2016 IEEE 8th International Conference on Intelligent Systems Smart Learning Environment for the Development of Smart City Applications Manuella Kadar Department of Computer Science 1 Decembrie 1918 University Alba Iulia, Romania mkadar@uab.ro; manuellakadar@yahoo.com and the benefits of learning out of the classroom in combination with smart classroom activities carried out in a smart environment. The paper is structured as follows: section 2 presents the key features of smart learning environments and ubiquitous learning, section 3 presents the concept and architecture of the SCL computing system, followed by the functionalities of SCL computing system, section 4 presents learning scenarios and means of learning evaluation, while section 5 points out on conclusions and further work. Abstract—The paper presents the results of the “Smart People for Smart Cities!” project funded by European Structural Funds Human Resources. “1 Decembrie 1918” University of Alba Iulia, Romania in partnership with UTI SA Romania, UNINOVA Portugal and INFOR ELEA Italy have collaborated to bridge and enhance people’s skills and competences for the development of smart city applications within the university curricula of Applied Electronics, Computer Science and Environmental Engineering specialties. The primary goal was to create an enhanced learning and teaching environment, goal that was achieved through specific objectives: (i.) to set up a research testing laboratory where real conditions of a smart city is simulated; (ii.) to develop and implement a software instrument for students’ evaluation; (iii.) to develop and implement a software solution for processing data provided by various types of sensors used in the smart city environment. The project successfully complemented the academic curricula with a module of practical applications, which simulate existing automation projects on the market, integrates data obtained from sensors and provides statistical models. Smart City Lab (SCL) computing system covers all stages of data processing in automated systems II. LEARNING Smart learning (s-learning) means a new learning paradigm which serves learners to have an efficient learning environment that offers personalized mobile contents and easy adaption to current education model. And it also allows learners to have a convenient communication environment and rich resources [1], [2]. Table 1 shows some learning options with specific features and tools, namely: e-learning, m-learning, u-learning, and slearning. Electronic (e-learning) is defined as the use of computer network technology, primarily over an intranet or through the Internet, to deliver information and instruction to individuals [3]. Mobile learning (m-learning) includes the ability to learn everywhere at every time without permanent physical connection to cable networks. This can be achieved by the use of mobile and portable devices such as PDA, cell phones, portable computers, and Tablet PC [4]. Smart learning (s-learning) is an important and new paradigm of learning today. The concept of s-learning plays an important role in the creation of an efficient learning environment that offers personalized contents and easy adaptation to current education model. It also provides learners with a convenient communication environment and rich resources [1], [2]. Ubiquitous (u-learning) environment is a situation or setting of pervasive or omnipresent education or learning. Education is happening all around the student, but the student may not even be conscious of the learning process. Source data is present in the embedded objects and students do not have to do anything in order to learn. They just have to be there [5]. Keywords—Urban computing, ubiqitous learning, smart city computing system, smart environment I. INTRODUCTION Modern smart cities offer the necessary infrastructure and services to foster a number of formal and informal learning activities in the city landscape. This paper presents our proposal on how urban computing system may leverage the learning process and improve the university curricula for the specialties of Applied Electronics, Computer Science and Environmental Engineering. We have developed the Smart City Lab (SCL) computing system together with various educational scenarios and activities to be performed by exploiting SCL. In this paper we present the concept and architecture of the SCL, the methodology we followed for designing learning scenarios and the experience we gained. Learning is a cognitive process of acquiring skills or knowledge. In our approach both students and teachers have experimented novel technological means and methods in the process of exploration of the effects of urban computing on the learning process. We applied our methodology in the city of Alba Iulia. Results from our experience demonstrate the potential of exploiting urban computing in the learning process 978-1-5090-1353-1/16/$31.00 ©2016 SMART LEARNING ENVIRONMENT AND UBIQUITOUS 59 This approach is supported also by [8] stating that a ubiquitous learning environment provides an interoperable, pervasive, and seamless learning architecture to connect, integrate, and share three major dimensions of learning resources: learning collaborators, learning contents, and learning services. Therefore ubiquitous learning is characterized by providing intuitive ways for identifying right learning collaborators, right learning contents and right learning services in the right place at the right time [8]. Table 1. Learning Systems Learning system Features Tools E-learning Computer-based training Intranet, Internet M-learning Flexibility of location Mobile and portable devices S-Learning Personalised content, efficient and effective education Smart mobile devices, PDA, PC U-Learning Pervasive education Mobile and PC devices, sensors, camera III. SMART CITY LAB CONCEPT AND ARCHITECTURE The Smart City Lab (SCL) computing system aims at simulating within the laboratories of "1 December 1918" University Alba Iulia situations and events based on specific sensors that identify smart city features. By definition, cities in which all vital systems (transport, parking, water, sewer, lighting, waste management) are "smart" or "connected", are called "smart cities". For the scope of this paper we consider the definition of a smart city as in [9]: “…are territories with a high capacity for learning and innovation, which is built-in to the creativity of their population, their institutions of knowledge creation and their digital infrastructure for communication”. …. [and are concerned] with people and the human capital side of the equation, rather than blindly believing that IT itself can automatically transform and improve cities.” From the technological perspective, smart cities have been defined as cities with great presence of ICT technologies. These have permeated commercial applications of intelligentacting products and services, artificial intelligence, and thinking machines to enter and leverage the city’s infrastructure. Smart homes and smart buildings are example of systems equipped with a multitude of mobile terminals and embedded devices as well as connected sensors and actuators [10]. In this context a smart city becomes the extension of a smart space to the entire city scale. The SCL computing system is based on specific hardware such as: a large variety of sensors, data concentrators, gateway routers, database server, etc. The hardware covers four areas of interest in a smart city: (i.) smart management of waste, (ii.) air pollution, (iii.) noise pollution and (IV.) intelligent management of street lighting. The software has been designed as modules that interface with hardware and backoffice modules to interpret the collected data, and to display and interface with additional applications such as student evaluation and counselling applications, student physical and emotional status evaluation applications. Data collected from sensors can be analyzed, processed and presented as reports and graphics. Further on, event workflows and alarm transmissions can be defined. As a lab solution, it allows transfer of data to additional applications developed by students. The application is a secured solution based on data collection from sensors arranged in simulated environments of a smart city. The information read by sensors are recorded with a variable frequency that can be defined by the user and then can be stored in the data warehouse. Based on this information, graphs/reports can be generated, issuing of alarms and tracking of workflows for predetermined events can be also achieved. The main characteristics of context‐aware and ubiquitous learning are given in [6] and discussed under the following eight aspects: mobility, location awareness, interoperability, seamlessness, situation awareness, social awareness, adaptability, and pervasiveness. More detailed descriptions of these aspects are given as follows: • Mobility: The continuousness of computing while learners move from one position to another. • Location awareness: The identification of learners’ locations. • Interoperability: The interoperable operation between different standards of learning resources, services, and platforms. • Seamlessness: The provision of everlasting service sessions under any connection with any device. • Situation awareness: The detection of learners’ various situated scenarios, and the knowledge of what learners are doing with whom at what time and where. • Social awareness: The awareness of learners’ social relationship, including what do they know? What are they doing at a moment? What are their knowledge competence and social familiarity? • Adaptability: The adjustability of learning materials and services depending on learners’ accessibility, preferences, and need at a moment. • Pervasiveness: The provision of intuitive and transparent way of accessing learning materials and services, predicting what learners need before their explicit expressions. In addition to the context‐aware ubiquitous learning characteristics presented in [6], the potential criteria of a context-aware ubiquitous learning environment are formulated as in [7]: • Offers more adaptive supports to the learners by taking into account their learning behaviors and contexts in both the cyber world and the real world. • Actively provide personalized supports or hints to the learners in the right way, in the right place, and at the right time, based on the personal and environmental contexts in the real world, as well as the profile and learning portfolio of the learner. • Adapt the subject content to meet the functions of various mobile devices. 60 The main goal of such systems is to record immediately and to set-up parameters of interest for: • Real-time monitoring of living conditions; • Increasing efficiency and labor productivity; • Cost reduction; • Minimize response time in extreme situations The proposed solution is based on 4 sensors kits: intelligent street lighting kit, air quality testing kit, smart waste management kit and noise pollution monitoring kit. Each kit contains one data concentrator that collects data from sensors and send it to the Meshlium gateway using ZigBee communications protocol. This communicates with the 4 concentrators, aggregates data and send them to the server. The server runs Portal Services and Web Services. The functional modules of the application (back-office) are implemented on each user terminal and are connected to the server, providing real time access to the stored data as in Fig. 1. Fig. 2. Software architecture of SCL Data collections from sensors and storing into a database is done using a communication service. This service is bidirectional due to the fact that collects data from sensors and also sends commands to the sensor kits. Back-office is modular with components such as: - User administration module - Components/sensors management module - Communication settings module - Reports/graphics module - Events workflow module - Data export module A. User Administration Module This functionality allows information management of users and rights: • Users - operators who have access to the application. It manages the following information: Name, Code, Password, Email • Groups - allows creating groups that contain one or more users. • Rights - allows grant / revoke rights to various functionalities of the application. Fig.1. Hardware architecture of SCL The functionalities of the SCL computing system are: • Unique identification of sensors and unique identification of data type for each sensor; • Data collections from sensors and storing into a database • Users and rights management; • Handling data from the database to issue alarms in case of escaping from a range of preset parameters; • Defining alarms based on data collected from sensors; • Defining of event workflows; • Data exports into csv, xls formats B. Components/Sensors Management Module The application can uniquely identify each sensor node and each hardware architecture node. In hardware architecture, the nodes are represented by data concentrators and gateways. It defines the following entities: Sensor - at hardware level, each sensor is identified through a unique address. Starting from this unique address, other attributes can be defined for each sensor: sensor ID, name, type, measured parameter, measurement unit, status, etc. Node - is represented by the data concentrator or gateway. The attributes that can be defined for each node are: ID, name, type (concentrator or gateway), status, parent, etc. The system strictly identifies sensors in order to keep the accuracy of retrieved data. Nodes (concentrators) are identified through type of data transfers that are transmitted by the central system. Software applications ensure the implementation of business requirements through the sensor communication software and the back - office module (Fig. 2.). C. Communication Settings Module The application is able to control and modify the communication parameters between different components. Different components of the system and also the 61 eight educational affordances, namely: knowledge construction, applying, synthesis, evaluation, interactivity, collaborative learning, game based learning, and context aware learning. SCL is designed as ubiquitous learning that provides context aware information and self learning opportunities for learners. In SCL learning is not viewed only as a form of delivered instruction, undertaken within the confines of traditional educational environments. It is understood as a social process that happens at a time and place of the learner's choosing. It is collaborative, evolving and informed by a process of self paced development. It combines ‘aware’ classroom, capable of understanding about the context of its students; it is also a digitally enhanced outdoor space as a cityscape. SCL builds a real time interactive classroom with tele education experience by bringing pervasive computing technologies. The used approach is to move the user interface of a real time tele education system from the desktop into the 3D space of an augmented classroom so that in this classroom the teacher could interact with the remote students with multiple natural modalities just like interacting with the local students. Following we present a number of representative learning scenarios exploiting the smart city environment. Fig. 3. Presents the interface for the city of Alba Iulia that implements several Smart city applications. One application scenario is the Smart Factory. This application is focused on reducing maintenance costs and ensuring quality in the manufacturing process of porcelain. communication status is identified. For instance, one can select: - ZigBee communication settings between data concentrators and gateways; - Ethernet / Wi-Fi / 3G communication settings between gateway and database server; - Views communication status between different components. D. Reports / Graphics Module The application is able to issue specific reports for both parameters collected from sensors and communication area and the status of each sensor / node. It will also offer features for generate the following graphics: • Integration of collected parameters as time function; • Uptime communication; • Uptime sensors / nodes. E. Events Workflow Module Workflows module for event implementation is designed to define certain actions that the system must perform if a sum of conditions are met. In this case, there are defined workflow settings for normal operation and also tasks to be executed in the case of certain events. The module is very important, if the application is interfaced with additional systems to communicate and transmit tasks to third parties. Interface defining workflow site is graphical and intuitive. F. Data Export Module Data export module represents the module in which collected data can be exported in various formats such as: xls, csv, and pdf formats. This module is important for both data visualization, and data collection and transfer to additional applications, unless Web Services connectors are not used. G. Additional Application Interfaces using Web Services Given the open feature of the application, it is imperative to offer a mechanism to native interface with additional applications. This is designed by using external connectors of web service type. In this way the application can transmit information collected by additional applications and also receive information. Using this module in combination with the workflow event module, one can elaborate logic diagrams that perform certain actions using certain inputs, with bidirectional information running thought these connectors. An example is the connection with the Ambient Intelligence Environment for Curriculum Development [11] that integrates modules for student’s assessment, environment assessment and personalized and adapted teaching. IV. Fig. 3. SCL interface for Alba Iulia city A. Case study 1 The objective of this case study is to monitor critical processes, environmental variables throughout a porcelain factory, parameters that affect product quality and working conditions. It monitors air temperature around working robots and conveyors, light intensity in various working areas and CO2 concentration in the workers' area and in real time, using a PC, tablet, or smartphone and an Internet connection. The integration of sensor technology with Cloud platform allows data from sensor nodes to transmit directly to the Cloud, for analysis and use it in a number of applications [12]. EDUCATIONAL SCENARIO AND EVALUATION The educational scenarios focused on the educational affordances of the context aware ubiquitous learning environment. The methodology established types of educational affordances that can be provided by a context aware ubiquitous learning environment. SCL can provide 62 Battery, communications module: 802.15.4, 220V charger/adapter. The unit houses a battery in its enclosure to allow the system to measure and transmit data even without a connected power supply. The battery capacity is 6600mAh: this means that application autonomy can range from several days to several weeks without recharging the battery, depending on sampling frequency. The unit nodes can operate under battery power or on the grid to continuously recharge the battery [12]. A multiprotocol gateway and can be configured with different communication protocols such as Wi-Fi, Ethernet and 802.15.4, enabling the system to receive sensor data from the nodes, parse it and store it in a local database. The use of 802.15.4 communication technology allows sensor data to be stored or read as it is measured, so that users can visualize data “live” in real time. By introducing an IP address in a browser all the Smart Factory information can be visualized on the Web interface. Software is pre-configured – data is accessible anywhere, anytime. Each unit node is programmed and tested in laboratories so that, once in position, it can start monitoring when the “ON” switch is pushed. It monitors corresponding parameters at different intervals because not all processes have the same priority or critical level. After each measurement, data is transmitted to the gateway using 802.15.4. Other information such as battery level is sent to provide robustness to the solution. Using the cloud connection allows the unit to start sending data immediately without the need to configure a own server system. From the outset, the sensor data is available for download and local analysis, or it can be transferred to another server. Once the data is synchronized with the cloud platform, it can be accessed from anywhere. Fig. 4. The sensor solution in four key areas B. Sensors in the porcelain factory In the factory, the porcelain production process is performed throughout several phases mainly executed by robots. The final phase of the control is achieved manually by workers. At this point in the process, it is crucial to control the defects on the surface and structure, the quality of the color. Temperature control is required at a high sampling rate and high rate of accuracy in the readings. Because of all the above, costs of maintenance of these automated machines is high. Ensuring a right temperature control not only reduces the rejection of products but lowers the maintenance costs. Humidity sensors are necessary because humidity can alter the behavior of the object during the glazing phase and affect the final quality. Light sensors are needed to maintain a constant luminosity, and are calibrated for color analysis, since color looks different depending on ambient light. Sensors that can capture Volatile Organic Compound (VOC) readings are very important. In the process of producing the porcelain ware different types of solvents are used that volatilize during the drying process. Environmental rules state that evaporated VOCs from the solvents must be captured, recycled or destroyed in a controlled manner, to keep them from polluting the atmosphere. Sensors measure VOCs and ensure compliance. Finally, Noise sensors monitor working conditions in the factory. The hardware configuration consists of four types of sensor devices and a multiprotocol Internet gateway. The unit nodes used in this case study are modular and can monitor and measure a wide range of factors: up to seven initial parameters such as: temperature, CO2, VOC, humidity, luminosity (Luxes accuracy), microphone (dBA). Each unit node contains the appropriate sensors and electronics needed to control them, and are formatted from the factory to adapt to each specific use. The encapsulated nodes also contain: C. Evaluation and Lessons Learned In our evaluation we exploited both quantitative and qualitative data. Quantitative data from questionnaires that students filled in and qualitative data from observations, user trials and interviews. Our aim was to answer research issues in three different areas. The first issue was relevant with the usability of the SCL computing system. The second issue was to study whether people can actually learn by interacting with the application. Finally we studied the benefit of using an urban computing system like SCL in the learning process. Aiming to evaluate the usability of our system we asked 16 teachers to fill in a questionnaire after using the application. All participants answered that it was easy to use SCL, except two persons who were neutral. The overall rating of SCL performance (Fig. 5) was good as expected, as one participant rated it neutral. In order to assess the benefit of using an urban computing system like SCL in the learning process we interviewed both students and teachers. Students expressed their excitement about being able to use smart phones in learning activities and taking a class outside their classroom in combination with applications in the smart classroom. It was evident that students were eager to participate in our study and go out interacting with SCL in the smart factory application as well in the smart classroom. Teachers, also, expressed positive remarks; they found interesting to use novel technologies 63 teaching specific subjects such as Sensors, Measurement, Data Processing and were encouraged that their students were passionate on carrying out outdoors learning activities. which students can become totally immersed in the learning process. In extension, smart city learning leverages the infrastructure and services that modern smart cities offer in order to transform them into open learning environments. Further work will integrate new case studies and new web service applications will be developed by students. ACKNOWLEDGMENT The author acknowledge the European Social Fund and the Romanian Government for support and funding and the partners of the research project POSDRU/156/1.2/G/137166 entitled “Smart People for Smart Cities! – Adapting the study programs for Applied Electronics, Computer Science and Environmental Engineering to the requirements of the 21st century”. Fig. 5. Evaluation of SCL performance REFERENCES Our experience has shown the benefits of learning in a smart city environment exploiting an urban computing system. Such an environment provides effective and meaningful learning experiences by expanding learning beyond the walls of the classroom, as well as by allowing interaction in the city setting and bringing new interactions back into the classroom. We observed that students were urged to take an active role and work collaboratively. Embracing mobile and ubiquitous learning teachers and students communicate, arrange their activities and change their habits and practices freeing themselves from the confines of the desktop activities. As younger generations are “born” familiar with those systems there is an increasing benefit in adopting such technologies, as they facilitate students to continue their learning process outside classrooms, when and where they desire, through exploration and interaction. Therefore, an increasing interest in physical spaces, in the role of technology in city-wide environments and in passage of learning activities outside the classroom is revealed. V. [1] S. Kim, S.-M. Song, and Y.-I. Yoon, “Smart learning services based on smart cloud computing,” Sensors, vol. 11, no. 8, pp. 7835–7850, 2011. [2] T. Kim, J. Y. Cho, and B. G. Lee, “Evolution to smart learning in public education: a case study of Korean public education,” Open and Social Technologies for Networked Learning, vol. 395, pp. 170–178, 2013. [3] E. T.Welsh, C. R.Wangerg, K. G. Brown, and M. J. Simmering, “E-learning: emerging uses, empirical results and future directions,” International Journal of Training and Development, vol. 7, no. 4, pp. 245–258, 2003. [4] T. Georgiev, E. Georgieva, and A. Smrikarov, “M-learning-a new stage of @-learning,” in Proceedings of the International Conference on Computer Systems and Technologies, pp. IV.28-1– IV.28-5, 2004. [5] V. Jones and J. H. Jo, “Ubiquitous learning environment: an adaptive teaching system using ubiquitous technology,” in Proceedings of the 21st Australasian Society for Computers in Learning in Tertiary Education Conference (ASCILITE ’04), 2004. [6] S.J.H. Yang, T. Okamoto, S.S. Tseng, ”Context Aware and Ubiquitous Learning” (Guest Editorial), Educational Technology & Society, 11 (2), pp. 1 2, 2008. [7] G.J. Hwang, C.C. Tsai, S.J.H. Yang, “Criteria, Strategies and Research Issues of Context Aware Ubiquitous Learning”. Educational Technology & Society, 11 (2), pp 81 91, 2008. [8] S.J.H Yang, “Context Aware Ubiquitous Learning Environments for Peer to Peer Collaborative Learning”, Educational Technology & Society, 9 (2), pp. 188 201, 2006. [9] R.G. Hollands, “Will the real smart city please stand up?”, City: analysis of urban trends, culture, theory, policy, action, Vol. 12, No. 3, pp. 303320, 2008. [10] C. Klein, G. Kaefer “From smart homes to smart cities: Opportunities and challenges from an industrial perspective” In Proceedings of the 8th International Conference, NEW2AN and 1st Russian Conference on Smart Spaces, SMART 2008 (St. Petersburg, Russia, Sep 3-5), 2008. [11] M. Kadar, M. Muntean, L. Marina, “Developing a personalized and adapted curriculum for engineering education through an ambient intelligence environment”, Engineering Education: Curriculum, pedagogy and didactic aspects, CP, Elsevier, pp .25-63, 2014. [12] http://www.libelium.com/products/waspmote/ accessed May 2016. CONCLUSIONS AND FURTHER WORK In this work we have investigated how smart city technologies may enhance the learning process of students hearing courses of Applied Electronics, Computer Science and Environmental Engineering. We have developed the Smart City Lab computing system with various educational scenarios and activities were performed exploiting SCL. Results from our experience demonstrated the potential of exploiting urban computing in the learning process and the benefits of learning out of the classroom. This evolution and especially mobile and smart city teaching enables educators to get their classes out of their classrooms, facilitates learners to continue their learning in the city landscape and allows realization of the learning process when and where they desire. Teaching methods like projects and field trips can benefit from these technologies as they can enrich the smart city learning experience. Ubiquitous learning on the other hand means something more than mobility, a ubiquitous learning environment is any setting in 64