Smart Learning by Design: A Framework for IoT-Driven Adaptive Classrooms and Inclusive Education

Abstract

This research presents a novel conceptual framework for inclusive education by integrating Internet of Things (IoT)-driven real-time environmental and behavioral monitoring with adaptive teaching strategies. Unlike traditional methods, our model leverages sensor-based data collection to analyze classroom conditions, teacher mobility, and student interactions, enabling dynamic adjustments that aim to enhance engagement and inclusivity. While the framework is theoretical and has not yet undergone experimental validation, we discuss how optimizing spatial configurations, voice dynamics, and movement patterns could support student participation, particularly for learners with diverse needs. Pilot implementations and empirical testing are planned for future research. By merging data-driven insights with educators’ expertise, our approach offers a scalable vision for creating responsive, inclusive learning environments that proactively address barriers to education.

1. Introduction

1.1. Background and Motivation

Over recent decades, global initiatives such as Education for All (1990) and the Convention on the Rights of Persons with Disabilities (2006) have shaped inclusive education policies. Despite these efforts, classrooms today face persistent challenges in ensuring equitable learning experiences for all students, particularly those with diverse abilities. Advances in IoT (Dudhe et al., 2017; Ramlowat & Pattanayak, 2019) and adaptive learning technologies offer promising solutions, yet their integration into educational environments remains limited.

On the one hand, Education for All focused world attention on the basic learning needs of neglected groups and on learning achievement rather than on mere attendance (European Agency for Development in Special Needs Education, 2011). On the other hand, inclusive education promotes mutual respect and value for all persons and builds educational environments in which the approach to learning, the institutional culture, and the curriculum reflect the value of diversity (UNESCO, 2020). To bridge this gap, this study proposes an IoT-based framework to enhance adaptive learning environments. By leveraging real-time data on student engagement and environmental conditions, we provide educators with actionable insights to foster inclusive and responsive teaching strategies. The term IoT-driven adaptive practices is used to refer to the systematic use of IoT technologies to collect, analyze, and interpret real-time data on classroom environmental and behavioral conditions, and to translate these insights into timely instructional or environmental adjustments. Such practices aim to dynamically align teaching strategies, spatial configurations, and learning supports with students’ diverse needs, enabling inclusive, data-informed decision-making that complements educators’ professional judgment.

As supported by Embodied Cognitive Science (Clark, 1997; Varela et al., 1991), the mind and cognitive processes are in fact implanted un sensory–motor processes. The mind is no longer an entity independent of the body because the body itself (intended as a biological, cultural, reflective, spiritual context) is the subject of cognition. Learning is not only an individual achievement but a privileged social occasion during which students encounter, know, and value the heterogeneity of human existence, and thus constitutes a unique opportunity to test those values to desire and design communities that are increasingly inclusive and respectful of differences. In the creation of an Education for All, teachers play a complex and decisive role, in the attempt to pay attention to the differences in the students (cognitive styles, emotional experiences, etc.), to the creation of learning environments that are as universal as possible (so as to facilitate the different ways of functioning present in the classroom), and to the removal of barriers to learning and social participation (for all students). Thanks to initial and ongoing training courses, teachers can integrate specific evidence-based teaching strategies aimed at some students with teaching strategies for all. Teachers, together with family members, managers, and different professionals, both internal and external to the school, can mix and trigger strategies and interventions on multiple levels, ranging from individualized and personalized teaching to class teaching and to collaboration networks with the territory (Murawski & Scott, 2019). Teaching is increasingly intertwined and enriched with/by (new) technologies (applications available on PCs, mobiles, and tablets, multimedia and immersive production platforms and systems, etc.), which are crucial in some situations to access knowledge and to reduce potential inequalities in learning processes for everyone, with no one excluded.

1.2. Our Contribution

While significant progress has been made, traditional classroom settings often struggle to accommodate diverse learning needs effectively. This paper proposes an IoT-driven framework (Al-Emran et al., 2019; Zeeshan et al., 2022) designed to enhance classroom dynamics through real-time data collection and adaptive teaching strategies, enabling a more responsive and inclusive learning environment.

The proposed approach leverages environmental and behavioral IoT sensors to provide continuous insights into classroom conditions, teacher–student interactions, and student engagement. Unlike conventional methods, which rely on static adaptations, the proposed framework introduces a data-informed, real-time adjustment mechanism that supports differentiated instruction and minimizes barriers to learning. This paper specifically examines how this framework can be realistically implemented, focusing on its integration into existing educational settings, with an emphasis on the Italian school system as a case study.

The primary contribution of this study is the bridging of the gap between data-driven insights and practical teaching methodologies. Specifically, a framework is proposed for integrating IoT sensors and adaptive systems to monitor environmental conditions, track student engagement, and provide real-time feedback to educators. This integration enables the design of responsive learning environments that support equitable participation and minimize learning disparities. This study contributes to inclusive education by focusing on detecting and analyzing body and voice interactions to monitor and manage teaching–learning processes.

In the Italian context, nowadays, the technological infrastructure typically available in schools is quite basic. In most cases, it consists of a single computer connected to a video projector, an interactive whiteboard, or a touch-screen monitor, along with a Wi-Fi connection. Such a setup, while useful, exerts only a limited influence compared to what has been described in the literature or implemented in more technologically advanced educational environments. A pilot experiment of technology-enhanced classrooms is currently being implemented in Italy, and the results are intended to be shared in future publications.

This contribution specifically refers to the Italian school landscape where, from the 1970s to today, we have witnessed a process that has seen the concatenation of three concepts and key words: insertion, integration, and inclusion. This process did not stop at a cultural–ideological level but brought and brings with it changes of a political (such as law 104 of 1992 relating to the condition of disability, law 170 of 2010 relating to specific learning disorders, and the MIUR Directive of 27/12/2012 relating to Special Educational Needs), social (in Italy, there are almost no special institutes, differential classes, etc., remaining, and almost all pupils and students, with and without disabilities, attend the same classes and schools), and phenomenological nature (classes and schools represent networks of relationships characterized by a structured superdiversity where over 4% of pupils and students have a disability certification), with repercussions and pushes toward research and training actions increasingly oriented toward an application horizon that is (at least theoretically) fair, inclusive, and intercultural.

For these reasons, the proposed approach is inspired by Universal Design for Learning (UDL), which promotes inclusive teaching methods that inherently integrate specialized approaches, methodologies, and knowledge. By embracing UDL principles, the proposed framework aims to create learning environments that cater to the needs of all students while providing targeted support for those with specific challenges.

UDL is an educational framework aimed at creating accessible and engaging learning environments for all students, taking into account individual needs (CAST, 2011; Mace, 1985). The Internet of Things, with its ability to interconnect devices and collect real-time data, offers new opportunities to implement UDL principles in innovative and personalized ways.

The three core principles of UDL, supported by findings from neuroscience research (Meyer et al., 2014; Rose & Meyer, 2002, 2006), can be effectively integrated into IoT systems:

• Multiple Means of Engagement. This principle focuses on motivating and engaging students in the learning process by offering diverse options to spark their interest and sustain their attention. IoT devices can be employed to create interactive and immersive learning experiences. For example, environmental sensors can monitor classroom temperature, lighting, and noise levels, and automatically adjust these parameters to ensure an optimal learning environment for all students. Moreover, IoT-based applications and platforms can provide personalized feedback to enhance students’ motivation and engagement.
• Multiple Means of Representation. This principle highlights the need to present information in different ways to accommodate students’ varied learning modalities. Data collected by IoT devices can be displayed in multiple formats, such as graphs, tables, or interactive maps, enabling students to access information in ways that best suit their preferences. For instance, a student with reading difficulties may prefer data presented in graphical form, whereas a student with mild intellectual disabilities might benefit from an interactive map emphasizing the most relevant information.
• Multiple Means of Action and Expression. This principle emphasizes offering students various options to demonstrate what they have learned, considering their diverse abilities and preferences. IoT devices can support different modes of expression. For example, students can use sensors and actuators to develop interactive projects that showcase their understanding of scientific or technological concepts. In addition, IoT platforms can provide online collaboration tools that allow students to work together on projects and share their ideas in multiple formats.

2. The Role of the Body in Inclusive Education

2.1. Teacher and Student Movement in Learning Environments

As argued in the Innovative Learning Environment project (OECD, 2017), the physical context of the classroom can represent a barrier or a facilitator in the learning–teaching processes as well as in the quality of the different interactions between teachers and students. The presence of inadequate noise, temperature, and light levels, and uncomfortable seats and supports determine unfavorable conditions at an emotional, behavioral, and cognitive level. On the contrary, the following conditions may be considered particularly favorable for establishing a good classroom climate and more generally a stimulating school experience:

• Choosing and/or modifying visibility, ventilation, acoustics, and the arrangement of desks and furniture;
• Using comfortable areas, informal spaces, corners, and niches where one can dedicate oneself to individual moments of attention, elaboration, reflection, relaxation, and study;
• Easily accessing and moving between the inside and outside of the school, perhaps also thanks to the presence of sensory stimulation (chromatic, olfactory, tactile), functional to exploration and orientation;
• Using communicative, expressive, and technological forms to expand and experiment with knowledge, skills, and competences, enhancing individual differences;
• Modulating spaces for group activities and/or laboratory activities, thus creating contexts as suitable as possible to the type of discipline, the structuring of the various activities, and the teaching choices of the individual teacher.

When designing new buildings or adapting existing ones, maximizing flexibility and responding to specific needs could represent the ideal characteristics to reinforce fully student-centered learning, supporting motivation and participation (Higgins et al., 2005), while at the same time enhancing a sense of belonging to the class and the school community. Pursuing these characteristics, in addition to fully responding to the theoretical principles and operational indications of the UDL approach, recalls the concept (present in the CRPD) of systemic accessibility of physical places, digital and material resources, knowledge, and opportunities that give quality and significance to the daily experiences of a person.

2.2. Physical Proximity of Teachers

The physical proximity that teachers can have with their students plays a significant role in classroom dynamics and at the same time can tell us a lot about the didactics acted out by the teacher. When teaching is for the most part didactic, teachers tend to stay in the chair or in the area immediately surrounding it. Conversely, when the teacher adopts active didactics, creating a more ‘student-centered’ learning environment, they tend to pass more often between the desks of individual pupils or small groups to monitor the progress of the teaching activity and provide feedback while also distributing time and attention effectively among all students (An et al., 2018). Increased physical proximity contributes to more frequent interactions between teachers and students, but also between students and learners, by fostering the co-construction of knowledge (Spillane et al., 2017). Physical proximity between the subjects inhabiting the classroom is able to generate a more empathetic and welcoming environment (Mehrabian, 1967), capable of transforming the class, often made up of individuals who only share the same age, into a group favoring a climate of greater involvement and deeper relationships. Witt et al. (2004) provide a meta-analytical review offering robust empirical evidence on the relationship between teacher immediacy and student learning, emphasizing the role of nonverbal behaviors such as proximity in enhancing educational outcomes. Myers and Rocca (2018) provide a comprehensive overview of teacher immediacy research including both verbal and nonverbal dimensions, such as spatial proximity, and discuss their impact on student engagement and learning.

The support teacher, as a teacher of the whole class, will not sit at the desk of a pupil with disabilities, but, together with their curricular colleague, will increase their movements to be close to the pupils and effectively manage the class. The physical proximity of the teacher to one or more pupils during certain moments of the lesson can be an effective teaching tactic for both intervening in possible disruptive behavior and redirecting pupils to the given task by improving attention. In addition, numerous studies have shown that, during all types of communication, eye contact, close proximity, body posture leaning toward the interlocutor, and smiling represent important indicators that develop communicative intimacy and trust (Burgoon et al., 1984). Baum (2018) highlights how classroom spatial design, including seating arrangements, technological instrumentation, and teacher mobility, affects student behavior and interaction. The study provides a valuable perspective on how physical and technological elements of the classroom shape communication dynamics. Wang et al. (2016) explore the impact of digital equipment on the quality of classroom interactions in technology-rich learning environments. Drawing on observational data from 54 lessons conducted across Hong Kong, Beijing, and Shenzhen, the study analyzes 626 interaction processes to assess how tools such as interactive whiteboards and mobile devices influence communication dynamics. The authors highlight that while digital technologies can enhance engagement and facilitate diverse forms of interaction, their effectiveness depends on how they are integrated into pedagogical practices. The study underscores the need for thoughtful classroom design and teacher training to ensure that technology supports, not hinders, meaningful educational exchanges.

2.3. Voice and Communication Dynamics

In the design and management of teaching and learning pathways, particular importance is attached to the frequency and mode of interaction both between the teacher and students and between the students themselves. When designing, what are referred to as instructional architectures can be applied, i.e., strategies of instruction that also differ in the intensity and manner of interactions. In fact, the way in which the four architectures can be taxonomically listed meets this criterion (

Table 1. Comparison of instructional architectures and their implications for inclusive education.

Instructional Architecture	Interaction Level	Teacher’s Role	Student’s Role	Implications for Inclusive Education
Receptive Architecture	None	Transmissive	Passive recipient	Limited inclusivity; not adaptable to diverse needs.
Behavioral Architecture	Frequent	Stimulus–response guide	Active responder	Encourages engagement but may not fully support autonomy.
Situated Guided Discovery Architecture	High	Facilitator	Knowledge constructor in context	Supports inclusivity by adapting to specific learning contexts.
Exploratory Architecture	Optional	Minimal guidance	Central role in knowledge construction	High inclusivity; fosters autonomy and personalized learning.

).

After the first architecture, the receptive architecture, characterized by the absence of interaction and the transmissive role of the teacher, the next architecture is the behavioral model, based on the stimulus–response model, which implies forms of frequent interaction. In the third architecture—the situated guided discovery architecture, which, compared to predetermined forms of knowledge, emphasizes the construction of knowledge in relation to specific contexts—highly interactive learning environments are to be assumed from the design stage. The fourth and last architecture, the exploratory architecture, envisages the centrality of the learner’s mental process with respect to externally offered processes and makes interaction with the teacher optional. The presence and mode of interaction between the teacher and the learner make it possible to distinguish approaches and the management of learning paths. And, although interaction passes through a multiplicity of forms, such as those linked to proxemics and emotions, voice represents an ineliminable channel that can be monitored, giving an account of the type of interaction and its intensity—providing in this sense indications of not only the prevailing type of architecture but also the gap between the design setting and the actually existing situation. In this sense, studies on turn management in voice interaction (turn-taking) within the classroom can be retrieved. Through use of well-established models, such as the conversation analytic approach, it is possible to identify multiple types of interactions such as, for example, the Adjacency Pair, which includes pairs, such as in the exchange of greetings, the question–answer alternation (Sari et al., 2023), the types of pauses/silences in verbal interactions (Ingram & Elliott, 2014) or more active forms of participation as the years go by Maroni et al. (2008). In didactic terms, the study of such interactions can, on the one hand, constitute a form of evaluation of the courses, showing their coherence with or deviation from the design framework, and, on the other hand, can provide indications of modifications in itinerary to implement, for example, individualized actions toward specific students.

3. Monitoring Interaction in the Classroom Through IoT

3.1. Non-Intrusive Environmental IoT Sensors for Monitoring Interaction and Engagement

Advancements in Internet of Things (IoT) technology offer unprecedented opportunities to monitor and analyze classroom dynamics in a non-intrusive manner. By leveraging IoT sensors to collect real-time data on environmental conditions, educators can gain insights into the interactions and engagement levels of students without directly interfering in the learning process. These sensors can monitor parameters such as temperature, lighting, noise levels, and air quality, all of which significantly influence cognitive styles and emotional states, thereby affecting overall classroom engagement. For example, suboptimal lighting can lead to visual discomfort and reduced attention spans, while high noise levels may cause distraction and hinder communication. IoT-enabled environmental sensors can provide actionable data to adjust these variables, creating optimal learning conditions. Furthermore, by correlating environmental data with behavioral observations, patterns indicative of student engagement or disengagement can be identified. This non-intrusive approach integrates seamlessly into classroom activities, offering educators feedback to enhance the learning environment. By addressing the diverse needs of students, including those with disabilities, IoT systems support the creation of an inclusive and universally accessible educational setting. For instance, temperature and light adjustments based on real-time data can accommodate sensory sensitivities, while noise-reducing strategies informed by sound level monitoring can support students with auditory processing challenges. The integration of IoT in classrooms exemplifies the Universal Design for Learning (UDL) approach by removing barriers and fostering an inclusive, adaptive learning environment. This data-driven methodology provides educators with the tools to make informed decisions, ensuring that all students benefit equally from the educational process.

3.2. Ambient Sensors for Tracking Student Participation

IoT ambient sensors are pivotal in tracking student participation, offering a quantitative measure of involvement in classroom activities Miazek et al. (2024). These sensors, which can include motion detectors, proximity sensors, and acoustic analyzers, capture data on physical movements, spatial dynamics, and verbal interactions within the classroom, with more detail provided as follows:

• Movement and Engagement. Tracking movement through sensors can reveal patterns of teacher and student mobility, providing insights into teaching strategies and engagement levels. For instance, active teaching methods are often characterized by higher mobility as teachers circulate among students, facilitating discussions and providing feedback. Similarly, student movement during collaborative or hands-on activities can indicate high levels of participation. Conversely, a lack of movement might signal disengagement, prompting educators to reassess their methods.
• Proximity and Interaction. Proximity sensors can monitor the distance between teachers and students, highlighting areas of the classroom that may require more attention. Increased proximity often correlates with higher interaction levels, contributing to a more inclusive and empathetic learning environment. Teachers who frequently move closer to students foster a sense of involvement and accessibility, aligning with inclusive education goals.
• Acoustic Data Analysis. Voice and sound data collected through acoustic sensors provide insights into communication dynamics (Pleshkova et al., 2020). Turn-taking patterns, frequency of participation, and types of interactions can be analyzed to assess engagement. For instance, a balanced distribution of speaking turns among students suggests an equitable participation model, while excessive teacher talk may indicate a need for more student-centered approaches.

Ambient sensors offer a comprehensive view of classroom participation, bridging the gap between qualitative observations and quantitative data. By identifying participation trends, educators can tailor their strategies to foster a collaborative and dynamic learning environment. This approach not only supports the principles of inclusive education but also enhances the overall classroom experience by addressing the diverse needs of learners.

Table 2. IoT-based classroom monitoring aspects.

Aspect	Relevance to IoT-Based Classroom Monitoring
Movement and Engagement	- IoT sensors track movement to assess teaching strategies and engagement levels. - Higher teacher mobility fosters interaction and feedback. - Student movement signals participation; reduced movement may indicate disengagement.
Proximity and Interaction	- Proximity sensors identify areas needing attention. - Closer proximity enhances engagement and real-time intervention. - For support teachers, movement and proximity reflect inclusive participation.
Voice and Communication	- Acoustic sensors analyze voice dynamics for interaction quality. - Turn-taking, participation, and teacher talk time indicate discussion balance. - Silence patterns reveal engagement and the need for teaching adjustments.

highlights how IoT-driven insights on movement, proximity, and voice dynamics help teachers create responsive and adaptive learning environments that promote inclusivity and equitable participation.

3.3. Data-Driven Adaptive Learning from Classroom Dynamics

IoT technologies enable the collection of vast amounts of data on classroom dynamics, creating opportunities for data-driven adaptive learning. By analyzing patterns in the data, educators can design personalized and responsive learning experiences that cater to the unique needs of each student.

• Adaptive Interventions. Data collected by environmental and ambient sensors can inform targeted interventions. For instance, if data shows that certain areas of the classroom consistently exhibit lower engagement levels, educators can experiment with rearranging seating, adjusting lighting, or introducing interactive activities in those zones. Similarly, if participation data highlights disparities among students, teachers can implement strategies to encourage quieter students to contribute, such as structured turn-taking or smaller group discussions.
• Real-Time Feedback. IoT systems provide real-time feedback, allowing for dynamic adjustments to teaching methods and classroom conditions. For example, sensors detecting increased noise levels might indicate a need for redirection or a shift in activity. Likewise, monitoring trends in movement and proximity can help teachers evaluate the effectiveness of their strategies and adapt their approaches in the moment.
• Longitudinal Analysis. Beyond immediate interventions, IoT data supports longitudinal analysis, revealing trends over time that can inform curriculum development and instructional design. Patterns in engagement, participation, and environmental conditions can guide the development of strategies that are both effective and sustainable. For instance, consistent data showing that collaborative activities yield higher engagement could lead to the integration of more group-based projects into the curriculum.

3.4. Bridging the Gap Between Technology and Education

Pedagogical insights on classroom architecture, proxemics, and instructional models are not meant to act as an independent background; rather, they define the variables and interaction patterns that the IoT system is designed to measure, interpret, and ultimately support. For example, physical proximity between teachers and students, teacher mobility across learning zones, and the intensity of verbal interactions (as described in Table 1) are conceptualized as proxies for instructional architectures (from receptive to exploratory). By mapping these interaction models to quantifiable sensor data, such as proximity metrics, movement trajectories, and acoustic turn-taking patterns, IoT measurements can be aligned with pedagogical intent.

The proposed sensor measurements are explicitly informed by these educational models. Motion and proximity sensors monitor teacher and student movement patterns to distinguish, e.g., transmissive versus student-centered instructional modes. Acoustic analysis captures the distribution and rhythm of verbal interactions, identifying whether participation is balanced (as expected in situated or exploratory architectures) or dominated by the teacher (as in receptive or behavioral models). Environmental sensors address the physical classroom characteristics that research has linked to cognitive and emotional states (e.g., lighting, temperature, noise) and therefore to engagement and inclusivity.

In other words, the IoT system is not simply collecting raw data; it is structured around interpretive layers derived from established pedagogical frameworks. The classroom architecture, teacher–student proximity, and interaction intensity provide the interpretive scaffold that makes sensor outputs meaningful for adaptive instruction. This synergy enables educators to move from passive observation to active, data-informed decision-making, aligning real-time environmental and behavioral adjustments with the desired instructional architecture and inclusive teaching strategies.

4. Potential Application Scenario

4.1. IoT-Driven Adaptive Learning to Minimize Learning Gaps

This section aims to present a hypothetical case study where all the monitoring IoT components, previously described, can be implemented for enhancing the learning experience and for supporting inclusive education.

The goal is to hypothesize how IoT technologies can help educators create responsive and inclusive learning environments by addressing individual and group needs. By leveraging data from environmental and participation sensors (both wearable and stationary), the system aims to foster equitable engagement and performance across all students.

Adaptive learning systems, which rely on Artificial Intelligence (AI) algorithms, adjust learning processes to the behavior and performance of individual students, generating a tailored learning environment (Garavaglia & Petti, 2022). These systems are designed to provide flexible learning activities, content, and timing to meet the diverse needs of learners. Promising results have been shown, particularly in terms of their capacity to support learning on a large scale through the application of big data and learning analytics techniques.

Adaptive learning systems, while effective in personalizing learning, are limited without environmental and behavioral data. Integrating IoT sensors provides real-time insights into factors such as student interactions, movements, and emotional states and cognitive styles within the classroom. This added layer of information helps identify learning gaps and offers a more comprehensive understanding of the variables influencing student outcomes. Teachers and administrators can use these insights to pinpoint areas where gaps may arise due to environmental or behavioral factors, fostering a truly responsive learning environment. By combining adaptive learning with IoT-driven data, the educational experience is enhanced for individual students, helping to reduce learning disparities across the broader student population (Burunkaya & Duraklar, 2022).

The proposed implementation relies on IoT sensors to monitor environmental variables such as lighting, noise levels, and air quality, along with ambient sensors tracking student interactions and movements, and physiological sensors to monitor emotional states and cognitive styles. This data allows educators to recognize where learning gaps may emerge from environmental or behavioral influences.

Figure 1 illustrates a conceptual framework for an IoT-driven adaptive learning system aimed at minimizing learning gaps in classroom settings, providing an overview of the key components and interactions necessary to create a responsive and inclusive learning environment. The system consists of three main components:

• Local IoT Data Acquisition Component. This includes the deployment of IoT sensors to track environmental condition variables, as well as student and teacher interactions, movements, and acoustics, providing insights into behavioral and participation patterns. This component may include a local preliminary data processing unit, such as IoT gateways, which aggregate data before transmitting it to the cloud-based platform.
• Local IoT Actuator Component. This includes the deployment of IoT actuators to automatically manage the environmental conditions based on the system real-time feedback.
• Cloud-Based Data Management and Processing Component. This is part of the in-cloud platform, and it includes centralized AI algorithms that process the collected environmental and behavioral data, as well as additional inputs, delivering real-time feedback to teachers (e.g., through an interactive dashboard). Teachers can use this feedback to dynamically adjust classroom strategies, such as transitioning to quieter activities during noise spikes or incorporating movement-based learning when inactivity is detected.

The bidirectional link between the local and cloud system components is depicted in Figure 1, highlighting how communication flows between IoT devices, the internet, and the processing components.

The overall operation of the system involves continuous data collection via IoT sensors, which are then processed by AI algorithms to deliver actionable insights. These insights empower teachers to address individual and group needs in real time, fostering equitable engagement and improved learning outcomes. The framework emphasizes the importance of leveraging IoT technologies to create adaptive learning environments that reduce educational disparities and support diverse learning needs, addressing the following:

• Environmental Optimization: Data from sensors could identify classroom zones with less favorable conditions, such as poor lighting or excessive noise. Adjustments like rearranging seating, optimizing lighting, or installing acoustic panels would then be tested to improve engagement.
• Participation Insights: Ambient sensors would track student involvement by analyzing verbal and physical participation. Trends such as unequal turn-taking or disengagement in specific groups could inform tailored teaching strategies, such as targeted prompts or peer-assisted activities.
• Real-Time Feedback Mechanism: Teachers would receive real-time updates via an interactive dashboard, enabling immediate adjustments to teaching methods or classroom settings. For instance, sudden spikes in noise levels might prompt a shift to quieter activities, while low-movement patterns could signal the need for active learning interventions.

Although the framework has yet to be implemented, the expected outcomes include the following:

• Improved Engagement: Students are anticipated to show increased participation and interaction when learning environments align with their cognitive and sensory needs.
• Reduced Disparities: The system is designed to bridge participation and achievement gaps, particularly for students who may face challenges due to environmental or socio-emotional factors.
• Data-Driven Insights: Teachers and administrators could use aggregated data to refine curricula and make informed decisions about classroom design and instructional strategies.

This conceptual case study demonstrates the potential of IoT-driven adaptive learning systems to address learning gaps and support inclusive education. By hypothesizing the integration of IoT technologies with data-informed teaching practices, this model outlines a pathway toward more equitable and personalized education. Future research and pilot implementations will be essential to validate these hypotheses and refine the approach for widespread application.

A robust IoT-driven adaptive classroom framework relies on a diverse ecosystem of sensors, each selected for its ability to capture specific environmental, behavioral, and physiological variables that influence learning. The integration of these technologies is not merely a technical choice but a pedagogical one, as each sensor type provides actionable insights that can inform inclusive, data-driven teaching strategies.

Sensing and networking technologies can be divided into the following categories:

• Environmental Monitoring Sensors:

  –

  Temperature and Humidity Sensors: Maintain thermal comfort, which is directly linked to cognitive performance and attention. Real-time readings can trigger heating, ventilation, and air conditioning (HVAC) adjustments to prevent discomfort that may distract learners.

  –

  Light Sensors: Detect suboptimal lighting conditions that can cause visual strain or reduce alertness. Data can inform automated lighting adjustments or seating rearrangements to improve visibility for all students, including those with visual impairments.

  –

  Noise Level Sensors: Monitor ambient sound to identify disruptive noise spikes or sustained high levels that may hinder communication, particularly for students with auditory processing challenges.

  –

  Air Quality Sensors: Track pollutants, CO₂ concentration, and particulate matter, which can affect concentration and wellbeing. Alerts can prompt ventilation or air purification interventions.

• Motion and Proximity Sensors:

  –

  Passive Infrared Motion Sensors: Detect movement patterns of teachers and students, offering insights into mobility, engagement, and instructional style.

  –

  Ultrasonic Distance Sensors: Measure proximity between teachers and students, highlighting interaction zones and identifying areas of the classroom that may be underserved.

  –

  Radar Sensors: Provide high-precision detection of micro-movements, gestures, and breathing patterns without capturing visual imagery, making them ideal for privacy-sensitive environments. Radar can track presence and subtle activity even in low-light or visually obstructed conditions, enabling continuous monitoring of engagement and movement while minimizing personal data collection risks.

  –

  Computer Vision with AI (via Cameras or LiDAR): Recognize gestures, posture, and clustering behavior, enabling analysis of group dynamics and participation equity.

• Voice and Communication Sensors:

  –

  Microphone Arrays: Capture spatial audio data to determine the source and distribution of speech, supporting analysis of whether discussions are teacher-dominated or student-centered.

  –

  Directional Audio Sensors: Identify turn-taking patterns, conversational flow, and participation frequency, helping educators adjust strategies to encourage quieter students.

  –

  Speech-to-Text AI Integration: Transcribe and analyze classroom dialogue for content coverage, inclusivity of participation, and alignment with lesson objectives.

• Wearable and Physiological Sensors:

  –

  Heart Rate Sensors: Detect stress or heightened arousal during specific activities, enabling timely interventions for anxious or disengaged students.

  –

  EEG Brainwave Sensors: Provide insights into cognitive load and attention levels, useful for research-driven adaptive learning environments.

  –

  Galvanic Skin Response Sensors: Measure physiological arousal, which can be correlated with engagement or anxiety levels.

• Integration and Data Processing Components:

  –

  IoT Gateways: Aggregate multi-sensor data locally before transmitting to cloud platforms, reducing latency and enabling real-time feedback loops.

  –

  Edge AI Modules: Perform preliminary analysis on-site, allowing immediate environmental adjustments without reliance on constant internet connectivity.

Table 3. Summary of IoT sensors for classroom monitoring.

Category	Types	Applications
Environmental Sensors for Monitoring Classroom Conditions	- Temperature and Humidity Sensors - Light Sensors - Noise Sensors (Acoustic Level Monitors) - Air Quality Sensors	- Ensure thermal comfort for students. - Detect temperature variations affecting cognitive performance. - Can trigger HVAC (heating, ventilation, and air conditioning) adjustments in smart classrooms.
Motion and Proximity Sensors for Tracking Teacher and Student Movement	- Infrared Motion Sensors (PIR—Passive Infrared Sensors) - Ultrasonic Distance Sensors - Radar Sensors - Computer Vision with AI (via Cameras and LiDAR Sensors)	- Detect movement in designated classroom zones. - Identify students who are inattentive or disengaged. - Track the distance between teachers and students to study proximity-based engagement. - Detect whether students are clustering or staying apart. - Recognize gestures, movement, and body posture. - Track student interactions and emotional responses through facial expression analysis.
Voice and Communication Sensors for Monitoring Classroom Interaction	- Microphone Arrays - Speech-to-Text AI Integration - Directional Audio Sensors	- Analyze speech patterns and participation frequency. - Detect whether classroom discussions are teacher-dominated or student-centered. - Identify turn-taking patterns and conversational flow. - Detect hesitation or long silences, signaling disengagement. - Track the location of sound sources in the classroom. - Help determine whether students at the back are participating less.
Wearable and Physiological Sensors for Personalized Monitoring	- Heart Rate and Stress Sensors - EEG Brainwave Sensors - GSR (Galvanic Skin Response) Sensors	- Monitor stress levels in students during lectures or exams. - Identify students struggling with anxiety or cognitive overload. - Assess cognitive load and attention levels. - Used in research-driven adaptive learning environments. - Detect physiological arousal and stress responses.
Integration and Data Processing	- IoT Gateway	- Aggregates sensor data and transmits it to cloud servers.

summarizes the sensors that can be used for classroom monitoring in IoT-based smart learning environments.

4.2. Integrating Sensors into Didactic Design: A Practical Example

This section presents a structured example of lesson design that integrates technological tools for monitoring and analysis. The model divides the lesson into three phases (introductory, central, and concluding), each associated with specific activities, objectives, and evaluation elements. By leveraging audio and motion sensors, the approach aims to provide real-time feedback and long-term insights into classroom dynamics, supporting both teachers in instructional regulation and students in developing awareness of their engagement levels. Figure 2 illustrates an example of a sensor-enhanced educational framework across three lesson phases.

Whenever the teacher enters the classroom, they implement a prearranged micro-didactic design, which dynamically unfolds within the school context. This design includes didactic phases, understood as segments of teaching action characterized by the interaction between the teacher’s activity and that of the students (Laurillard, 2010). While acknowledging the impossibility of defining a single teaching approach, given the flexibility with which the teacher chooses and adapts methods, methodologies, techniques, and tactics according to specific objectives, it is possible to outline a paradigmatic lesson structured in three phases: introductory, central, and concluding.

Suppose that in the introductory phase, the teacher adopts the seminar-style lecture strategy. Audio sensors could monitor in real time the duration of interventions, the degree of student interaction, the prevalence of teacher monologue, and the presence of background noise. At the same time, motion sensors could detect the mobility of the teacher and students, signaling excessive staticity or movements indicative of inattention. These data, when shared and analyzed, provide the teacher with tools for didactic regulation and give students greater awareness of concentration levels in the classroom.

In the central phase, the teacher might assign a collaborative activity in small groups. Audio sensors could monitor interaction levels among students, identifying possible exclusion dynamics, as well as teacher–student interactions. Motion sensors could track the teacher’s movements among groups and analyze students’ postures, inferring the level of engagement (relaxed or slouched postures might indicate less interest compared to attentive postures oriented toward peers).

Finally, in the concluding phase, a debriefing could benefit from monitoring participation levels through audio sensors, detecting interactions between the teacher and students and among students themselves. Even silence, interpreted in context, can provide indications of class engagement. Motion sensors could detect whether groups have the opportunity to move and present their results to the class.

Table 4. Summary of a hypothetical lesson scheme in a smart classroom.

Didactic Phase	Activity	Architecture	Elements to Evaluate	Specific Goals	General Goals
Introductory	Seminar-style lecture	Transmission-based	Teacher’s speaking time; Noise level; Student interactions; Teacher–student interactions	Identify optimal speaking time; Assess attention; Verify interactions	Check consistency between design and actions
Central	Group work	Collaborative	Teacher–student and student–student interactions (proximity and audio)	Identify optimal timing for group management and interaction	Check consistency between design and actions
Concluding	Debriefing	Collaborative	Teacher–student and student–student interactions; Silence duration	Identify optimal timing for whole-group management and responses	Check consistency between design and actions

provides a summary of a hypothetical lesson, highlighting the elements that technologies would monitor. Collecting data on the entire class also makes it possible to assess the participation of all students and teachers involved, and the data collected could be used in real time or for long-term analysis.

5. Challenges

IoT technologies are evolving rapidly, offering significant opportunities to enhance education. However, their implementation comes with challenges that require careful consideration, such as effectively integrating data into teaching practices, addressing ethical concerns like privacy, and ensuring inclusivity and accessibility. Overcoming these hurdles is essential to fully realizing the potential of IoT in creating adaptive and equitable learning environments.

5.1. Integrating IoT Data with Educator Expertise for Holistic Insights

To unlock the full potential of IoT technologies in education, it is essential to integrate the quantitative data provided by sensors with the qualitative insights of educators. Teachers’ contextual understanding and professional judgment are critical for refining and enriching data interpretation, ensuring that decisions are tailored to the unique dynamics of each classroom. For example, acoustic sensors detecting elevated noise levels might signal productive group work rather than student disengagement, and patterns in movement or proximity could reflect specific teaching strategies or cultural norms. By combining IoT data with educators’ observations, schools can move beyond surface-level indicators to address underlying issues and create targeted interventions.

5.2. Optimizing Teaching Practices

The integration of IoT data with teacher-led observations offers a powerful mechanism for improving pedagogical strategies. Examples include the following:

• Spatial Dynamics: Proximity data can inform seating arrangements that encourage equitable participation and collaboration.
• Communication Patterns: Acoustic analysis of speaking turns can identify opportunities to foster more inclusive dialogue, ensuring that quieter students are heard.
• Environmental Adjustments: Sensor data on lighting, temperature, and noise levels can be used to create tailored accommodations for sensory sensitivities, particularly for students with disabilities.

By synthesizing IoT-generated insights with qualitative inputs, educators can create more adaptive and equitable classroom environments, aligning with Universal Design for Learning (UDL) principles.

To effectively leverage IoT data, educators need training in data analysis and ethical data management. Professional development programs should equip teachers with the skills to interpret sensor outputs and integrate them with classroom observations. Additionally, ethical considerations, such as maintaining student privacy and fostering transparency, must be integral to these programs.

Effective integration of IoT technologies into teaching practices requires targeted professional development that equips educators with both the technical competencies and the pedagogical strategies necessary to leverage these tools for inclusive, adaptive learning. Training programs should adopt a blended approach, combining hands-on workshops on IoT device operation, data interpretation, and troubleshooting with collaborative sessions that explore pedagogical applications aligned with UDL principles. Scenario-based simulations can help teachers practice responding to real-time sensor feedback, such as adjusting environmental conditions, modifying seating arrangements, or initiating targeted interventions based on participation data, while maintaining instructional flow. Peer-learning communities and ongoing coaching can further support the translation of IoT insights into differentiated instruction, ensuring that technology serves as an enabler rather than a distraction. Embedding ethical data management and privacy compliance into all training modules will also foster responsible use, building educator confidence in applying IoT-driven strategies to enhance engagement, equity, and learning outcomes.

5.3. Ethical Considerations and Privacy

While the potential of IoT in education is vast, it is crucial to address ethical considerations, particularly regarding data privacy and security. Transparent policies and robust encryption protocols must be implemented to ensure that sensitive data is protected and used responsibly. Involving stakeholders—including students, parents, and educators—in discussions about data use fosters trust and ensures alignment with ethical standards.

To ensure ethical integrity and legal compliance in IoT-driven adaptive learning environments, robust data privacy measures must be embedded into the system’s design from the outset. In alignment with the General Data Protection Regulation (GDPR), all personal and behavioral data collected through environmental, proximity, and acoustic sensors should be subject to explicit, informed, and revocable consent protocols. These protocols must clearly communicate to students, parents, and educators the nature of the data being collected, its intended use, the retention period, and the rights of data subjects, including access, rectification, and erasure. Data minimization principles should guide the collection process, ensuring that only information strictly necessary for pedagogical purposes is gathered. Furthermore, anonymization or pseudonymization techniques should be applied wherever possible to reduce the risk of re-identification, and secure storage with end-to-end encryption should be implemented to safeguard against unauthorized access. Regular privacy impact assessments, coupled with transparent reporting to stakeholders, can reinforce trust and accountability, ensuring that the integration of IoT technologies in education not only enhances inclusivity and engagement but also upholds the highest standards of data protection.

5.4. Inclusive Education Through Data

IoT systems can significantly enhance inclusive education by identifying and addressing barriers to learning and participation. For example, data highlighting patterns of disengagement among students with specific needs can prompt the development of tailored interventions, such as alternative communication methods or sensory accommodations. By embracing data-driven practices, educators can create environments that truly reflect the diversity of their students.

5.5. Technological Considerations

The integration of IoT technologies into educational settings presents both opportunities and challenges, particularly when fostering inclusive and adaptive learning environments. Key technological aspects to consider include system design, infrastructure requirements, accessibility, and data security.

A significant concern is the integration of diverse technologies within existing infrastructures. Many educational institutions lack the high-speed internet, modern hardware, and network capacity required to support IoT systems effectively, leading to potential compatibility and scalability issues.

Data privacy and security also pose substantial challenges, as IoT systems often involve continuous monitoring and data collection. Establishing robust encryption, data anonymization, and secure storage practices is essential to protect sensitive information while complying with regulations like GDPR.

Moreover, maintenance and reliability of IoT systems can strain resources. Ensuring that sensors, devices, and software function seamlessly requires continuous monitoring and technical expertise, which may not always be readily available in educational settings.

Finally, there is a need for educator training to maximize the benefits of these technologies. Without adequate professional development, teachers may struggle to interpret data or effectively integrate technological tools into their pedagogy, limiting the impact of IoT-driven solutions.

Addressing these challenges will require strategic planning, investment, and collaboration across stakeholders to unlock the transformative potential of technology in education.

6. Validation Strategy and Future Implementation

The proposed IoT-based classroom monitoring framework is designed to enhance inclusive education by leveraging well-established theories and advanced IoT technologies. While real-world deployment is the ultimate goal, a structured validation and implementation strategy has been defined to ensure its feasibility and effectiveness (

Table 5. Phases of validation and implementation strategy.

Phase	Description
Stakeholder Engagement	Structured interviews and surveys with educators, administrators, students’ parents, disabled people associations, and special needs specialists to gather empirical insights and refine system design.
Simulated Environment Testing	Controlled experiments in simulated classrooms to assess IoT sensor reliability, data collection accuracy, and initial correlations between classroom conditions and student engagement.
Pilot Classroom Deployment	Gradual implementation in real educational settings to evaluate key performance indicators such as engagement levels, participation rates, and adaptability.
Longitudinal Evaluation and AI-Driven Adaptation	Continuous data collection and AI-driven analysis to optimize adaptive learning strategies, ensuring scalability and long-term effectiveness.

).

This multi-phase approach ensures a robust transition from theoretical development to real-world application, paving the way for an inclusive, data-driven learning environment.

7. Conclusions

This study introduces a theoretical IoT-based framework for monitoring and enhancing inclusive classroom dynamics. By integrating environmental, motion, proximity, and acoustic sensors, the approach provides real-time insights into teacher–student interactions, engagement levels, and learning conditions. The findings emphasize the importance of teacher mobility, proximity-based interactions, and voice dynamics in fostering an adaptive and inclusive learning environment. Although no empirical testing has been conducted, the proposed framework is built upon well-established educational theories and IoT methodologies, positioning it as a strong candidate for future real-world validation.

Moving forward, research should focus on pilot implementations in authentic classroom settings to gather empirical data and refine engagement models. Additionally, incorporating AI-driven adaptation could further enhance the system’s ability to dynamically optimize teaching strategies based on real-time sensor inputs. Further validation of the framework will involve collaboration with a large group of educators and utilizing existing classroom engagement datasets for preliminary conceptual validation. These steps will bridge the gap between theoretical development and real-world application, ensuring that the framework can be effectively implemented and refined.

Ethical considerations related to student data privacy must be carefully addressed to ensure compliance with regulations and build trust in IoT-driven educational tools. Longitudinal studies will also be necessary to assess the long-term impact of this approach on student performance and inclusivity. By bridging technological advancements with pedagogical principles, future work can translate these insights into practical, data-driven strategies that promote equitable participation and personalized learning.