Source-linked AI summary
SmartSantander: IoT Experimentation over a Smart City Testbed
Luis Sanchez, Luis Muñoz, Jose Antonio Galache, Pablo Sotres, Juan R. Santana, Veronica Gutierrez, Rajiv Ramdhany, Alex Gluhak, Srdjan Krco, Evangelos Theodoridis, Dennis Pfisterer
TL;DR
IoT research needs open, realistic, and scalable experimentation beyond closed or narrowly tested deployments. This paper presents SmartSantander’s city-scale architecture, deployment, and support mechanisms, combining experimentation with smart-city services. The resulting facility is designed to support large-scale evaluation of IoT concepts under real-world conditions, while some nodes impose operational limits on experimentation.
Problem
Existing IoT deployments are closed and vertically integrated, while tested solutions have limited applicability across contexts, motivating an open real-world experimentation facility.
Method
The paper develops and deploys a multi-tier SmartSantander IoT testbed with management, experimentation, and service-support mechanisms for a heterogeneous urban infrastructure.
Results
SmartSantander provides a city-scale experimentation environment that supports experiment execution, smart-city service applications, and evaluation under realistic operational conditions.
Takeaways & Limitations
The facility integrates scale, heterogeneity, mobility, realism, and end-user involvement into a holistic environment for IoT experimentation and service evaluation.
Takeaways & Limitations
Some IoT nodes impose experimentation limits because their deployment or operational constraints restrict the functions that can be performed on them.
Abstract
from arXiv · showhide
This paper describes the deployment and experimentation architecture of the Internet of Things experimentation facility being deployed at Santander city. The facility is implemented within the SmartSantander project, one of the projects of the Future Internet Research and Experimentation initiative of the European Commission and represents a unique in the world city-scale experimental research facility. Additionally, this facility supports typical applications and services of a smart city. Tangible results are expected to influence the definition and specification of Future Internet architecture design from viewpoints of Internet of Things and Internet of Services. The facility comprises a large number of Internet of Things devices deployed in several urban scenarios which will be federated into a single testbed. In this paper the deployment being carried out at the main location, namely Santander city, is described. Besides presenting the current deployment, in this article the main insights in terms of the architectural design of a large-scale IoT testbed are presented as well. Furthermore, solutions adopted for implementation of the different components addressing the required testbed functionalities are also sketched out. The IoT experimentation facility described in this paper is conceived to provide a suitable platform for large scale experimentation and evaluation of IoT concepts under real-life conditions.
1. INTRODUCTION
SmartSantander addresses the limited realism, openness, and cross-domain applicability of existing IoT deployments by creating a large-scale urban experimentation facility. The paper presents its reference architecture, deployment experience, support mechanisms, and experimentation capabilities.
- Existing IoT deployments are mainly closed, vertically integrated solutions whose optimizations are constrained to tested application contexts.
- Interoperable IoT requires an open architectural reference model and enabling services that support cross-domain resource reuse.
- SmartSantander creates a European smart-city facility for large-scale IoT experimentation and evaluation under real-world operational conditions.
- The paper describes an open reference model, city-scale heterogeneous deployment, experimenter-support solutions, and implemented experimentation capabilities.
- The paper is organized around related work, architecture and requirements, deployment, testbed management, and supported experimentation solutions.
2. RELATED WORK
Existing IoT testbeds provide partial support for realistic experimentation but commonly lack the combined scale, heterogeneity, mobility, and user involvement required for urban IoT research. SmartSantander extends these capabilities within a city deployment.
- Many existing testbeds are lab-based and therefore have shortcomings in evaluating IoT solutions under realistic operational conditions.
- Oulu Smart City and CitySense do not adequately provide heterogeneity, realistic mobility, or sufficient scale for large experiments and user trials.
- KanseiGeni, SensLab, and iLab.t offer heterogeneity, but their deployment environments and management mechanisms differ from urban outdoor settings.
- SmartSantander builds on WISEBED while adding wireless reprogramming, improved resource selection, robustness, and reduced management configuration overhead.
- Its testbed embeds mobile nodes in buses and public service vehicles and can involve real citizens in experimentation.
3. IOT TESTBED REQUIREMENTS AND ARCHITECTURE
SmartSantander defines a multi-tier, multi-plane architecture for realistic, scalable, heterogeneous IoT experimentation while concurrently supporting testbed management and smart-city service provision. The design addresses mobility, end-user involvement, reliability, and operational tractability.
- Requirements: The requirements include realism, adequate scale, heterogeneity, realistic mobility, end-user involvement, and reliable service operation.
- Scale: SmartSantander offers access to thousands of IoT experimentation nodes for advanced large-scale scenarios.
- Mobility: The facility deploys infrastructure on buses, public service vehicles, and taxis to support realistic mobility patterns.
- User support and end user involvement: The facility targets researchers, commercial service developers, and citizens, enabling assessment of technical performance, service usability, and user acceptance.
- Architecture: Its architecture consists of IoT device, gateway, and server tiers, with cloud virtualization supporting reliable and available server components and services.
- Architecture: The reference model separates testbed observation and management from IoT experimentation while allowing logical functions to coexist on physical devices.
- Architecture: The platform supports both experiment execution and smart-city service applications, reflecting an experimentation/service-provisioning duality.
4. SANTANDER TESTBED DEPLOYMENT
The Santander deployment serves both as a scientific testbed and as an environment for evaluating the socioeconomic acceptance, usability, and performance of IoT services with end users involved.
- The deployed infrastructure concurrently supports scientific experimentation and assessment of IoT service acceptance, usability, and performance with end users.
4.1 USE CASES AND SCENARIOS
The Santander testbed supports smart-city scenarios using distributed IoT sensing, including environmental monitoring, parking management, precision irrigation, augmented reality, and participatory services.
- Environmental Monitoring: Environmental monitoring combines fixed and mobile sensors to assess pollution through air quality, noise, and luminosity measurements.Large numbers of lower-cost sensors provide finer-grained measurements than conventional urban monitoring stations.
- Outdoor Parking Management: Outdoor parking management uses buried ferromagnetic sensors to monitor bay occupancy and disseminate parking information to displays and mobile applications.Historical occupancy data can also support municipal analysis of parking provisioning.
- Precision Irrigation: Precision irrigation evaluates plant water requirements to replace timetable-based watering with more precise on-demand irrigation.The use case targets parks and gardens whose existing systems operate without considering weather or local vegetation needs.
- Augmented Reality: Augmented reality uses NFC tags at points of interest, shops, and public spaces to expose context-sensitive information and services.Collected location and visitor-frequency data can inform site-management strategies and citizen feedback services.
- Participatory Sensing: Participatory sensing uses smartphones to provide physical measurements and user observations while distributing subscribed city-event alerts.Users can submit text, images, and video alongside sensed data.
4.2 DEPLOYED INFRASTRUCTURE
The Santander deployment combines fixed, mobile, and tagged IoT resources in a city-scale architecture designed for realistic experimentation, while addressing power, connectivity, and management requirements.
- Deployment Strategy: The deployment proceeds cyclically, beginning with fixed wireless-sensor-network locations and expanding toward broader experimentation and smart-city services.The first cycle focused on a meshed WSN across three geographically significant areas.
- Deployment Constraints: A hybrid power solution uses lamppost electricity and rechargeable batteries, charging at night while supporting daytime node operation.Electrical adaptations and protections were added to comply with municipal regulations.
- Deployment Constraints: Buried parking nodes cannot access permanent power, so energy-efficient mechanisms provide device lifetimes exceeding 3 years while restricting experimentation to car-presence information.This constraint follows from the nodes’ installation beneath asphalt.
- Fixed Infrastructure: 740 points of presence include more than 50 noise, 600 temperature, 500 light-intensity, and 30 carbon-monoxide sensors, plus 390 parking nodes and 23 gateways.The gateways connect the IoT node tier with the server tier.
- Expanded Infrastructure: The second cycle added approximately 50 fixed nodes, 150 mobile devices on public and municipal vehicles, and 2,000 QR/NFC tags across the city.The added resources support irrigation, mobility-pattern experimentation, environmental monitoring, and alternative IoT technologies.
- Connectivity and Gateways: The topology groups sensors around gateways, using repeaters when nodes fall outside gateway radio range and embedded PCs for routing, experimentation, and management.Gateways require both constant power and Internet connectivity, supplied through municipal premises, fibre, or 3G connectivity.
- Hardware and Planes: Nodes provide separate wireless channels for experimentation and management or service functions, avoiding the impracticality of a wired backhaul across geographically distributed sensors.Lamppost devices use two XBee-Pro radio modules operating at 2.4 GHz, with distinct IEEE 802.15.4 configurations.
5. LARGE-SCALE IOT TESTBED MANAGEMENT
SmartSantander manages a large-scale IoT testbed through automated resource discovery, monitoring, and reconfiguration. Its multi-tier components communicate asynchronously through event channels to support dependable operation and resource selection.
- Dynamic management: Automated management covers resource discovery, monitoring, and testbed reconfiguration across changing node states and failures.Human intervention is retained for decision-making and policy specification while routine event response and remediation are automated.
- Resource discovery: Resource discovery registers new IoT resources and stores standardized descriptions so users can select devices matching their criteria.The Resource Directory supports storage and lookup through a REST interface, including criteria such as sensed phenomena and locality.
- Management components: The management architecture distributes responsibilities across portal-server, gateway, and IoT-node tiers.Components include the Resource Directory, IoTResourceManager, configurators, participatory-sensing managers, NodeManagers, and node APIs.
- Gateway and node functions: NodeManager and the Node API detect, register, and monitor nodes by exchanging service frames, commands, and device-property queries.NodeManager can identify new or dead nodes and maintain gateway status through periodic beacon messages.
- Event-based coordination: Management components communicate through dedicated event channels for registration, monitoring, and reconfiguration.The distributed event bus uses asynchronous, distributed, multi-party bindings and request-reply topics to support remote IoT management.
- Resource discovery: The Resource Directory is a central SmartSantander building block because it stores searchable resource descriptions for device selection.Its lookup function supports selecting resources according to user-defined criteria.
6. TESTBED USAGE: IOT EXPERIMENTATION
SmartSantander supports both service-level and IoT-device experimentation in a city-scale facility, combining real-world smart city services with mechanisms for selecting, reserving, monitoring, and reprogramming resources. A precision irrigation trial illustrates how sensor data supported remote operational decisions.
- Experimentation approaches: SmartSantander supports heterogeneous experimentation, including service-level applications and scientific experiments requiring direct IoT-device control.Service experiments generally access gathered information, whereas scientific experiments may reprogram nodes with experimental binaries.
- Service-level experimentation: City-scale deployment enables assessment of both IoT technologies and smart city services through interaction with real end-users.The platform is used in ongoing trials involving municipality technicians, citizens, and other users.
- Precision irrigation: 48 IoT nodes equipped with agricultural sensors were deployed for the precision irrigation use case.Measurements included air and soil conditions, atmospheric pressure, solar radiation, wind, and rainfall.
- Precision irrigation: The precision irrigation service used real-time field information to help park technicians adjust irrigation strategies across deployment subareas.Technicians reported that the irrigation status was accurate enough to support watering decisions, while heatmaps and reports enabled remote assessment.
- Precision irrigation: During February 2013, a one-week dry period raised soil moisture tension to 28 centibars, but managers did not start irrigation.The service supported this decision and avoided in-field inspection visits during the month.
- Experiment lifecycle: The scientific experiment lifecycle includes resource specification, setup, execution, and corresponding mechanisms for managing experiments on the testbed.Resource discovery uses static and dynamic properties, topology, semantic descriptions, reservation, and interference-aware allocation.
- Resource management: Resource management is designed to maintain suitable conditions for simultaneous experiments and services, with the Reservation System API reserving devices by properties such as type, sensors, and mobility.The API identifies resources using URNs and supports selection of subsets of IoT devices.
- IoT node control: Reprogrammability is treated as critical, and SmartSantander implements reliable multihop over-the-air programming for repeated unicast, multicast, or broadcast flashing.The mechanism was implemented because available literature mechanisms lacked implementations for the deployed devices.
7. CONCLUSIONS
The paper presents SmartSantander’s architecture, deployment experiences, and management mechanisms for realistic, large-scale IoT experimentation. It reports a holistic city facility intended to address limitations of existing testbeds while supporting further federation with other sites and experimentation facilities.
- Conclusions: SmartSantander supports testing IoT protocols, services, and configurations in a realistic setting at an appropriate scale.The facility addresses shortcomings in existing testbeds concerning scale, heterogeneity, mobility, realism, and end-user involvement.
- Conclusions: The paper presents the testbed architecture, deployment issues, experiences, and mechanisms needed to keep track of resources and support dependable facility management.The management solutions address the scale and variety of events in a large testbed.
- Conclusions: The facility is planned for federation with additional sites and other Future Internet experimentation testbeds, providing access to more numerous and varied IoT devices.The federation is aligned with existing GENI and FIRE activities.