Geospatial Sensor Web Resource Management System ... - IEEE Xplore

4 downloads 221 Views 1MB Size Report
Geospatial Sensor Web Resource Management. System for Smart City: Design and Implementation. Jia Li, Nengcheng Chen*. State Key Lab for Information ...
2014 14th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

Geospatial Sensor Web Resource Management System for Smart City: Design and Implementation Jia Li, Nengcheng Chen* State Key Lab for Information Engineering in Surveying, Mapping and Remote Sensing Wuhan University 129 Luoyu Road, Wuhan Hubei, China, 430079 [email protected], *[email protected]

other natural or social problems. The “Smart City” concept then came up.

Abstract—It is a needful and challenging work to integratedly manage the massive, distributed and heterogeneous sensor resources for a cooperative observation task in smart city context, especially for the cooperative task of dynamic environment monitoring, timely emergency response and citizen safety guarantee. Regarding the heterogeneous sensor resources that immersed in the cities, there lacks an open and comprehensive platform for the management of this sensing infrastructure of the smart cities. In this paper, a Geospatial Sensor Web Resource Management System (Geo-SWRMS) is designed and implemented to tackle this issue. With the requirement investigation of the resources management of the Geospatial Sensor Web for the smart city, four main functions including resource modeling and registry, resource query and discovery, resource visualization and resource access and control are considered and implemented leveraging the standard Sensor Web Enablement information model and service interface of Open Geospatial Consortium. The proposed system is flexible, accommodating heterogeneous sensor resources and can be utilized in multiple application fields in the smart cities such as sensing resource management, cooperative observation task for emergency and so on. The Geo-SWRMS promotes the sharing and interoperability of the sensor resources, and facilitates to provide a solid information base for observing task scheduling and decision making in the smart city application.

There are many definitions of the “Smart City”[3]-[7]. The “Smart City” concept was derived from the “Smart Planet” firstly proposed by IBM[8]. It is referred to the application of advanced information technologies to all works of our city life, embedding sensors and actuators to city public buildings, private houses, traffic routes, underground pipelines, air environment, water system and other circumstance in every corner of the city, and forming the “Internet of Things” via the Internet, which forms the strongest interact layer like a cyberskin of the city. Then through utilization of super computer, cloud computing and other processing and analyzing method, people can deal with a city’s problems and manage production and life in a more meticulous and dynamic way, ultimately achieving the state of smart. A smart city cannot be achieved without the smart sensing infrastructure. They function as the data sources of upper analyzing procedure and is critical to decision making of many city problems. Geospatial Sensor Web can function as important sensing infrastructure of the smart city in the fields such as facilities management, environment and traffic monitoring, public safety guarantee, emergency response and so on. This sensing infrastructure for the smart cities includes tens of thousands of in-situ sensors around the city such as air quality sensors, traffic sensors, video monitors as well as large scale remote sensing satellite sensors for natural and human disasters response or general planning and managing of the city. Under the Geospatial Sensor Web environment, these spaceair-ground platform based earth observing sensors possess multiple standard, multi-type, multi-scale characteristics, heterogeneous and vast in quantity, which hinder the realization of sharing management of these sensor resource via the web for the broader range of applications in the smart city. Moreover, the widely access and control of these heterogeneous sensing resources of the web is considered a development trend and huge challenge. There is a strong need for a comprehensive tool to integratedly manage these sensors utilizing a unified sensor information sharing model and a standard service interfaces regardless of the diversity and heterogeneity of these resources to form, thus helping promote an open and scalable Geospatial Sensor Web for the smart city application .

Keywords—Geospatial Sensor Web; resource management; OGC; visualization; cooperative observation; sensor modeling; smart city

I.

INTRODUCTION

With the change of the earth's environment, the demand for earth observation data is becoming more and more complicated, and the earth observation systems appear to be more and more various. The Geospatial Sensor Web concept is formed, which means a coordinated observation infrastructure composed of distributed resources. These resources can behave as a single, autonomous, task-able, reconfigurable observing system that performs earth observation (EO) and provides observed and derived data along with the associated metadata by using a set of standards-based service oriented interfaces [1]-[2]. With the process of accelerating urban development in the globe, the cities is facing more and more new problems such as traffic congestion, energy shortage, air pollution, water environment deterioration, urban surface subsidence and many 978-1-4799-2784-5/14 $31.00 © 2014 IEEE DOI 10.1109/CCGrid.2014.70

819

the city sensors, ranging from the meteorological sensors, air quality sensors to traffic sensors. They are very different in observing phenomenon and observing mechanism, but generally gives a simple observed value. These sensors generally can be easily accessed and integrated managed. A special important in-situ sensor is the video camera which can be seem around the cities. It is complicated in observing mechanism and large in data volume. More will be considered when managing them.

OGC Sensor Web Enablement [9] Group has made a general framework for sensor resource sharing and interoperability, but there is no definition of unified shared metadata set of sensors or a general method manage these abundant sensor resources information. For the discovery and management of sensor resources on the web, OGC has developed the catalogue service for Web (CSW)[10] and then published the ebRIM( ebRIM registry model)[11] as the core information model for publish and discovery of the geo-spatial web service and spatial data, but there is still a lack of directly registry and an efficient discovery mechanism for vast heterogeneous sensors. That is standardized and unified management of decentralized sensor resources is not yet achieved.

2. Remote sensing sensor: Remote sensing plays an important role in the Geospatial Sensor Web for smart city applications such as weather forecast, city environment monitoring, city planning and managing, emergency response and so on. The remote sensing sensors also have several different types (camera, scanner, SAR, etc.)with different usage. They can be cooperated with each other or the in-situ ground sensor for complex observation tasks for the smart city application. The function of remote sensing sensor, especially the high spatial and temporal resolution sensors, largely promote the geographic information base for the smart city.

Although there are some application projects or systems based on the OGC Sensor Web Enablement framework using SOA architecture, such as the sensor web project of 52north [12], GeoSWIFT [13], PULSENet [14], NASA Sensor Web 2.0 [15], Sensor web agent platform (SWAP) [16], which all are based on a Sensor Web Enablement standard service interface to realize the sensor resource planning and access service, they are all using their own limited heterogeneous sensor resources sharing information model and may partly implemented the standard OGC service interfaces, thus resulting in a lack of an overall efficient management for these resources from describing the sensor resources, connecting to the sensors, discovering the sensors, visualizing the sensors to even controlling the sensor resources for particular task.

B. System Requirement for Management of Sensor Resources The goal of the Geospatial Sensor Web Resource Management System is to provide a platform for modeling (describing), discovering, visualizing, accessing and controlling the vast and heterogeneous sensor resources in the Geospatial Sensor Web for smart city. To achieve this goal, the system has to meet four main functional requirements: Sensor modeling, Sensor discovery, Sensor visualization and sensor access and control.

This paper elaborates on the development of Geospatial Sensor Web Resource Management System(Geo-SWRMS) and its aim to support the management of resources sharing and interoperability, both for the sensors and the sensed data for the smart city application. The development of the system follows the software development lifecycle phases used in Rapid Application Development [17], including: requirement identification and analysis (Section 2), system design and implementation (Section 3). In Section 4, we summarize the outstanding features of the system. Finally, we draw our conclusion in Section 5.

II.

Fig. 1 is a map showing the system functional requirements and the sub-module to be considered to satisfy those requirements. 1. Sensor modeling covers the construction of unified sensor description information model for heterogeneous sensor

SYSTEM REQUIREMENT

A. Sensor Resources Types for Smart Cities For the development of the smart cities, vast and various sensors play an important role in obtaining the latest sensed data about our cities, both Indoor and outdoor, from underground up into the air. This sensing infrastructure provides great source of live data for smart city applications such as environment monitoring, public safety guarantee, emergency response and so on. Because of its diversity, two main types of sensors are classified as the typical sensor resources to managed in our Geospatial Sensor Web for the smart city application.

Fig. 1. System requirement map.

resources, checking its validity and registering it to the registry center. Types of sensors include in-situ sensors (e.g. meteorological sensors, hydrological sensors, GPS, RFID etc.), remote sensing sensors (e.g. aerial camera, scanner, radar, etc.). This function mainly unified represent the heterogeneous sensor resources for widely integration and sharing of them .This requirement is the information foundation of the system.

1. Ground in-situ sensor: The in-situ sensor means the sensor that is directly deployed at the observing location and in contact with what they are sensing. They account for most of

820

Fig. 2. Data flow between system components.

4. Make the distributed sensors fast available to the users across the web regardless its heterogeneous physical interfaces and provide necessary control operations for observation scheduling. This sensor access and control requirement ensures the system’s operability and functionality.

2. Sensor discovery means the query among the wellmodeled sensors based on the sensor information models above, which involves basic fields query (e.g. sensor name, type, application, contacts etc.) and combination query (e.g. combined fields query, spatial and temporal query of satellite image sensors etc.). This will be of great help for efficiently discovering and utilizing the sensor resources. It can provide a target sensor resource collections for the users ( decision maker).

III.

SYSTEM DESIGN AND IMPLEMENTATION

This section outlines the design and implementation of the Geospatial Sensor Web Resource Management System(GeoSWRMS). As show in the Fig. 2, the system mainly includes three parts: the core management platform (Geo-SWRMS-P) of the Geo-SWRMS, the Registry Center and the SOS Center. The core management platform covers the identified four main functional modules, namely Resource Modeling and Registry, Resource Query and Discovery, Resource Visualization and

3. Sensor visualization considers the visualizing of both insitu and remote sensing sensors. It includes geolocation of the in-situ sensors, the visualization of the real-time satellite orbit, the covering area of the sensors, the basic information of the sensors and so on. This visualization requirement will greatly enhance explicitness and usability of the sensor resources.

Fig. 3. System overall architecture of Geo-SWRMS.

821

Fig. 4. Main user interface of Geo-SWRMS.

Resource Access & Control. The Registry Center is responsible for the repository of the sensor resources as well as provides query interface for the models. The SOS Center is set up to receive and store the live data and ready to response the requests for the data.

A. Sensor Modeling and Registry In order to represent heterogeneous sensor resources, a SensorML [18] based sharable and interoperable information model [19] is utilized for the sensor modeling process. Different modeling wizards (corresponding to the mentioned Sensor Type Tree) encapsulating this model are developed to give a friendly user interface for fulfill these models. Fig. 5 gives a glance at one of the modeling wizard. The overall sensor information is divided into four part, namely the identification information, characteristics and capabilities, service information and input and output information, and through filling the sensor information elements on the four taps a whole model is formed.

The Fig. 3 shows the system overall architecture of GeoSWRMS. There are main four layers for the Geo-SWRMS: 1. Data Layer: mainly contains sensor models have been established, the SOS data, 3-D model data, satellite orbit data (SGP4 TLE) as well as basic geographic data, etc.; 2. Components Layer: mainly contains SensorModel component, SensorRegistry component, SensorAccess component, SensorControl component, OrbitTools component for calculating the dynamic position of the satellite, GeaeExplore component for 3-D display and related Windows controls; 3. Business Layer: mainly contains the basic business functions such as sensor modeling, registry, query, visualization, access and control; 4. Presentation Layer: the satellite sensors’ dynamic position and in-situ sensors’ fixed position and their basic information mainly visualized in the browser window or the 3D visualization sphere. The system main interface is shown in Fig. 4 . The upper area is the menu bar with the above proposed functions, middle-left is Sensor Type Tree through which a type-based wizard will instruct the users to fast model the target sensor. The middle-right part is the browser and edit area, in which model instance can be browsed and edited. The button is the status bar showing the status of the system.

Fig. 5. Modeling wizard of a video camera sensor

The following subsections describe each of these parts with an emphasis on the Geo-SWRMS platform.

822

Fig. 6. Interface of batch registry.

to facilitate diversified query demand. As shown in Fig. 7, they are: basic information criteria (Unique Id, name, sensor type, etc.), characteristics and capabilities criteria (observing range, observing resolution, revisit time, etc.) and spatial-temporal criteria. sensor resource query (valid time, covering boundary, etc.).

For the sensor registry, a direct sensor registry service method [20] is implemented. The model can be registered after saved through the modeling wizard or using a batch registry interface shown in Fig. 6 B. Sensor Query and Discovery Sensor Query is a very important function of the GeoSWRMS. The abundant information encoded in the resource models greatly promotes the discovery of required sensor resources. Three different facets of query criteria are designed

For the remote sensing satellite sensors, when check the spatial-temporal criteria, an orbit and covering algorithm runs to give a precise answer. Moreover, these criteria can be combined to figure out the required available sensor resources

Fig. 7. Query interface for the remote sensing satellite sensors.

823

Fig. 8. (a)The visualization of satellite orbit and sensor coverage

(b)The visualization of in-situ sensor and its basic information.

regardless of their physical locations and interfaces, a Sensor Observation Service approach [21] is used in implementation of the Geo-SWRMS. For the in-situ sensors, as soon as it can be connected to the Internet, a SOS-based access can be achieved by the workflow of modeling the sensor, registering it to the center and inserting the data to the SOS center using a matched SOS toolkit. The present SOS toolkit can match more than 10 types of in-situ sensors.

collection. C. Sensor Visualization For each sensor resource, there is a static or dynamic position attached to it, which can be geolocated on the map. For the static in-situ sensor, they are represented as points owning basic sensor information while for the movable aerial and satellite sensors, a path or an orbit is given to show their positions. Especially for the satellite sensors, their orbit are calculated based on the latest orbit data as well as their dynamic coverage information.

Fig. 9 shows the latest observation value and the time period observations from this SOS approach. As for the satellite sensor, we use a method of harvesting data from existing open data centers by means of SOS. Fig. 10 shows the interface of satellite sensor access and its results.

Fig. 8-(a) shows the visualization of satellite orbit and sensor coverage and Fig. 8-(b) illustrates an in-situ sensor situation.

The access and control of video camera is focused in the implementation of the Geo-SWRMS. Fig. 11 shows the interface of access and control of video camera sensor. Live video streaming is acquired, with screen captures and video clips at a certain interval inserting to the SOS center. Many

D. Sensor Access and Control To obtain the live data sensed by the sensor resources

Fig. 9. Latest observation and time period observations through SOS access of a insitu soil moisture sensor.

824

Fig. 9. Satellite sensor access and its results.

unified sensor information model is used for describing the heterogeneous sensors. The physical characteristics and observation capabilities are all unified model for more precisely discovery and deeper analysis. In this way, almost all sensors can be modelled and shared through Geo-SWRMS in Geospatial Sensor Web.

control operations are implemented by the SPS[22] approach including simple zoom, rotate to complex path planning control. IV.

FEATURES OF GEO-SWRMS

Compared to the state-of-the-art of the existing sensor resource management system, We summarize the outstanding features of the system as follows:

B. A Direct Sensors Registry Service Method Through a direct sensors registry service method implemented in the Geo-SWRMS, a user can store sensor information directly with more complete sensor information

A. A Unified Sensor Information Model To implement the sensor modeling, a SensorML-expressed

Fig. 10. Interface of access and control of video camera sensor.

825

Based on our work thus far, we have identified the following research challenges that must be met to advance the Geo-SWRMS: 1. Deriving information from the sensor information model for high-level decision making, such as observing capability evaluation. 2. A more adaptive sensor access and control module (not flexible for all sensors presently).

and higher efficiency for discovery compared to existing registry systems such as the 52N-SIR registry[20]. The users can register personalized sensor information and enable a efficient sensor publish and discovery. C. Spatial-Temporal Combined Dynamic Query A orbit and covering calculation algorithm has been implemented for dynamically calculate the position and covering area of the satellite sensor based on the orbit parameters and observing capability recorded in the sensor model instance. This is of great help for remote sensing image data searching and scheduling, especially for complex cooperative task among the remote sensing sensors or rather including the in-situ sensors.

ACKNOWLEDGMENT This work has been supported in part by National High Technology Research and Development Program of China under Grant 2013AA01A608, by the National Basic Research Program of China under Grant 2011CB707101, by the National Natural Science Foundation of China under Grant 41171315, and by the program for New Century Excellent Talents in University under Grant NCET-11-0394.

D. SOS-based Sensor Access and SPS-based Sensor Control Based on OGC SOS and SPS service interfaces, a sensor access and control module is developed in the Geo-SWRMS. The simple in-situ sensors can be accessed easily and some can even execute the web user’s control commands. Through these standard OGC service interfaces, the sharing of distributed live observation data is achieved and this enable the Geo-SWRMS open and interoperable characteristic. V.

REFERENCES [1]

CONCLUSION

The management of vast and heterogeneous sensor resources in the smart cities is of great importance and there lacks a comprehensive and open platform to . Geospatial Sensor Web can function as important sensing infrastructure of the smart city.

[2]

[3]

In this paper, we have proposed a Geospatial Sensor Web Resource Management System(Geo-SWRMS) for the smart city application from requirement analysis to system design and implementation to tackle the issue of a lack of a unified sensor information model and versatile management platform for describing, connecting to , discovering, visualizing and controlling the vast and heterogeneous sensor resources using the OGC standard information models and service interfaces under the Geospatial Sensor Web Environment in the smart city context. The proposed Geo-SWRMS provides a wizard modeling and registry tool to model heterogeneous sensors and directly register it for repository or updating. Through the multiple query interface, precise dynamic discovery is achieved considering the spatial-temporal criteria. The live data can be accessed based on the Sensor Observation Service and the remote sensing sensors can be dynamic simulated for its coverage capability and some in-situ sensors can be remotely controlled and planned for specific task using an SPS approach. All these help promote an open and scalable Geospatial Sensor Web for the smart city application .

[4] [5]

[6]

[7]

[8] [9]

[10] [11] [12]

In summary, Geo-SWRMS greatly promotes the sharing and interoperability of the resources of Geospatial Sensor Web. Considering its outstanding features, Geo-SWRMS has great potential for comprehensive application in many fields in the smart city such as sensor management for environment and traffic monitoring, public safety guarantee, emergency response and so on.

[13]

[14]

826

K. Delin and S. Jackson, “The Sensor Web: a new instrument concept,” Proceedings of the SPIE International of Optical Engineering, vol. 4284, pp. 1–9, 2001. L. Di, “Geospatial Sensor Web and Self-adaptive Earth Predictive Systems (SEPS),” ESTO/AIST Sensor Web PI Meeting 2007, San Diego, February, 2007. Palmisano, Samuel J. "A smarter planet: the next leadership agenda." IBM. November 6 (2008). Glaeser, E. L., & Berry, C. R. (2006). Why are smart places getting smarter? Taubman Center Policy Briefs, PB-2006-2. C. Harrison , B. Eckman , R. Hamilton , P. Hartswick , J. Kalagnanam , J. Paraszczak , P. Williams, “Foundations for smarter cities”, IBM Journal of Research and Development, v.54 n.4, p.350-365, July 2010 Giffinger, R., Fertner, C., Kramar, H., Kalasek, R., Pichler-Milanovi, N., & Meijers, E. (2007). “Smart Cities: Ranking of European MediumSized Cities”. Vienna, Austria: Centre of Regional Science (SRF), Vienna University of Technology. Washburn, D., Sindhu, U., Balaouras, S., Dines, R. A., Hayes, N. M., & Nelson, L. E. (2010). Helping CIOs Understand "Smart City" Initiatives: Defining the Smart City, Its Drivers, and the Role of the CIO. Cambridge, MA: Forrester Research, Inc. Harrison, Colin, et al. "Foundations for smarter cities." IBM Journal of Research and Development 54.4 (2010): 1-16. M. Botts, G. Percivall, C. Reed, et al. "OGC® sensor web enablement: Overview and high level architecture," GeoSensor networks. Springer Berlin Heidelberg, pp. 175–190, 2008. Nebert, Douglas, Arliss Whiteside, and P. Vretanos. "Opengis catalogue services specification." Implementation Specification (2007). Lesage, N. "OGC® Cataloguing of ISO Metadata (CIM) Using the ebRIM profile of CS-W." OGC document 07-038 (2007). A. Bröring, J. Echterhoff, S. Jirka, I. Simonis, C. Everding, Stasch, S. Liang and R. Lemmens, “New Generation Sensor Web Enablement,” Sensors, vol. 11(3), pp. 2652–2699, 2011. S. Liang, V. Tao and A. Croitoru, “Sensor Web and GeoSWIFT An Open Geospatial Sensing Service,” In:International Society for Photogrammetry and Remote Sensing XXth Congress–GeoImagery Bridging Continents, pp. 12–23, 2004. S. Fairgrieve, J. Makuch, and S. Falke, “PULSENet: An implementation of Sensor Web standards,” International Symposium on Collaborative Technologies and Systems, Baltimore ,pp. 64-75, 2009.

[21] A. Na and P. Mark, “Sensor observation service,” Implementation Standard OGC, 2007. [22] I. Simonis and P. Dibner, “OpenGIS® Sensor planning service implementation specification (Version 1.0),” Open Geospatial Consortium (OGC) document number (2007): 186, 2007.

[15] D. Mandl, P. Cappelaere, S. Frye, R. Sohlberg, Ong L., S. Chien, D. Tran, A. Davies, S. Falke, S. Kolitz, P. Zhao, L. Di, N. Chen, G. Yu, D. Smithbauer, S. Ungar, L. Derezinski and M. Botts, “Sensor Web 2.0: Connecting Earth’s Sensors via the Internet,” NASA Earth Science Technology Conference, Adelphi, 2008. [16] D. Moodley and I. Simonis, “A New Architecture for the Sensor Web: The SWAP Framework,” Lecture Notes in Computer Science. 5th International Semantic Web Conference (ISWC 2006), Athens, GA, USA, pp. 4273, 2006. [17] N. Ruparelia, “Software development lifecycle models,” ACM SIGSOFT Software Engineering Notes, vol. 35(3), pp. 8-13, 2010. [18] M. Botts and A. Robin, “OpenGIS sensor model language (SensorML) implementation specification,” OpenGIS Implementation Specification OGC, 07-000, 2007. [19] N. Chen and C. Hu, “A sharable and interoperable model for atmospheric satellite sensors and observations,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 5(5), pp. 1519-1530, 2012. [20] N. Chen, X. Wang and X. Yang, “A direct registry service method for sensors and algorithms based on the process model,” Computers & Geosciences, vol. 56, pp. 45–55, July, 2013.

827