RFID-embedded mattress for sleep disorder detection for athletes in sports psychology

Abstract

Sleep disorders like sleep apnea and insomnia significantly impair athletes’ recovery and performance. Sleep apnea, exacerbated in supine positions due to increased airway resistance, and insomnia, evidenced by fragmented sleep and restlessness, highlight the necessity of monitoring sleep postures. This study introduces a novel RFID-embedded smart mattress capable of non-invasive monitoring and detection of these disorders by capturing body postures and movements using passive RFID sensors. A multi-layered mattress design integrates advanced RFID technology with machine learning algorithms—Gaussian process regression (GPR) and linear regression (LR)—to classify postures and detect movement anomalies. Evaluated with data from five participants in supine and prone positions, the system achieved a posture recognition root mean square error (RMSE) of 0.42 and movement detection RMSE of 0.15. Data processing included standardization and Gaussian filtering for enhanced accuracy, with a 5-fold cross-validation framework ensuring robust performance. The results demonstrate the mattress’s effectiveness as a cost-efficient, non-intrusive alternative to traditional polysomnography, offering insights for early detection and management of sleep disorders. This approach shows significant potential for sports psychology applications, enabling personalized recovery strategies and performance optimization. Future work will focus on expanding the dataset, integrating additional biometric sensors, and refining algorithms to improve diagnostic accuracy and real-time usability in clinical and home settings.

Introduction

Monitoring sleep quality and detecting sleep disorders are essential for athletes, as both play a major role in recovery and overall performance. When sleep is disturbed, it can negatively impact an athlete’s physical and mental health, leading to poor performance and an increased risk of injury¹. Quality sleep allows muscles to repair, restores energy levels, and prepares the body for peak performance. Without quality sleep, athletes often experience reduced strength, speed, and endurance, all of which are critical in high-pressure sports^1,2. Sleep deprivation can also slow reaction times and diminish strength, making it harder for athletes to compete at their best³.

Sleep disorders such as sleep apnea, insomnia, and restless leg syndrome disrupt normal sleep patterns, leading to fatigue, reduced cognitive function, and impaired physical performance. For athletes, this can translate into slower reaction times, diminished endurance, and an increased risk of injuries. These disorders can also lead to long-term health problems. For example, athletes with sleep apnoea may experience mood disturbances like depression and anger, and the condition can worsen cognitive issues, especially in those with a history of head injuries⁴. Early detection and monitoring of sleep disorders are vital for both immediate and long-term health. These practices not only support physical recovery but also optimize mental well-being, leading to better athletic performance.

Monitoring vital signs such as heart rate and respiration rate during sleep is key to identifying sleep issues and assessing overall recovery. By managing sleep more effectively, athletes can improve their cardiovascular health, boost endurance, and sharpen their mental focus, all of which are essential for staying at their peak. Athletes’ sleep patterns, or chronotypes—whether they are early risers or night owls—also influence how well they perform. Those with a late chronotype often experience lower sleep quality and reduced athletic performance compared to early risers⁵. Sleep also affects cognitive functions like attention, learning, memory, and decision-making, which are key in sports⁶. A lack of sleep can impair an athlete’s ability to make sound decisions, especially in high-pressure situations^2,7. This link between sleep and brain function highlights how vital rest is for learning new skills and strategies, which are critical for strategic decision-making in sports⁷. Sleep deprivation not only impacts decision-making but also increases the likelihood of injury and slows the recovery process. Athletes who don’t get enough sleep are at higher risk for injuries like concussions due to weakened physical and cognitive performance. Proper rest is also necessary for recovering the nervous and endocrine systems, both of which are essential for reducing injury risk and speeding up recovery after intense exercise¹.

Sleep apnea, characterized by interruptions in breathing during sleep, often worsens in supine sleeping positions due to increased airway resistance^8–12. Similarly, insomnia frequently manifests through fragmented sleep and restlessness, which can be detected by monitoring posture changes and frequent movements during the night^13–15. Understanding the relationship between sleep posture and these disorders is critical for targeted interventions. For example, certain postures, such as prolonged supine sleeping, have been associated with sleep apnea due to mechanical compression of airways. Conversely, frequent posture shifts and restless movements are indicative of insomnia or discomfort. These connections highlight the importance of posture monitoring in diagnosing and managing sleep disorders.

Early detection of sleep disorder helps athletes avoid performance impairments caused by poor sleep quality and frequent breathing interruptions^16–18. Prior research has explored various sensor technologies for sleep monitoring, including pressure sensors, ballistocardiographs, and radar-based systems, which have demonstrated the potential to detect sleep postures and disturbances. One promising development in this area is the use of smart beds equipped with advanced sensor technology to detect sleep apnoea. These beds integrate sensors and Internet of Things (IoT) technologies to monitor important physiological signals, offering a non-invasive and affordable alternative to traditional sleep studies. For athletes, these smart beds provide continuous monitoring, helping them improve sleep quality, recovery, and overall performance. Smart beds use sensors like ballistocardiograph, pressure sensors, and microphones to track heart rate, breathing, and sleep posture. These sensors can detect subtle changes that may indicate sleep apnoea, enabling early detection and helping athletes avoid the performance problems caused by poor sleep. The data collected is then analyzed using machine learning algorithms to identify patterns linked to sleep disorders, giving a detailed view of how sleep disturbances affect recovery and performance. For example, logistic regression models using demographic and physiological data can achieve high sensitivity for detecting apnoea. This analysis provides a more comprehensive view of how sleep disturbances might affect athletes’ recovery and performance¹⁶.

However, these methods often involve intrusive setups, high costs, or limited adaptability to diverse environments. RFID-based systems offer a novel solution by enabling non-invasive, continuous monitoring of sleep posture and movements. This approach leverages the interaction of radio frequency signals with pressure distribution and body movement, providing actionable insights into sleep quality and associated disorders. This study introduces an RFID-embedded mattress designed to detect and analyze sleep disorders in athletes, focusing on the interplay between sleep posture and disorders such as sleep apnea and insomnia. By integrating advanced RFID technologies with machine learning algorithms, this system enables precise classification of postures and movements, offering a non-intrusive, cost-effective alternative to traditional polysomnography. Furthermore, the relevance of posture in sleep disorders and the potential applications of this system in sports psychology and health optimization are emphasized.

Background

Sleep posture plays a critical role in managing sleep disorders, such as sleep apnoea, as supine sleeping positions often worsen the condition. Smart beds monitor and adjust sleep posture to mitigate apnoea symptoms using advanced technologies, such as spatiotemporal convolutional neural networks (S3CNN). Athletes benefit from this by maintaining optimal breathing patterns and reducing disruptions during rest¹⁹. Another method involves the use of pressure sensors strategically placed in the mattresses. By concentrating sensors in key areas, such as the shoulders and chest, this approach reduces costs and computational demands, while maintaining effective posture classification^20,21. Fiber Bragg grating sensors integrated into smart mattresses have been used to classify sleep postures. Machine learning models such as SVM, DT, LSTM, and 2D CNN have been employed to achieve high accuracy in distinguishing between different postures, such as axial versus lateral²². FMCW technology addresses the limitations of camera and contact sensor methods by using radar to estimate user posture without the need for new user calibration data, thereby achieving high generalisation and accuracy²³. By monitoring changes in temperature and humidity, sensors attached to pillows can effectively determine sleep postures. This method, combined with CNN-based modelling, has shown promising results for posture discrimination²⁴. Wearable devices measuring wrist kinematics provide an interpretable framework for detecting postural changes and inactivity, offering reliable home-based sleep analysis²⁵.

Although these advancements in sleep posture sensing offer promising solutions, challenges remain in terms of deployment, user comfort, and data privacy. Non-contact methods, such as radar and temperature sensors, provide non-invasive alternatives, but they may face limitations in specific environments or require complex data processing. Conversely, contact- based methods, although potentially more accurate, can affect user comfort and increase costs. Balancing these factors is crucial for developing effective and user-friendly sleep-posture monitoring systems. This study explored the use of radio frequency identification (RFID) sensors embedded in a smart mattress to detect sleep disorders in athletes. The system is designed to provide a passive, non-invasive method for monitoring sleep posture and movement, which are critical indicators of sleep quality. Previous studies have highlighted the role of sleep posture in modulating breathing patterns and musculoskeletal stress; however, few have utilized RFID-based systems for this purpose. Addressing this challenge, this study explored the use of Radio Frequency Identification (RFID) sensors embedded in a smart mattress to detect sleep disorders in athletes.

RFID technologies are widely used for tracking objects and sensing systems^26–30. They operate across various frequency bands, including low-frequency (LF), high-frequency (HF), very-high-frequency (VHF), ultrahigh-frequency (UHF), and super-high-frequency (SHF). An RFID system consists of an RFID reader and tags, each containing an antenna, modu- lator/demodulator, and an integrated circuit with a microcontroller. RFID tags can be either passive (battery-free) or active (powered by a battery), and each carries a unique identification number known as an electronic product code (EPC). RFID tags communicate with readers via backscattered signals, thereby enabling seamless tracking and monitoring. The RFID reader emits interrogation RF signals that energise the tags upon receipt. Passive RFID tags, which lack an internal power supply, rely on energy from these signals for activation. In contrast, active RFID tags, which have their own power sources, are activated when they detect a signal from an RFID reader. Once activated, each RFID tag transmits specific data stored within it to the reader by generating backscattered radio waves encoded with information, including the tag’s unique ID and additional data related to its status or type²⁸.

Similar to other RF systems, the response of an RFID system includes variations in the amplitude, frequency, and phase. These variations depend on factors such as propagation medium, distance, frequency, antenna orientation, tag location, interference, and object composition²⁸. Such variations enable RFID systems to be effectively utilised as wireless sensors, making them valuable for real-time tracking, monitoring, and data collection in various applications. RFID technology uses electromagnetic fields to automatically identify, and track tags attached to objects that, in this context, are embedded in the mattress. RFID’s capability of RFID to provide non-contact data collection makes it an ideal candidate for continuous health monitoring, particularly during sleep.

RFID-based sleep posture recognition is an innovative approach that leverages radio-frequency identification technology to monitor and analyse sleep postures. This method offers a noninvasive, privacy-preserving, and cost-effective solution for sleep monitoring, which is crucial for diagnosing sleep disorders and improving sleep quality. The integration of RFID with other technologies, such as AI and IoT, enhances the accuracy and functionality of sleep-posture recognition systems. In addition, RFID-based systems offer several advantages over traditional methods, such as camera-based or contact sensor systems, which can be intrusive or affect the user comfort. For example, radar-based systems provide a noninvasive alternative but require calibration for new users, whereas RFID systems can be seamlessly integrated into existing environments²³. Pres- sure detection systems, such as those using air mattresses or force sensing resistors, provide high accuracy in sleep posture recognition, but may involve higher production costs and complexity compared to RFID systems^19,31. These advantages have increased the use of RFID technology.

In the literature, RFID tags were embedded into bedsheets to create a contactless sleep monitoring system³². This setup is part of a low-cost and low-power microsystem that uses random forest classification to recognize sleep postures. The data are then uploaded to a server for analysis and can be used for sleep self-management at home or in medical settings³². Another study highlighted the use of RFID for both sleep-posture recognition and body-movement detection. By embedding RFID tags into a bed cloth, the system applies a CNN algorithm to identify sleep postures and a k-means algorithm to detect movements, thereby demonstrating the versatility of RFID in sleep monitoring³³.Thus, integrating RFID with machine learning provides enhanced accuracy in detecting patterns related to sleep posture and disorders. Prior research demonstrates the utility of RFID for body movement detection, but its application in athlete-specific sleep monitoring remains underexplored. By embedding passive RFID tags in a mattress, this study builds upon existing methodologies to deliver actionable insights into athletic recovery.

In addition, a novel design approach was employed in this study, incorporating a conductive layer that acts as a protective barrier, shielding users from unnecessary electromagnetic exposure by reflecting or absorbing radiation. By addressing this gap in existing research, the development of a smart mattress enables continuous, non-invasive monitoring with potential applications in sports psychology and health optimization.

Methods

Smart RFID-embedded mattress design

A mattress (152 mm × 196 mm) embedded with passive RFID sensors (7 × 49 = 343) was developed to detect and track an athlete’s sleep posture. These sensors were positioned as a matrix to monitor pressure points and movements throughout the sleep cycle. Data collected from the RFID sensors are processed using machine learning algorithms to classify different postures and detect irregularities in sleep behaviour. Figures 1 and 2 illustrates the test setup and mattress layers.

Fig. 1.

RFID-embedded mattress for the sleep behavior test setup.

Fig. 2.

RFID-embedded mattress layers (side view).

The multi-layered design of this smart system represents an advanced integration of comfort, sensing, and safety, combining human-centric engineering with state-of-the-art technology. At its core, the mattress layer is designed with the user’s comfort in mind, acting as the primary interface for physical interaction. It ensures the even distribution of body pressure, offering support while also protecting the more sensitive layers beneath it from excessive strain. Directly below this is the elastic and conductive layer, a multifunctional component that serves two critical purposes. On one hand, it acts as a dynamic sensor by reflecting electromagnetic waves within the first Fresnel zone. Changes in pressure—such as those caused by body movement or shifts in weight—alter the distance between this layer and the sensor grid, modulating the amplitude of the signal. This precise interaction allows the system to detect subtle variations in pressure and deformation. On the other hand, the conductive layer serves as a protective barrier, shielding the user from unnecessary electromagnetic exposure by reflecting or absorbing radiation. This dual role not only enhances sensing accuracy but also prioritizes user safety, a particularly important consideration for applications involving long-term physical contact, such as sleep monitoring or health diagnostics.

The sensor-grid layer, located beneath the conductive material, operates as the heart of the sensing system. It captures detailed data on pressure distribution and movement, translating these mechanical inputs into digital signals that can be analyzed for insights into posture, motion, or even overall health. Its integration with the conductive layer ensures that the system takes advantage of electromagnetic wave interactions to maximize accuracy. The furniture frame, or the nonconductive base layer, provides essential structural stability while isolating the sensing components from external electromagnetic interference. This layer ensures that the readings remain precise and unaffected by environmental factors, such as metallic objects or other conductive materials in the vicinity. Finally, the RFID reader, which underpins the entire system, collects data wirelessly and processes it in real time. This non-invasive and seamless approach to data collection eliminates the need for intrusive wiring, offering a practical and scalable solution for remote monitoring.

This design exemplifies the seamless blending of technology and user-centered design. Each layer has been carefully crafted to fulfill a specific purpose, yet all work in harmony to create a system that is both functional and safe. The uppermost layers focus on human comfort and safety, while the underlying components deliver precision and reliability in data collection. Together, they provide a comprehensive solution for applications such as sleep tracking, health diagnostics, and ergonomic assessments. The system not only ensures that the user’s experience is comfortable and non-intrusive but also addresses critical safety concerns by minimizing electromagnetic exposure. However, the success of this design depends heavily on meticulous material selection, precise calibration, and thoughtful consideration of environmental factors to ensure its long-term accuracy and reliability. This layered approach serves as a compelling example of how advanced technologies can be humanized to serve both functional and well-being needs in real-world applications.

Data collection and processing

An RFID reader antenna placed strategically under the bed continuously collected data during sleep. Data regarding movement, posture shifts, and pressure distribution were analyzed to determine patterns indicative of sleep disorders. The system aims to detect anomalies such as frequent tossing and turning, which are indicative of restless sleep, or prolonged periods in specific postures that may suggest discomfort or sleep apnoea. To achieve this, the system follows a structured monitoring process (Fig. 3): it begins by checking if someone is present on the bed. If no one is detected, the system loops back to continuously check for presence. Once someone is detected, a 10-second timer is initiated before data recording begins. During the recording phase, the system monitors for any movement. If movement is detected, recording stops, and the system transitions to checking if the person has left the bed. If the person has left, a counter is incremented to track the number of times the bed is entered and exited. After this, the system resets to start the monitoring process again. This structured approach ensures accurate tracking of sleep patterns and entry/exit behaviors, providing valuable insights for identifying potential sleep disorders.

Fig. 3.

Flow chart for data collection procedure for posture recognizing.

The procedural methodology used in this study consisted of several stages. The first stage involves the acquisition of radio-frequency identification (RFID) data, where the amplitude and phase information are collected from RFID sensors embedded in a mattress under carefully controlled environmental conditions. Next, the data were exported and organised into Microsoft Excel or CSV format to ensure that they were accessible and structured for further analysis. The third stage involves parsing the data using Visual Basic (VB) scripts within Excel, specifically focusing on the sensor identifier (EPC), amplitude, and phase data, which is crucial for the accurate interpretation of RFID information. An interface program was then employed to integrate additional sample characteristics, such as weight, height, age, and health conditions, with the parsed data. Once the data is processed, it is categorized into two primary types: ’Normal (Without Load)’ and ’Various Human and Human Postures.’ In the final stage, machine learning algorithms in MATLAB were applied to the categorised data, enabling the system to recognise and differentiate between various human postures. In this study, data from five volunteers with varying height, weight, age, and gender were analyzed across two different postures (supine and prone).

For this project, a monostatic RFID reader (ST25RU3993) with a single-polarisation antenna was used, both of which have been utilised in previous studies. The RFID reader antenna, featuring a 6 dBi directivity gain, was installed beneath the mattress. Passive RFID sensors, specifically 40 × 40 mm long-range ISO18000-6 C H47 inlay labels, were selected. Initially, for a queen-sized mattress measuring 150 × 200 cm, 343 passive RFID sensors arranged in a 7 × 49 grid were planned. The RFID reader antenna was placed underneath the sofa for optimal data collection.

During each measurement, approximately 530 data points were captured from 343 individual sensors, resulting in nearly 1,200,000 data points. The reader has a read rate of 700 tags per second, and each measurement lasted for 10 s. The measurements captured the x and y coordinates of the sensors, average amplitude and phase levels, standard deviation, and maximum and lowest levels for each sensor. A critical step in data processing was conducted prior to the implementation of the machine learning algorithms. This included standardising the data and applying a Gaussian filter at the 5% level for smoothing²⁸. This preliminary phase is essential to guarantee the dependability and precision of the ensuing algorithmic evaluations.

Machine learning for sleep disorder detection

In this study, a total of five individuals participated, with two specific postures—supine (lying on the back) and prone (lying on the stomach)—analyzed for human recognition and posture identification. To achieve this, five datasets were created for individual recognition, capturing unique pressure distribution patterns across the mattress for each participant. Additionally, two datasets were specifically developed for posture recognition, focusing on the distinct Z-level pressure variations associated with the supine and prone postures. Given that the sensor coordinates (X and Y) were fixed and the Z-level varied as a function of distance due to applied pressure, regression algorithms were identified as the most suitable method for analyzing the data. The changes in the Z-level effectively represented the pressure distribution across the mattress, serving as a crucial metric for estimating the sleeper’s posture and movement. This data-driven approach enabled a detailed understanding of how pressure variations corresponded to positional shifts and movements during sleep. To detect sleep disorders and recognize postures, two machine learning approaches were employed. The first, Gaussian process regression (GPR), is a non-parametric Bayesian algorithm renowned for its ability to model complex, non-linear relationships. GPR’s probabilistic framework provided nuanced insights into subtle variations in pressure data, making it particularly effective for identifying sleep disturbances such as fragmentation or irregular breathing patterns. The model’s capacity to quantify uncertainty further enhanced its applicability in analyzing dynamic sleep patterns. The second approach, Linear Regression (LR), is a straightforward yet computationally efficient algorithm used to model linear relationships within the data. This method provided a solid baseline for posture recognition, enabling the rapid processing of pressure-based features with minimal computational overhead. Together, these algorithms facilitated a comprehensive analysis of sleep quality and posture recognition, enabling the identification of critical issues such as sleep fragmentation, disturbed breathing, and distinct postures.

The system was implemented using MATLAB, a robust computational platform ideal for processing complex datasets and executing advanced regression algorithms. The implementation included critical stages of data preprocessing, model training, and performance evaluation. To ensure the reliability of the results, a 5-fold cross-validation technique was applied, where the dataset was divided into five equal parts. During each iteration, four parts were used for training, and the remaining part served as the test set. For each fold, 20% of the data was used as the test set, while the remaining 80% was used for training. This process was repeated five times, ensuring that each data point was tested exactly once, providing a robust evaluation of the model’s generalization ability. Performance metrics, specifically the Root mean square error (RMSE), were calculated for each fold and averaged to assess overall model accuracy. Table 1 shows the summary of the algorithms used for the system.

Table 1.

Algorithms used for the system.

Function	Dataset number	Test data (%)	Training data (%)	Model	Performance metric
Human	5	20	80	Gaussian process regression (GPR)	RMSE
Human/sleep posture	2	20	80	Linear regression (LR)	RMSE

Ethical considerations

This study was conducted in accordance with the ethical guidelines of the Declaration of Helsinki. All participants were informed about the study’s purpose, procedures, and their right to withdraw at any time without consequences. Written (or verbal) informed consent was obtained from all participants prior to their participation. As the study did not involve sensitive personal data, medical interventions, or vulnerable populations, institutional ethical approval was not deemed necessary in accordance with the policies of Mersin University.

Results

The RFID-based smart mattress demonstrated its ability to accurately detect sleep postures and movements, providing valuable insights into athletes’ sleep quality. The performance of the machine learning algorithms used for these detections was evaluated using the root mean square error (RMSE), a widely recognized metric in predictive modeling. RMSE quantifies the standard deviation of prediction errors by measuring the square root of the average squared differences between predicted and actual values³⁴. As a key indicator of model accuracy, RMSE provides a clear sense of how well the algorithm performs, with lower values indicating closer alignment between predictions and observations. This metric is particularly valuable in sleep monitoring systems, as it enables the objective assessment of detection reliability, ensuring the system’s outputs are both actionable and precise. By leveraging this evaluation framework, the smart mattress ensures its utility in real-world applications, including athlete-specific recovery and performance optimization.

Key findings include

Sleep posture identification achieved an RMSE of 0.42165, meaning the predicted postures closely aligned with the actual positions, with only minimal deviations. This level of accuracy is critical for identifying problematic sleep postures that could impact recovery. Classification of sleep-related movements showed a validation RMSE of 0.14501, demonstrating even higher precision in detecting movements associated with sleep disorders, such as restless legs or frequent awakenings. The system effectively detected potential sleep disorders by analysing movement frequency and postural changes, with RMSE helping to quantify how well the system predicted these deviations from normal sleep behaviour.

The use of RMSE in these results (Table 2) highlights the system’s ability to make reliable predictions, with lower RMSE values confirming the effectiveness of the smart mattress in enhancing sleep monitoring for athletes.

Table 2.

Machine learning algorithms prediction rates for sleep disorders in the study.

Function	Model	RMSE
Human	Gaussian process regression (GPR)	0.14501
Human/sleep posture	Linear regression (LR)	0.42165

Discussion

The integration of RFID technology into a smart mattress offers a nonintrusive method for monitoring athletes’ sleep behaviour. By detecting sleep disorders early, sports psychologists can intervene to optimize an athlete’s recovery process and improve their overall performance. This RFID-based system represents a cost-effective and practical solution compared to traditional sleep studies, which often require uncomfortable monitoring devices. Although RFID-based systems offer numerous benefits, they also face challenges such as signal interference and the need for precise tag placement to ensure accurate posture detection. The development of robust algorithms and models is crucial for improving the accuracy and reliability of RFID-based sleep-posture recognition systems. Techniques such as data enhancement and multitask learning can enhance the model performance and generalization.

Sleep posture plays a critical role in the onset and progression of sleep disorders. For example, supine sleeping positions are known to exacerbate conditions like sleep apnea by increasing airway resistance, leading to disrupted breathing patterns. By detecting and analyzing such postural trends, the RFID-embedded mattress can provide critical insights into these issues and inform targeted interventions.

Frequent postural changes or restless movements during sleep, identified through the system, may indicate disorders such as insomnia or restless legs syndrome. These patterns are directly linked to fragmented sleep and reduced recovery, which are detrimental to athletic performance. By monitoring these movement patterns, the system can offer valuable insights into potential disruptions in athletes’ recovery processes and overall health, further highlighting its utility in sports psychology and health optimization.

Conclusion

This study demonstrates that the RFID-embedded mattress offers a multifaceted and innovative approach to sleep monitoring, combining advanced technology with machine learning to significantly improve the understanding and enhancement of sleep quality. By continuously tracking movements, posture shifts, and pressure distribution, the system generates a comprehensive sleep profile. It identifies different sleep stages—light, deep, or REM sleep—while analyzing sleep duration, quality, and efficiency. Additionally, it detects abnormalities such as fragmented sleep caused by frequent movements and disorders like restlessness due to stress or discomfort, as well as sleep apnea, indicated by prolonged periods in specific postures. By pinpointing pressure points and discomfort areas, it also helps identify issues such as joint pain or mattress-related problems.

The findings highlight the system’s capability to identify clinically relevant postural patterns associated with sleep disorders. For instance, extended periods in a supine position may indicate sleep apnea, while frequent posture changes are linked to restlessness and insomnia. These insights underline the system’s potential for early detection and management of sleep disorders, offering significant benefits for clinical applications.

In this study, data from five individuals with varying heights, weights, ages, and genders were analyzed across two different postures. The results showcase the system’s effectiveness in monitoring and recognizing body movements and postures. Future research will expand on these findings by incorporating biometric data and conducting studies in hospital settings to validate and enhance diagnostic capabilities.

Beyond sleep pattern analysis, the system provides valuable insights into health by monitoring behaviors such as bed entry and exit. These metrics can reveal conditions like nocturia, insomnia, or other health concerns. Over time, tracking these patterns enables the evaluation of interventions, such as medications or sleep therapies. By integrating sleep analysis with movement tracking, the system offers a holistic view of physical and mental well-being, identifying trends that could signal chronic sleep issues or early symptoms of disorders like restless leg syndrome.

The advanced features of the system enhance usability and user experience. RFID sensors wirelessly transmit data to a central processing unit for real-time analysis and remote monitoring. Integration with IoT platforms allows users and healthcare professionals to access data via smartphones or computers. Algorithms analyze movement and posture data to detect anomalies indicative of potential sleep disorders, triggering alerts when abnormalities are identified. Customizable thresholds ensure adaptability to individual sleep habits and conditions, such as sleep apnea or insomnia. Unlike traditional methods that rely on intrusive wearable sensors, this RFID-based system provides a non-intrusive, contactless solution, ensuring maximum comfort during sleep.

Future advancements in the system’s capabilities could be achieved through the integration of additional biometric sensors and advanced machine learning algorithms. These enhancements would improve diagnostic accuracy, facilitate more precise detection of sleep stages, and enable long-term trend analysis for the early identification of potential health complications. Integration with smart home technologies could further allow the system to communicate with devices such as smart lighting and climate control, optimizing the sleeping environment. Additionally, future developments may incorporate the monitoring of health metrics such as heart rate and breathing patterns, providing a more comprehensive evaluation of overall health.

In conclusion, the RFID-embedded mattress represents a breakthrough in sleep monitoring technology, combining RFID sensors with machine learning to deliver personalized and effective health solutions. Its transformative applications in sports psychology and healthcare hold great promise for improving sleep quality and overall well-being.

Author contributions

M.P wrote the main manuscript, All authors reviewed the manuscript.

Data availability

Data that support the findings of this study are available upon request by contacting the corresponding author. These data are not publicly available due to privacy or ethical restrictions.

Declarations

Competing interests

The authors declare no competing interests.