Bugra Aydin, T. Tolga Sari, Sema F. Oktug
©SHUTTERSTOCK.COM/GOGOFVECTOR
Advancing information and communication technologies (ICT) enable the implementation of better solutions that increase the efficiency and quality of the systems in almost every aspect of life, including the urban environment. ICT enables the creation of smart environments by utilizing intelligent parking systems, advanced waste management, and energy-efficient infrastructure to enhance the quality of life of its citizens. At the same time, such enhancements cause less harm to the environment and increase energy efficiency. Unfortunately, the large geographies and complex structures of cities pose challenges in developing new ICT projects.
On the other hand, campuses provide favorable conditions for developing such projects, thanks to their relatively controllable geography and similarity to cities. For this reason, researchers develop smart applications to enhance campus residents’ lives or consequently offer new solutions for smart cities. Compared to cities, the limited scale of campuses allows for the implementation of numerous smart campus applications.
Smart campus applications can be classified into six main categories: smart people, smart living, smart mobility, smart environment, smart economy, and smart governance. Smart people applications generally aim to improve learning/teaching activities and contribute to the design of the campus based on its residents’ expectations. Online learning applications and campus-specific social platforms are examples of these. Furthermore, smart canteens and libraries, surveillance applications, and recommendation apps contribute to Smart living. Such projects enhance the routine services available on campus. The construction infrastructure of campus and building information management systems could be considered as a part of smart living.
Smart Mobility apps that include all aspects of campus transportation include smart bus stops, intelligent parking spaces, shuttle monitoring applications, bicycle sharing projects, electric vehicle charging, etc. Smart environment projects, such as smart solar microgrids, air quality monitoring, and waste management systems, aim to reduce or monitor environmental side effects. Moreover, they aim to use natural resources in a sustainable way. Smart economy is the dimension that is related to the campus budget; investments; scholarships for students; project grants; and, moreover, concepts like innovation and entrepreneurship. The final domain of the smart campus projects is smart governance. In short, data platforms, open data projects, and standardization studies facilitate the management of a smart campus.
When developing an application related to one of these dimensions, we could employ various technologies like big data, artificial intelligence (AI), augmented reality (AR), blockchain, computer vision, wireless and/or broadband communication technologies, the Internet of Things, autonomous systems, robotics, cloud computing, cybersecurity, geographic information systems (GISs), etc. as shown in Fig. 1. The outer ring shows some smart campus applications in the figure.
Fig 1 The main categories of smart campus applications.
This article first presents general information about sensors and data collection, communication technologies, data platforms, and the use of AI in smart campuses. Then, the functions and features of the test platform that was built in this context to promote the development of smart campus capstone projects in the Department of Computer Engineering at Istanbul Technical University (ITU) is explained.
Sensors play a vital role in data collection. They enable the fetching of informative measurements that can contribute to increasing the quality of life on campuses. Air quality, temperature, soil moisture, position, magnetic, and vibration sensors are commonly used to gather information. Air quality sensors can measure ratios of hazardous gases, such as CO2 and SO2, in the atmosphere. Moreover, some sensors collect the density in the air of particle matter that is smaller than a certain value. These tiny particles can damage the lungs, and, in fact, some of the particles can enter the bloodstream and cause severe harm to our bodies. Gathering these values from the air assists residents and administrative units in deciding on the location of leisure areas or new dormitories. Likewise, soil moisture sensors can provide crucial information about irrigation schedules, which is very valuable to maintain the campus flora. These sensors help with the development of various applications serving various smart campus dimensions.
As these sensors are, for the most, readily available and inexpensive, developers unreservedly use them. However, some types of sensors, like lidar and those with high precision, can be costly and rare.
On campuses, data are not only collected via sensors. The number of students enrolled in a course or the images of students entering a specific campus building could be used to develop smart campus applications by following ethical rules and obtaining necessary permissions.
After collecting sensor data, they should be transferred to a common database. Due to the higher cost of wired options and ease of wireless environments, smart campus applications mainly utilize wireless communication solutions to transmit data. Various wireless communication technologies can be used in these projects. When selecting the transmission technology, we must consider the following criteria: energy requirement, range, and data rate.
Zigbee radio-frequency modulation is regarded as the most appropriate technology due to its long battery life and affordability for applications that need short-range and low transmission rates, like digital attendance checking. Bluetooth and Wi-Fi (IEEE 802.11) provide better communication solutions if an application needs high data transmission in short ranges. Indoor video transmission requiring applications can be given as an example. Based on the campus size, long-range transmission may be necessary. When an application needs to transfer data over long ranges, and low energy expenditure is the primary requirement, it should use low-power wide-area technologies, such as LoRa, SIGFOX, INGENU, TELENSA, or QOWISIO. On the other hand, applications requiring long range and high throughput can use GSM technologies to transmit data.
Energy consumption is another important criterion for selecting communication infrastructure. Generally, the larger the range or data rate of the communication technology, the more energy is needed to transfer data. Technologies like Wi-Fi and GSM (with high energy demands) are not feasible for applications requiring low energy consumption.
In an ideal smart city or campus, there are a massive number of sensors that regularly measure and send data. As most applications are battery powered and do not require frequent data transmission, regular implies every few hours or once a day in some cases. Due to the amount of data generated by these assorted types of sensors, smart campuses are prime examples of the domain of “big data.” Managing big data is challenging for smart city or campus administrators, so employing data platforms is preferred. Data platforms effectively analyze, visualize, use, and make future predictions based on data sources under a single software framework. Moreover, these data platforms simplify the installed sensors’ observation, supervision, and maintenance. Managers can either develop a unique data platform or use ready-made platforms, such as DeviceHive, Cumulocity, and SensorCloud. Generally, these platforms support multiple communication technologies and visualization options to satisfy the requirements of different types of sensor applications.
The main purpose of collecting data in a smart campus is to make inferences based on the information gathered. AI-based solutions are employed in some applications since they are more effective than traditional approaches. For example, face recognition applications and other image recognition projects, like surveillance applications, commonly use AI. Furthermore, AI-based solutions are highly promising in interpolating the campus-scale data from a few measured values in tasks such as soil moisture collecting. Moreover, to predict the future values of properties measured by sensors based on previously collected values, AI regression algorithms, like support vector regression and deep learning, could be used. Predicting air quality, temperature, and rainfall on campuses can be equally advantageous for decision makers and residents. Additionally, predicting the garbage level in bins and the human density of buildings/locations is also possible. Decision makers can take action beforehand, and residents can make short-term plans according to these predictions
The Department of Computer Engineering at ITU has an application development environment for smart campuses. Fig. 2 shows the general structure of the ITU smart campus development environment. The environment can be examined in three phases: data collection, transmission, and decision making.
Fig 2 The general structure of the ITU smart campus development environment. API: application programming interface; SCA: smart campus application.
First, an air quality-monitoring project has been developed using this environment. Sensors at numerous places on the campus measure air quality data. Then, with the help of a mobile application specially designed for this environment, these data are collected using Bluetooth technology. The general structure of the developed smart campus project is as follows: The sensor measuring air quality sends the obtained value to the microprocessor. Then, the microprocessor processes and forwards the received data to a Bluetooth module. Finally, the Bluetooth module connects to a nearby smartphone with the necessary application and transmits the air quality data. Smartphones deliver incoming data to the database, and all collected data from the various places are visualized over the campus map on the web and mobile applications. Later, a second version of the application, which enables long-range, low-power data transmission on the campus with the help of LoRa technology, was integrated into the environment.
Air quality monitoring is not the only application in development using this environment. The environment will be enriched with other smart campus applications developed by the students in the future. Integrating and managing new sensors in this environment should be simple; therefore, it was decided to create a data platform that facilitates data management. The data platform defines the data format and constructs the communication medium, so the developers do not need to design a custom communication environment and data format. Furthermore, this platform is expected to accelerate the implementation of smart campus projects.
The aim of the designed data platform is to serve the capstone project developers in the department. The platform administrator registers the application developers to allow them to use the system. After they are registered on the system, developers can add their sensors to the environment from the web application of the platform and fetch data from the sensors to their applications via an application programming interface. The developers own all collected data from their sensors. Each developer has the right to control the visibility of the collected data. A developer can give and withdraw privileges for other developers to use her/his data at any time.
The communication infrastructure of smart campuses should cover the whole campus area to collect data from sensor nodes placed all over the campus. Our platform uses LoRa technology to reach the sensors placed in the farthest corners of the campus. A LoRa end device can send data to the centrally placed LoRa gateway. Then, the data collected can reach the platform through a wired Internet connection. However, due to the cost of LoRa modules, installing a LoRa module on each far sensor node is not feasible. Mobile LoRa modules are used to solve this problem. These mobile modules collect data using the Bluetooth modules connected to sensors and then transfer the data to the LoRa gateway via the LoRa end devices. These mobile LoRa modules circulate the campus, collecting data from remote sensors. One possible way to achieve mobility is to install LoRa modules on bicycles or shuttle buses. After fetching the data from the sensor node using Bluetooth technology, the LoRa module forwards it to the LoRa gateway, as shown in Fig. 3. The LoRa gateway sends the data to the web application of the data platform through the Actility ThingPark LoRa-enabled server.
Fig 3 Collecting data with a mobile LoRa end device. ITU ComNet Server: ITU Computer Networks Lab Server.
Sensors should be in the open air for some smart campus projects, such as air quality monitoring. Therefore, these sensors should be protected from adverse outdoor conditions, such as rain and wind. In spring 2022, a group students from the Department of Industrial Design designed and 3D printed a protective case for sensor nodes to prevent possible damage. As can be seen, smart campus applications are nice examples of interdisciplinary collaboration.
Campuses are reasonable environments for developing smart applications by using information and telecommunication technologies because of their spatial and structural advantages. Moreover, smart campus applications are good starting points for those planning to develop smart cities. Therefore, researchers give rising attention to the implementation of smart campus applications. In this article, after discussing basic issues to be considered on smart campuses, we introduce a platform designed to facilitate the development of these applications by students on the ITU main campus. The platform also allows the management of the data collected by the application developers. In the future, the designed data platform can be improved by extending its communication network and adding new functionalities, such as the ability to send control messages to sensor nodes from the platform, download code to some modules, and enhance secure data transmission.
S. Oktug, Y. Yaslan, and H. Gulacar, “A prediction module for smart city IoT platforms,” in Transportation and Power Grid in Smart Cities: Communication Networks and Services, vol. 12. Hoboken, NJ, USA: Wiley, 2019, pp. 269–290.
N. Chagnon-Lessard et al., “Smart campuses: Extensive review of the last decade of research and current challenges,” IEEE Access, vol. 9, pp. 124,200–124,234, Aug. 2021, doi: 10.1109/ACCESS.2021.3109516.
U. Raza, P. Kulkarni, and M. Sooriyabandara, “Low power wide area networks: An overview,” IEEE Commun. Surveys Tuts., vol. 19, no. 2, pp. 855–873, Secondquarter 2017, doi: 10.1109/COMST.2017.2652320.
ThingPark Community. Accessed: Aug. 31, 2022. [Online] . Available: https://community.thingpark.org/index.php/thingpark-interop-engine/
T. M. Fernández-Caramés and P. Fraga-Lamas, “Towards next generation teaching, learning, and context-aware applications for higher education: A review on blockchain, IoT, fog and edge computing enabled smart campuses and universities,” Appl. Sci., vol. 9, no. 21, Oct. 2019, Art. no. 4479, doi: 10.3390/app9214479. [Online] . Available: https://www.mdpi.com/2076-3417/9/21/4479
Bugra Aydin (aydinbu17@itu.edu.tr) earned his B.Sc. degree in computer engineering at Istanbul Technical University. He continues his M.Sc. degree in computer engineering at the same university. He is a software engineer at the Digital Transformation Office of Turkey, Istanbul 34906, Turkey. His research interests include the Internet of Things.
T. Tolga Sari (sarita@itu.edu.tr) earned his M.Sc. degree from Istanbul Technical University (ITU) in June 2021. He is a research assistant and Ph.D. student in the Computer Engineering Department at ITU, Istanbul 34469, Turkey. He received a best paper award at the 6th IEEE International Workshop on Wireless Communications and Networking in Extreme Environments and the 2022 ComSoc Turkey Master Thesis Competition Special Achievement Award. His research interests include next-generation ad hoc networks and media access control protocols.
Sema F. Oktug (oktug@itu.edu.tr) earned her B.Sc., M.Sc., and Ph.D. degrees in computer engineering from Bogazici University, Istanbul, Turkey, in 1987, 1989, and 1996, respectively. She was a visiting researcher in the Department of Electrical and Computer Engineering, New York University Polytechnic School of Engineering, in 1996. Currently, she is a professor with the Department of Computer Engineering, Istanbul Technical University, Istanbul 34469, Turkey. She served as the dean of the Faculty of Computer and Informatics Engineering, Istanbul Technical University, during the period of 2015–2022. She has also served as an evaluator of the Association for Evaluation and Accreditation of Engineering Programs in Turkey since 2014. Her research interests include communication protocols; modeling and analysis of communication networks; wireless networks; low-power, wide-area networks; and the use of wireless communication technologies in smart cities. She is a Senior Member of IEEE.
Digital Object Identifier 10.1109/MPOT.2023.3250968