Department of Natural Resources and the Yukon River Inter-Tribal Watershed Council could easily find and use the water quality data. Software tools implemented as Web services could scan vast quantities of metadata to pull water quality data from countless servers and aggregate the data into thematic layers involving broad geographic areas and complex GIS algorithms. Such searches could become the basis for investigations and models based on vast quantities of legacy data and live sensor data, avoiding the need for, or perhaps helping to focus collaborative plans for, the collection of new data.

Fortunately, implementing these concepts is technically practical, thanks to the OGC’s SWE standards. The cost of storage and bandwidth keeps dropping. All that’s missing are institutional will and supportive legal models (see GeoRM standards below) that support sustainable — that is, profitable — business models.

Coming Into Wide Use, But Work Remains
The SWE standards are becoming widely used in many programs. The international oceans science community has been particularly organized in embracing OGC standards, with SWE standards playing an important role in the EU’s InterRisk project (Interoperable GMES  Services for Environmental Risk Management in Marine and Coastal Areas of Europe), ESA Service Support Environment (SSE) portal , SeaDataNet , Delivering Environmental Web Services (DEWS), EU Coastal Union (EUCC), the EU Marine Overlays on Topography for Annex II Valuation and Exploitation (MOTIIVE), Australian Oceans Portal, Gulf of Maine Ocean Observing System (GoMOOS), Ocean Observing System (OpenIOSS), Southeastern Universities Research Association (SURA) Coastal Ocean Observing and Prediction (SCOOP), Marine Metadata Interoperability (MMI), OOTHethsys , and the Global Earth Observing System of Systems (GEOSS).

Though the SWE framework is fully capable of addressing a wide range of requirements in communities such as the oceans community, further standards work is ongoing in a number of areas:

— Collaboration with the Open Grid Forum (OGF) will result in improved standards harmonization, tools and best practices for grid-based processing, such as complex environmental and ecological modeling, that could be used with live and stored sensor data. The OpenGIS Web Processing Service (WPS) Interface Standard plays an important role in connecting geospatial data to grid processes and models.

— Environmental activities associated with the built environment will benefit from the new OpenGIS® CityGML Encoding Standard (http://www.opengeospatial.org/standards/citygml and www.citygmlwiki.org ), an open data model framework and XML-based encoding standard for the storage and exchange of virtual 3D urban models. CityGML figures importantly in the Architecture, Engineering, Construction, Owner, Operator (AECOO-1) testbed that has been organized by the OGC with participation and sponsorship by a number of AECOO stakeholder organizations, agencies and businesses.

— The standards that govern how geospatial digital rights management will operate in a Web services environment are working their way forward in the OGC’s Geospatial Rights Management (GeoRM) Working Group (http://www.opengeospatial.org/projects/groups/geormwg). This Working Group plays a pioneering role in an evolutionary process that has advanced from the 19th century copyright model through the current popular Creative Commons model and that is now moving to a more flexible “rights management” regime. GeoRM standards must address a wide range of inescapable real world requirements that are not addressed by simple permissive/restrictive, open/closed approaches. The standards must make it possible to provide data — including environmental data — under different terms for different circumstances.

— Concerted effort is needed to apply metadata standards, frameworks and associated applications to processes involving information semantics. Interfaces applicable to knowledge retrieval and query are needed in addition to the current OGC interfaces that address information retrieval.  The OGC’s Geosemantics Working Group (http://www.opengeospatial.org/projects/groups/semantics ) is playing a role in this as they elaborate the concepts of the Semantic Web (http://www.w3.org/2001/sw ) and of spatial data infrastructures into the Geosemantic Web (http://colab.cim3.net/file/work/Expedition_Workshop/2006-02-21_GeoSpatialandDRMCoPs/JLieberman02212006.ppt ).

Conclusions
The standards infrastructure necessary to make the Web work for the geosciences is in place. Enhanced now with the SWE set of OGC standards for sensor webs, this infrastructure enables the publishing, discovery, assessment, access, control and use of environmental sensors of any kind, as well as stored data they have produced and Web services for processing sensor data. Beginning with the OGC membership’s 2006 adoption of several SWE standards, uptake among environmental researchers has accelerated. As different communities of researchers and practitioners continue to use this infrastructure, it will expand and improve through the ongoing OGC standards process.

If the last century was a century of scientific specialists focusing reductively and ever more exclusively on narrow topics, this new century is a century of studying relationships in systems — ecosystems, social systems, complex adaptive systems, and the environmental and social “externalities” of economic activities. Going forward, reductive analysis will surely be balanced by holistic analysis of “the known world as one and continuous,” as Ptolemy put it. When this kind of science involves data that has geospatial importance, it can now proceed more effectively through Web-based sharing of sensors, data and services.

* Evidence of the Open Science trend:
— NSF: In March 2008, an NSF discussion of the key issues appeared at http://www.cendi.gov/minutes/pa_0308.html . See “Impacts and Changes on the Horizon for Scientific Communications and STI”, Dr. Clifford Lynch, Director, Coalition for Networked Information (CNI).
— Nature Magazine: On the Science Commons website (http://sciencecommons.org/weblog/ ) scroll down to a 10 minute July 14th, 2008 video talk by Timo Hannay, a publisher at Nature: “Timo Hannay on CC-based science publishing”. Towards the end Hannay promotes publishing of research data. (Science Commons (http://sciencecommons.org/ ) is a project of Creative Commons, a non-profit initiative that promotes policy and technology that remove unnecessary legal and technical barriers to scientific collaboration and innovation.)
— Science Magazine: “The publishing journal, Science, convened a committee to review editorial procedures. The committee recommended that more extensive information be included in the published supporting material and asserted that primary data are essential and should therefore be made available to reviewers and readers (http://www.sciencemag.org/cgi/content/full/314/5804/1353/DC1 ). … The practice of open data-sharing isn’t as unlikely as one might think.
— Recent agreements for data-sharing in genetic science allowed the development of the Human Genome Project…”  March, 2008 (www.openmedicine.ca/article/viewArticle/205/104 ). Note that the human body and the Earth are similar in that each has a single geometry framework on which many “layers” of information are superimposed.

 

Sam Bacharach is executive director for Outreach and Community Adoption at the Open Geospatial Consortium, Inc.; e-mail: sbacharach at opengeospatial.org

Leave a Reply

Your email address will not be published. Required fields are marked *