Study of Wash-Induced Performance Variability in Embroidered Antenna Sensors for Physiological Monitoring

Abstract

This paper presents a study on the repeatability of washing effects on two antenna-based sensors for breathing monitoring. One sensor is an embroidered meander antenna-based sensor integrated into a T-shirt, and the other is a loop antenna integrated into a belt. Both sensors were subjected to five washing cycles, and their performance was assessed after each wash. The embroidered meander antenna was specifically compared before and after washing to monitor a male volunteer’s different breathing patterns, that is, eupnea, apnea, hypopnea, and hyperpnea. Stretching tests were also conducted to evaluate the impact of mechanical deformation on sensor behavior. The results highlight the changes in sensor performance across multiple washes and stretching conditions, offering insights into the durability and reliability of these embroidered and loop antennas for practical applications in wearable health monitoring. The findings emphasize the importance of considering both washing and mechanical stress in the design of robust antenna-based sensors.

1. Introduction

Electronic textiles (E-textiles) are increasingly becoming a part of our daily lives, transforming ordinary clothing into smart, functional garments that enhance convenience, safety, and efficiency [1]. By embedding antennas, sensors, and communication devices into fabrics, e-textiles enable real-time health monitoring, fitness tracking, and even environmental sensing, offering significant benefits for personal well-being and productivity [2]. This integration not only improves individual quality of life but also opens new opportunities for innovation across industries. E-textiles find applications in various fields, including healthcare, sports, defense, and environmental monitoring, where their adaptability and functionality meet diverse needs [3]. In healthcare, e-textiles can facilitate the early detection of medical conditions and enable continuous monitoring of patients without the need for bulky equipment, enhancing both comfort and accessibility. Moreover, they enable the seamless collection and transmission of vital data, making healthcare more proactive and personalized [4]. Their potential also extends to defense and security, where they can enhance the safety of police officers and firefighters by monitoring vital signs and environmental conditions. E-textiles’ ability to combine flexibility, wearability, and advanced functionality positions them as a foundational element of the wearable technology revolution, driving progress in connected living and smart environments [5].

In recent years, the demand for wearable technology has led to significant advancements in the development of flexible textile antennas, particularly antenna-based sensors. These sensors integrate seamlessly into garments, enabling applications in healthcare, fitness tracking, and security operations [6]. Wearable systems with antenna-based sensors enable real-time monitoring and wireless communication with nearby devices, delivering essential data for safety and performance enhancement. Textile antenna-based sensors are notable for their unique ability to combine advanced functionality with user comfort. Typically, textile antennas are constructed using conductive materials like conductive-coated fabrics, embroidered conductive threads, or conductive inks [7]. The integration of antenna-based sensors into e-textiles has also advanced the market of wearable technology by acting as an intersection between advanced communication systems and user convenience. E-textiles embedded with antenna functionality allow clothing to serve not only as passive clothing but also as a device capable of collecting and transmitting data. This dual-purpose capability has expanded the possibilities for wearable technology, making e-textiles an essential component in the Internet of Things (IoT) [8].

The exploration of e-textiles in antenna sensors also highlights the critical aspect of washability, which is essential for their integration into daily life. Since these garments are designed to function like traditional clothing, their ability to withstand regular washing cycles while maintaining performance is crucial [9]. Washing can present significant challenges, such as degradation of conductive materials, loss of adhesion, and changes in the mechanical or electrical properties of the antenna. This makes it necessary to design antenna sensors that are flexible and resistant to repeated laundering [10]. As such, the repeatability of washing effects on antenna performance becomes a crucial factor in determining their long-term viability for wearable applications. The impact of washing on the integrity of the electrical and mechanical properties of the antenna can influence both the accuracy of the collected data and the overall reliability of the e-textile system [11].

In recent years, the development of embroidered antennas integrated into wearable textiles for health monitoring applications has attracted considerable attention. While many studies have explored the design, performance, and mechanical properties of textile-based antennas, there is a significant gap in the literature regarding the long-term wash durability of embroidered antennas, particularly in terms of their resonance frequency after repeated washing cycles [12]. Existing research tends to focus on the initial performance and material properties of wearable antennas, yet the impact of washing cycles on antenna durability remains poorly understood. Although some studies have investigated the effects of environmental factors and mechanical stress on antennas, there is little emphasis on the degradation of embroidered antennas under repeated laundering, a common condition for wearable systems. For example, research on washability in e-textiles e.g., Refs. [13,14] typically addresses more general fabric-based sensors or non-embroidered antennas, leaving a gap in understanding how embroidered conductive threads perform under washing and stretching forces. Additionally, most literature lacks detailed insights into how repeated washing alters signal integrity and long-term stability of antenna-based sensors. This study aims to fill this gap by providing a comprehensive evaluation of the wash durability of embroidered meander dipole and loop antenna-based sensors, focusing on how multiple washing cycles impact their performance, including resonance frequency shifts and mechanical degradation.

In this context, it is important to assess how different washing cycles impact the functionality of embroidered antenna-based sensors, especially in terms of signal integrity and durability. These antenna sensors, integrated into wearable garments such as T-shirts and belts, are exposed to a range of environmental factors that could degrade their performance over time. This paper aims to investigate the effects of multiple washing cycles on the performance of these antenna sensors, providing insights into their reliability and potential for real-world applications in monitoring health. Specifically, we will examine how washing affects key parameters such as resonance frequency shift and overall sensor stability, which are critical for accurate and continuous monitoring of physiological signals like breathing patterns. These antenna sensors must be washed following established ISO norms to ensure that they undergo standardized testing for washability and performance, simulating real-life use cases in which the garments will be subjected to regular washing.

2. Materials and Methods

In this work, the main figure of merit used to assess the influence of washing cycles and mechanical deformation on antenna performance is the resonance frequency shift of the reflection coefficient. The reflection coefficient was systematically measured using a Vector Network Analyzer after each washing cycle to track any performance degradation and to evaluate the repeatability and stability of the antenna sensors under realistic conditions. In order to assess the stability of the antenna reflection coefficient after multiple washing cycles, an experiment was carried out using two different antenna-based sensor designs and materials. The fabrication process of these two antennas is shown in Figure 1. The first was a loop antenna-based sensor, recognized for its adaptability, consistent performance under various conditions, and its ability to maintain stable electromagnetic properties even under mechanical stress. The second was a meander dipole antenna, selected for its robustness and broad range of applications, serving as a baseline for performance evaluation [15]. Its compact design makes it a suitable candidate for wearable health monitoring systems. The sensors were subjected to multiple washing cycles to simulate real-life usage, assessing the impact of washing on key performance parameters, such as the reflection coefficient, which is crucial for signal integrity and accurate monitoring. The study aimed to determine whether the reflection coefficients $\left|S_{11}\right|$ of these antenna designs remained stable after repeated exposure to washing and mechanical stress, primarily in terms of resonance frequency, which serves as the key indicator of antenna performance degradation or stability under such conditions.

The materials used for the two antennas were carefully selected to ensure flexibility, durability, and suitability for wearable applications. The embroidered antennas were fabricated using a Singer Futura XL550 embroidery machine, employing 99% silver-coated nylon thread (Shieldex 117/17 dtex 2-ply). This machine offers a stitching resolution on the order of 1 mm, ensuring consistent fabrication accuracy and repeatability for the embroidered conductive patterns. The T-shirt substrate offered a relative dielectric constant of 1.3 and a loss tangent of 0.0058, ensuring seamless integration and comfort for wearable use. The elastic belt dielectric constant and loss tangent are ${\epsilon }_r=1.61$ and $tan\delta =0.0089$, respectively, with a thickness of 1.6 mm, serving as the substrate, providing the necessary flexibility and durability for real-time monitoring applications. These material choices, combined with the precision of the embroidery process, ensured optimal conductivity, comfort, and adaptability for wearable health monitoring systems. Both antenna sensor designs were embroidered using a weave pattern with a stitch spacing of 0.40 mm and a stitch length of 4 mm. The thread tension was adjusted to 50% of the maximum tension to ensure optimal thread tension, maintaining consistent thread placement and conductivity during the embroidery process. Before embroidery, the cotton T-shirt and elastic belt substrates were carefully prepared to ensure uniformity and repeatability in the experiments. The substrates were pre-treated by performing a gentle washing cycle to remove any contaminants or residues that could affect the conductivity of the conductive threads and the adhesion of the embroidery. The samples were securely clamped at both ends to ensure uniform stretching. The strain rate was manually controlled, with stretching values ranging from 2 to 8 mm to simulate realistic conditions that wearable textiles might experience during typical daily activities.

3. Washing Protocols for Antenna Durability Testing

To evaluate the durability and performance stability of antennas under repeated washing conditions, a standardized washing protocol was applied using carefully controlled parameters. This protocol was developed to replicate real-world washing scenarios while adhering to ISO standard norms to ensure consistency and reliability in the testing process. The washing parameters, as detailed in

Table 1. Washing parameters and settings for antenna durability testing.

Parameter	Setting
Washing temperature	40 °C
Washing duration	40 min
Number of towels washed at the same time	6
Washing detergent	ECE77 P/ENSAYO
Spin speed	1400 RPM

, included specific settings for temperature, duration, detergent, and spin duration to simulate typical household washing conditions. The washing process adhered to the UNE-EN ISO 6330:2021 [16] standard, using a washing machine and a neutral ECE-color detergent ISO 105-C06 ECE77 soap (Barcelona, Spain) [17]. The washing process was carried out using a washing machine (7000 ProSteam^® AEG, Electrodomesticos, Berlin, Germany); this machine performed both the washing and drying procedures, eliminating the need for manual air-drying. The antennas were then visually inspected for any signs of physical damage, such as cracking, fraying, or detachment of the conductive elements from the substrate. The effect of washing cycles on the electrical properties was assessed by measuring the resonance frequency using a vector network analyzer (VNA) before and after washing the samples. These measurements allowed for the evaluation of the antenna’s functionality and sensitivity post-wash, providing insights into their durability and suitability for prolonged use in wearable applications. This comprehensive washing protocol ensured that the antennas’ performance could be reliably tested under realistic and standardized conditions, facilitating a robust assessment of their material integrity and electrical stability after exposure to repeated washing cycles.

4. Results

Characterization of Embroidered Antennas Under Washing and Stretching Conditions

The impact of washing cycles on the performance of the embroidered antenna sensors was assessed by analyzing their reflection coefficient ($S_{11}$). The results provide insight into how repeated washing affects the antenna’s resonance frequency, which is a critical parameter for its functionality in sensing applications. By examining the evolution of the resonance frequency after successive washing cycles, we can evaluate how these effects impact the reliability and performance of the sensor over time. Understanding these changes is essential to assess the repeatability and long-term stability of embroidered antenna-based sensors, particularly for wearable applications where exposure to washing is inevitable. To illustrate these effects, Figure 2 presents the measured $S_{11}$ of the embroidered meander dipole antenna sensor before and after multiple washing cycles. The comparison highlights variations in the resonance frequency, which may be attributed to mechanical deformations or material degradation. As observed in the figure, the resonance frequency of the antenna shifts progressively toward lower frequencies with an increasing number of washes. This shift suggests variations in the antenna’s electrical properties, which may be due to changes in the integrity of the conductive threads, fabric deformation, or alterations in dielectric properties caused by detergent absorption. After the fourth wash, the resonance frequency stabilizes, indicating that the sensor performance has reached a plateau. However, while the resonance frequency stabilizes after the fifth wash cycle, the amplitude of $S_{11}$ worsens, indicating an increase in return loss. This behavior suggests that repeated washing cycles have affected the impedance matching of the antenna, leading to a decline in performance. The increased return loss could be attributed to factors such as the degradation of the conductive threads, an increase in contact resistance, and mechanical deformation resulting from the washing process. These variations provide valuable insights into the extent of washing-induced modifications, highlighting that even though the resonance frequency stabilizes, the antenna’s overall performance deteriorates due to the combined effects of these factors.

In addition to mechanical stress and repeated washing, potential confounding factors, such as the absorption of detergent during washing, could also affect the dielectric properties of the fabric. Detergent residues might alter the dielectric constant and loss tangent, influencing the electrical performance of the embroidered antennas. While this study primarily focuses on the effects of mechanical deformation and wash cycles, it is important to consider that detergent absorption may contribute to some degree of variability in the observed performance. Consequently, no additional washing cycles were conducted, as further testing was deemed unnecessary given the consistent behavior observed after the fifth wash. These findings underscore the implications of long-term washing on the functionality of the antenna-based sensor. As seen in Figure 2, as the number of wash cycles increases, the Q-factor of the antenna decreases. Initially, for the no-wash condition, the antenna exhibits the sharpest resonance and the highest Q-factor of 7.6, meaning it operates efficiently. However, with each wash cycle, the resonance becomes less sharp, and the Q-factor drops. This decrease is due to the degradation of the conductive material and mechanical deformation caused by washing. As a result, the bandwidth increases, leading to a lower Q-factor. After five washes, the antenna shows the lowest Q-factor of 7.2, reflecting the greatest degradation in performance. The resonance frequency variation under different stretching levels after repeated washing cycles, with linear regressions and error bars indicating measurement variability at each stage, is shown in Figure 3. The results demonstrate a consistent decrease in resonance frequency as the stretching increases, following a linear trend across all washing conditions. However, successive washing cycles induce a progressive downward shift in the resonance frequency, with the most significant degradation observed after five washes. This suggests alterations in the antenna’s electrical properties, possibly due to degradation of the conductive thread or increased resistivity from washing residues. The sensitivity of the meander dipole antenna, defined as the slope of the linear regression of resonance frequency versus stretching, was analyzed for each washing cycle. In the unwashed condition, the antenna showed a sensitivity of −41.5 MHz/mm, which remained relatively stable after the first two washing cycles, with values of −41.0 MHz/mm and −40.0 MHz/mm, respectively. A slight reduction to −39.5 MHz/mm was observed after the third wash, suggesting the onset of minor changes in the antenna’s physical or electrical structure. More pronounced degradation appeared after the fourth and fifth washes, where the sensitivity dropped to −31.0 MHz/mm and −27.5 MHz/mm, respectively. This trend reflects a gradual loss of sensitivity with repeated washing, potentially due to increased resistive losses or deformation in the embroidered conductive paths. Each data point is accompanied by an error bar representing the variability or uncertainty in the measured resonance frequency at a given stretch level. Notably, the error bars tend to increase in length as the number of washing cycles progresses, particularly for the 4 mm and 8 mm stretching cases. This indicates a growing inconsistency in the antenna’s response, which can be attributed to progressive degradation of the conductive threads, fabric wear, or changes in dielectric properties after multiple washes. Thus, the error bars provide important insight into the stability of the antenna’s behavior, highlighting the impact of cumulative washing on sensor performance variability. Linear regression analysis confirms a strong correlation between stretching and frequency for up to four washes ($R^2>0.98$), indicating repeatable behavior. However, after five washes, the correlation weakens ($R^2=0.7460$), suggesting increased variability and reduced performance stability. These findings highlight the necessity of assessing embroidered antennas under realistic washing conditions to ensure their reliability in wearable applications.

Figure 4 presents the reflection coefficient ($S_{11}$) as a function of frequency for the loop antenna-based sensor before and after multiple washing cycles. The unwashed sample (0 cycles) exhibits a resonance frequency of around 2.4 GHz, with a deep reflection coefficient, indicating strong impedance matching. As the number of washing cycles increases, a downward frequency shift was observed, with the resonance frequency decreasing to approximately 2.32 GHz after five washing cycles. As with the previous plot, after the fourth wash, the antenna performance stabilizes with no further significant degradation. Continuing with more washing cycles beyond the fifth wash does not result in additional changes, making further washing unnecessary. Furthermore, the minimum $S_{11}$ value becomes less deep, indicating increased losses and potential deterioration in antenna efficiency. This shift suggests that repeated washing affects the antenna’s electrical performance, likely due to changes in thread conductivity or mechanical deformation. These findings highlight the importance of considering washing effects when designing embroidered antennas for long-term wearable applications.

Figure 5 presents the variation of the resonance frequency as a function of stretching for the embroidered loop antenna before and after multiple washing cycles, with linear regression lines and error bars illustrating measurement variability at each wash. As expected, increasing the stretching results in a downward frequency shift, demonstrating the mechanical deformation’s impact on the antenna’s electrical behavior. The sensitivity of the loop antenna, defined as the slope of the linear regression of the resonance frequency versus stretching, was evaluated across all washing conditions. In the unwashed state (0 cycles), the antenna exhibited a sensitivity of −44.3 MHz/mm, a value that remained consistent after the first three washing cycles, indicating high repeatability and structural stability. After the fourth wash, a slight increase in sensitivity to −44.9 MHz/mm was observed, suggesting minor alterations in the antenna’s material or electrical characteristics. Following the fifth wash, the sensitivity further increased to −47.4 MHz/mm, reflecting a more pronounced shift in the resonance frequency with mechanical deformation. These findings confirm that the loop antenna maintains stable performance across initial washing cycles, with gradual sensitivity changes emerging only at later stages. The pre-wash condition (0 cycles) exhibits a strong linear correlation between stretching and resonance frequency, with $R^2$ a value of 0.9922, indicating predictable behavior. After each washing cycle, the resonance frequency continues to follow a similar downward trend, but with progressively lower values across all stretching levels. After five washing cycles, a more pronounced reduction in resonance frequency was observed, with the regression slope becoming slightly steeper and an $R^2$ value of 0.9947, suggesting a minor deviation from the original linear trend. This shift implies that repeated washing introduces material degradation, potentially affecting the conductivity of the conductive threads and the integrity of the substrate. These findings highlight the importance of considering washing effects when designing embroidered loop antennas for long-term wearable applications. Compared to the previous scenario, the error bars in this figure are significantly shorter and more consistent across all washing states, including after the fifth cycle. This reduced spread suggests improved measurement stability and enhanced repeatability of the antenna performance, even after multiple washes. Notably, the error bars remain relatively uniform across the entire stretching range, with no abrupt increases at higher elongation levels. This consistency reflects a robust antenna design that maintains its electrical properties despite mechanical and environmental stresses.

5. Discussion

Although the loop antenna demonstrated better resilience to washing cycles, the meander dipole antenna was selected for the breathing monitoring experiments to evaluate sensor performance under more critical operating conditions. This choice was made deliberately to represent a worst-case scenario, where the effects of washing and mechanical stress are more pronounced. In this study, a healthy male volunteer (27 years, 185 cm, 70 kg) wore a cotton T-shirt with the integrated antenna-based sensor while performing various breathing patterns, such as eupnea, apnea, hypopnea, and hyperpnea, in a controlled laboratory setting. A vector network analyzer (VNA) was employed to capture the sensor’s response, with data transmitted in real-time to a PC host via a LAN interface for processing. The resonance frequency shifts associated with these breathing patterns were then analyzed using MATLAB 2024b allowing for an in-depth assessment of sensor behavior under different conditions. The experiment aimed to evaluate how repeated washing and mechanical deformation affected the sensor’s ability to detect breathing, providing crucial insights into the sensor’s durability and reliability for long-term wearable health monitoring applications. By understanding how these factors impact sensor performance, the study contributes to the development of more robust and reliable wearable sensors that can endure everyday challenges while maintaining accuracy and functionality. The experimental setup for real-time respiration detection using the meander dipole antenna sensor integrated into a T-shirt is illustrated in Figure 6.

Figure 7 presents the resonance frequency variations of the embroidered meander dipole antenna sensor during eupnea (normal breathing) before washing (left) and after five washing cycles (right). The recorded data highlights the cyclic frequency shifts corresponding to the inhalation and exhalation phases, demonstrating the sensor’s capability to track respiratory activity in real time. Before washing, the resonance frequency exhibits well-defined oscillations with distinct inhale peaks (red markers) and exhale troughs (green markers), indicating stable and accurate breathing detection. After five washing cycles, the oscillations become less pronounced, and the frequency shift between inhalation and exhalation is slightly reduced, suggesting a minor decrease in sensitivity. This could be attributed to variations in the electrical properties of the conductive threads, increased contact resistance, or structural modifications in the textile substrate due to washing. However, despite this attenuation, the T-shirt maintains the same overall behavior and remains capable of detecting different breathing patterns. The results confirm that the embroidered meander dipole antenna sensor continues to function effectively for respiration monitoring after multiple washes, demonstrating its robustness for long-term wearable applications.

Figure 8 illustrates the detection of hyperpnea and hypopnea breathing patterns using the embroidered meander dipole antenna sensor before washing (a) and after five washing cycles (b). The resonance frequency variations indicate distinct respiratory phases, with hyperpnea (highlighted in blue) corresponding to deeper and more frequent inhalations, and hypopnea (highlighted in red) indicating periods of shallow or reduced breathing. In the pre-wash condition (a), the antenna sensor effectively captures the transitions between normal breathing, hyperpnea, and hypopnea, demonstrating its capability for real-time respiratory pattern recognition. After five washing cycles (b), the frequency response remains consistent, with the sensor continuing to distinguish between hyperpnea and hypopnea despite a reduction in the overall resonance frequency. This result confirms that, even after multiple washes, the embroidered antenna sensor maintains its ability to detect and classify different breathing patterns, reinforcing its potential for long-term wearable respiratory monitoring applications.

For apnea breathing detection, Figure 9 presents the resonance frequency variations captured by the embroidered meander dipole antenna sensor before washing (a) and after five washing cycles (b). The figure highlights apnea episodes (shaded in red), where a temporary cessation of breathing leads to a stabilization of the resonance frequency with minimal fluctuations. In the pre-wash condition (a), the sensor effectively detects normal breathing cycles followed by a distinct apnea phase, characterized by a flat resonance frequency response. After five washing cycles (b), the antenna continues to identify the apnea event, despite a slight reduction in the overall resonance frequency. This confirms that, even after multiple washes, the embroidered antenna sensor remains capable of detecting apnea episodes.

This study demonstrates a stabilization in the performance of embroidered meander dipole antennas after five washes, which contrasts with findings from [18], who observed a significant shift in resonance frequency after washing textile antennas, especially with jeans, while materials like wool and corduroy exhibited minimal shifts. In our study, resonance frequency stabilization after five washes suggests a reduced degradation rate over time. This finding aligns with other studies, such as [19], which investigated the impact of washing on textile-based antennas for body-centric wireless communication, showing resonance frequency shifts after repeated washing, and [20], which reported similar degradation trends in wearable antennas after mechanical stretching and washing. Furthermore, studies like [21] on post-wash impedance changes in textile sensors indicate that degradation slows down after several cycles, which supports our observation of post-wash stabilization. Unlike [18], whose results showed continuous degradation, our study suggests that after a certain point, the degradation rate decreases, and the performance stabilizes. These differences highlight the influence of antenna design (meander dipole vs. microstrip) and textile substrate on long-term performance stability. Further research should investigate the combined effects of material properties, wash cycles, and antenna architectures to better understand their impact on the durability of wearable sensor systems.

6. Conclusions

This study provides a comprehensive assessment of the repeatability of washing effects on embroidered antenna-based sensors used for breathing monitoring. The results demonstrate that while the antenna performance degrades after each wash, stabilization occurs after five washes, meaning that no significant further degradation is observed in subsequent cycles. Mechanical stretching also impacts the performance of the antennas, with both sensors showing changes in their resonance characteristics. These findings underline the importance of considering both washing and mechanical stress in the design of durable and reliable antenna-based sensors for wearable health monitoring applications. The results contribute valuable insights into the longevity and practical applicability of embroidered and loop antenna sensors, guiding future developments in sensor design and performance optimization for health monitoring systems.