Passive UHF RFID-based real-time intravenous fluid level sensor.

Ultra high frequency (UHF) passive radio frequency identification (RFID) tag-based sensors are proposed for intravenous (IV) fluid level monitoring in medical Internet of Things (IoT) applications. Two versions of the sensor are proposed: a binary sensor (i.e., full vs. empty state sensing) and a real-time (i.e., continuous level) sensor. The operating principle is demonstrated using full-wave electromagnetic simulation at 910 MHz and validated with experimental results. Generalized Additive Model (GAM) and random forest algorithms are employed for each interrogation dataset. Real-time sensing is accomplished with small deviations across the models. A minimum of 72% and a maximum of 97% of cases are within a 20% error for the GAM model and 62% to 98% for the random forest model. The proposed sensor is battery-free, lightweight, low-cost, and highly reliable. The read range of the proposed sensor is 4.6 m.

Energy-Efficient Respiratory Anomaly Detection in Premature Newborn Infants.

Precise monitoring of respiratory rate in premature newborn infants is essential to initiating medical interventions as required. Wired technologies can be invasive and obtrusive to the patients. We propose a deep-learning-enabled wearable monitoring system for premature newborn infants, where respiratory cessation is predicted using signals that are collected wirelessly from a non-invasive wearable Bellypatch put on the infant's body. We propose a five-stage design pipeline involving data collection and labeling, feature scaling, deep learning model selection with hyperparameter tuning, model training and validation, and model testing and deployment. The model used is a 1-D convolutional neural network (1DCNN) architecture with one convolution layer, one pooling layer, and three fully-connected layers, achieving 97.15% classification accuracy. To address the energy limitations of wearable processing, several quantization techniques are explored, and their performance and energy consumption are analyzed for the respiratory classification task. Results demonstrate a reduction of energy footprints and model storage overhead with a considerable degradation of the classification accuracy, meaning that quantization and other model compression techniques are not the best solution for respiratory classification problem on wearable devices. To improve accuracy while reducing the energy consumption, we propose a novel spiking neural network (SNN)-based respiratory classification solution, which can be implemented on event-driven neuromorphic hardware platforms. To this end, we propose an approach to convert the analog operations of our baseline trained 1DCNN to their spiking equivalent. We perform a design-space exploration using the parameters of the converted SNN to generate inference solutions having different accuracy and energy footprints. We select a solution that achieves an accuracy of 93.33% with 18× lower energy compared to the baseline 1DCNN model. Additionally, the proposed SNN solution achieves similar accuracy as the quantized model with a 4× lower energy.

UHF RFID tag localization using pattern reconfigurable reader antenna.

Passive ultra high frequency (UHF) radio frequency identification (RFID) tags have the potential to find ubiquitous use in indoor object tracking, localization, and contact tracing. We propose a machine learning-based method for RFID indoor localization using a pattern reconfigurable UHF RFID reader antenna array. The received signal strength indicator (RSSI) values (from 10,000 tags) recorded at the reader antenna units are used as features to evaluate the machine learning models with a train-test split of 75%-25%. The training and testing data is generated by a wireless ray tracing simulator. Five machine learning models: random forest regressor, decision tree regressor, Nu support vector regressor, k nearest regressor, and kernel ridge regressor are compared. Random forest regressor has the lowest localization error both in terms of average Euclidean distance (AED) and root-mean-square error (RMSE). For random forest regressor, localization error results show that 90% of the tags are within 1 meter of their true position, and 67% are within 50 cm of their true position based on Euclidean distance.

Passive UHF RFID-based Knitted Wearable Compression Sensor.

One of the major challenges faced by passive on-body wireless Internet of Things (IoT) sensors is the absorption of radiated power by tissues in the human body. We present a battery-less, wearable knitted Ultra High Frequency (UHF, 902-928 MHz) Radio Frequency Identification (RFID) compression sensor (Bellypatch) antenna and show its applicability as an on-body respiratory monitor. The antenna radiation efficiency is satisfactory in both free-space and on-body operations. We extract RF (Radio Frequency) sheet resistance values of three knitted silver-coated nylon fabric candidates at 913 MHz. The best type of fabric is selected based on the extracted RF sheet resistance. Simulated and measured performance of the antenna confirm suitability for on-body applications. The proposed Bellypatch antenna is used to measure the breathing activity of a programmable infant patient emulator mannequin (SimBaby) and a human subject. The antenna is highly sensitive to respiratory compression and relaxation. Fluctuations in the backscatter power level/Received Signal Strength Indicator (RSSI) in both cases range from 6 dB to 15 dB. The improved on-body read range of the proposed sensor antenna is 5.8 m, about 10 times higher than its predecessor wearable knitted strain sensing Bellyband antenna (0.6 m). The maximum simulated Specific Absorption Rate (SAR) on a human torso model is 0.25 W/kg, lower than the maximum allowable limit of 1.6 W/kg.

An Adaptively Parameterized Algorithm Estimating Respiratory Rate from a Passive Wearable RFID Smart Garment.

Currently, wired respiratory rate sensors tether patients to a location and can potentially obscure their body from medical staff. In addition, current wired respiratory rate sensors are either inaccurate or invasive. Spurred by these deficiencies, we have developed the Bellyband, a less invasive smart garment sensor, which uses wireless, passive Radio Frequency Identification (RFID) to detect bio-signals. Though the Bellyband solves many physical problems, it creates a signal processing challenge, due to its noisy, quantized signal. Here, we present an algorithm by which to estimate respiratory rate from the Bellyband. The algorithm uses an adaptively parameterized Savitzky-Golay (SG) filter to smooth the signal. The adaptive parameterization enables the algorithm to be effective on a wide range of respiratory frequencies, even when the frequencies change sharply. Further, the algorithm is three times faster and three times more accurate than the current Bellyband respiratory rate detection algorithm and is able to run in real time. Using an off-the-shelf respiratory monitor and metronome-synchronized breathing, we gathered 25 sets of data and tested the algorithm against these trials. The algorithm's respiratory rate estimates diverged from ground truth by an average Root Mean Square Error (RMSE) of 4.1 breaths per minute (BPM) over all 25 trials. Further, preliminary results suggest that the algorithm could be made as or more accurate than widely used algorithms that detect the respiratory rate of non-ventilated patients using data from an Electrocardiogram (ECG) or Impedance Plethysmography (IP).

Channel Emulation for the Characterization of Wearable RFID Systems.

Wearable sensors with RFID (Radio Frequency Identification) tags are considered to be an integral part of the upcoming revolution in the IoT (Internet of Things) sector. As with many deployed IoT sensor systems, dynamic environment conditions present challenges in reliably measuring system performance; this difficulty is enhanced due to proprietary details about the sensors, such as an RFID chip embedded within a novel knitted antenna acting as a passive sensor. A repeatable and scalable platform is necessary to evaluate the performance of the entire system in the pre-deployment stage in order to compare the predicted effects of varying components, design, and integration of sensors in an integrated IoT device. This paper demonstrates the development of an RFID channel emulation testbed in the United States ISM band (902-928 MHz). The testbed includes a commercial RFID interrogator, a custom-built circuit board housing a commercial passive RFID chip, and a dynamic spectrum environment emulator (DYSE) for wireless channel emulation. A single link scenario was considered where the DYSE emulates the antenna gain fluctuation due to the sensing of breathing with a fabric-based RFID. Two regular and one irregular breathing scenarios were emulated, and breathing rate or anomaly was detected from post-processed RSSI (Received Signal Strength Indicator) data received by the RFID interrogator.

UHF RFID Channel Emulation Testbed for Wireless IoT Systems.

We propose an Ultra High Frequency (UHF) Radio Frequency Identification (RFID, 902-928 MHz in the US) channel emulation testbed that is capable of simultaneously emulating unique wireless channels. The proposed system can potentially be an invaluable tool in the design and validation of RFID-based Internet of Things (IoT) sensors and systems. Emulation of ray-tracing-based wireless channels enables the evaluation of inherently difficult and complex RF scenarios, particularly in situations when in-person experimentation is not feasible or desirable (e.g., during a pandemic or in a critical care facility). Furthermore, the emulation testbed is able to generate a large amount of sensor data in a limited time period. Machine learning techniques used in wireless IoT can be greatly enhanced by a large amount of data extracted from the emulation of dynamic and challenging environments. The proposed multi-channel emulation testbed is therefore a valuable solution for experimentation on real hardware and a convenient tool for pre-clinical-trial system validation.

Solution-Processed Ti3 C2 Tx MXene Antennas for Radio-Frequency Communication.

Highly integrated, flexible, and ultrathin wireless communication components are in significant demand due to the explosive growth of portable and wearable electronic devices in the fifth-generation (5G) network era, but only conventional metals meet the requirements for emerging radio-frequency (RF) devices so far. Here, it is reported on Ti3 C2 Tx MXene microstrip transmission lines with low-energy attenuation and patch antennas with high-power radiation at frequencies from 5.6 to 16.4 GHz. The radiation efficiency of a 5.5 µm thick MXene patch antenna manufactured by spray-coating from aqueous solution reaches 99% at 16.4 GHz, which is about the same as that of a standard 35 µm thick copper patch antenna at about 15% of its thickness and 7% of the copper weight. MXene outperforms all other materials evaluated for patch antennas to date. Moreover, it is demonstrated that an MXene patch antenna array with integrated feeding circuits on a conformal surface has comparable performance with that of a copper antenna array at 28 GHz, which is a target frequency in practical 5G applications. The versatility of MXene antennas in wide frequency ranges coupled with the flexibility, scalability, and ease of solution processing makes MXene promising for integrated RF components in various flexible electronic devices.

Pattern Reconfigurable UHF RFID Reader Antenna Array.

The growing research interest in passive RFID (Radio Frequency Identification)-based devices and sensors in a diverse group of applications calls for flexibility in reader antenna performance. We propose a low-cost, easy-to-fabricate, and pattern reconfigurable UHF (Ultra High Frequency) RFID reader antenna in the RFID ISM band (902-928 MHz in the US). The antenna offers a 54 MHz bandwidth (890 - 944 MHz) and 8.9 dBi maximum gain. The proposed reconfigurable antenna can radiate four electronically switchable radiation beams in the azimuth plane. The antenna is LHCP (Left Hand Circularly Polarized) with axial ratio (AR) in the ranging from 0.45 dB to 7 dB in the RFID ISM band. Simulation and measurements are presented, and they are in good agreement. The proposed reader array performance is compared against a commercially available reader antenna. The pattern reconfigurable UHF RFID reader antenna not only increases the coverage area for conventional RFID applications but also opens the door to on-body RFID sensor implementation and indoor localization applications.

Efficiency measurement of the flexible on-body antenna at varying levels of stretch in a reverberation chamber.

Flexible antennas have the potential to transform wearable and fabric-based wireless sensing technologies. The antenna discussed in this study is part of a sensing system that uses the back-scattered power level as the decision metric. For a good wireless sensor, it is necessary to offer a feasible read range and maintain good distinctions in the back-scattered power levels between the different states (i.e. level of stretch) of the antenna. Moreover, effects due to human body proximity should be minimised. For these reasons, the radiation efficiency is a crucial parameter to investigate. This study presents the radiation efficiency measurement of the proposed flexible knitted 'Bellyband' antenna at two different levels of stretch in a reverberation chamber. This work validates the reverberation chamber measurements through comparison with simulations and anechoic chamber measurements at 900 MHz. Moreover, this work demonstrates how the approach can be used to quantify bellyband antenna efficiency in the vicinity of a human body. Finally, the efficiency results were used to predict the read range of Bellyband radio frequency identification technology.

On the Effect of Sweat on Sheet Resistance of Knitted Conductive Yarns in Wearable Antenna Design.

Researchers are looking for new methods to integrate sensing capabilities into textiles while maintaining the durability, flexibility, and comfort of the garment. One method for imparting sensing capabilities into garments is through coupling conductive yarns with the radio frequency identification (RFID) technology. These smart devices have exhibited promising results for short-term use. However, long-term studies of their performance are still needed to evaluate their performance over a longer period. Like all garments, wearable sensors are susceptible to environmental factors during use. These factors can lead to dielectric coupling and corrosion of conductive yarns, which has the potential to degrade the performance of the device. This letter analyzes the effect of sweat and moisture on silver-coated nylon yarn by extracting the sheet resistance at 913 MHz from transmission line measurements. HFSS simulation shows the level of perturbation in antenna performance as sheet resistance increased with each cycle of sweat-immersion, washing, and drying.

Fusion Learning on Multiple-Tag RFID Measurements for Respiratory Rate Monitoring.

Future advances in the medical Internet of Things (IoT) will require sensors that are unobtrusive and passively powered. With the use of wireless, wearable, and passive knitted smart garment sensors, we monitor infant respiratory activity. We improve the utility of multi-tag Radio Frequency Identification (RFID) measurements via fusion learning across various features from multiple tags to determine the magnitude and temporal information of the artifacts. In this paper, we develop an algorithm that classifies and separates respiratory activity via a Regime Hidden Markov Model compounded with higher-order features of Minkowski and Mahalanobis distances. Our algorithm improves respiratory rate detection by increasing the Signal to Noise Ratio (SNR) on average from 17.12 dB to 34.74 dB. The effectiveness of our algorithm in increasing SNR shows that higher-order features can improve signal strength detection in RFID systems. Our algorithm can be extended to include more feature sources and can be used in a variety of machine learning algorithms for respiratory data classification, and other applications. Further work on the algorithm will include accurate parameterization of the algorithm's window size.

Ensemble Learning Approach via Kalman Filtering for a Passive Wearable Respiratory Monitor.

OBJECTIVE: Utilizing passive radio frequency identification (RFID) tags embedded in knitted smart-garment devices, we wirelessly detect the respiratory state of a subject using an ensemble-based learning approach over an augmented Kalman-filtered time series of RF properties. METHODS: We propose a novel approach for noise modeling using a "reference tag," a second RFID tag worn on the body in a location not subject to perturbations due to respiratory motions that are detected via the primary RFID tag. The reference tag enables modeling of noise artifacts yielding significant improvement in detection accuracy. The noise is modeled using autoregressive moving average (ARMA) processes and filtered using state-augmented Kalman filters. The filtered measurements are passed through multiple classification algorithms (naive Bayes, logistic regression, decision trees) and a new similarity classifier that generates binary decisions based on current measurements and past decisions. RESULTS: Our findings demonstrate that state-augmented Kalman filters for noise modeling improves classification accuracy drastically by over 7.7% over the standard filter performance. Furthermore, the fusion framework used to combine local classifier decisions was able to predict the presence or absence of respiratory activity with over 86% accuracy. CONCLUSION: The work presented here strongly indicates the usefulness of processing passive RFID tag measurements for remote respiration activity monitoring. The proposed fusion framework is a robust and versatile scheme that once deployed can achieve high detection accuracy with minimal human intervention. SIGNIFICANCE: The proposed system can be useful in remote noninvasive breathing state monitoring and sleep apnea detection.

Activity Segmentation Using Wearable Sensors for DVT/PE Risk Detection.

Using a wearable electromyography (EMG) and an accelerometer sensor, classification of subject activity state (i.e., walking, sitting, standing, or ankle circles) enables detection of prolonged "negative" activity states in which the calf muscles do not facilitate blood flow return via the deep veins of the leg. By employing machine learning classification on a multi-sensor wearable device, we are able to classify human subject state between "positive" and "negative" activities, and among each activity state, with greater than 95% accuracy. Some negative activity states cannot be accurately discriminated due to their similar presentation from an accelerometer (i.e., standing vs. sitting); however, it is desirable to separate these states to better inform the risk of developing a Deep Vein Thrombosis (DVT). Augmentation with a wearable EMG sensor improves separability of these activities by 30%.

On the Use of Knitted Antennas and Inductively Coupled RFID Tags for Wearable Applications.

Recent advancements in conductive yarns and fabrication technologies offer exciting opportunities to design and knit seamless garments equipped with sensors for biomedical applications. In this paper, we discuss the design and application of a wearable strain sensor, which can be used for biomedical monitoring such as contraction, respiration, or limb movements. The system takes advantage of the intensity variations of the backscattered power (RSSI) from an inductively-coupled RFID tag under physical stretching. First, we describe the antenna design along with the modeling of the sheet impedance, which characterizes the conductive textile. Experimental results with custom fabricated prototypes showed good agreement with the numerical simulation of input impedance and radiation pattern. Finally, the wearable sensor has been applied for infant breathing monitoring using a medical programmable mannequin. A machine learning technique has been developed and applied to post-process the RSSI data, and the results show that breathing and non-breathing patterns can be successfully classified.

High-data-rate ultrasonic through-metal communication.

A link-adaptive frequency division multiplexing (OFDM) ultrasonic physical layer is proposed for high-data-rate communications through metal walls. The ultrasonic link allows for communication without physical penetration of the metal barrier. Link-adaptive OFDM mitigates the severe frequency- selective fading of the ultrasonic channel and greatly improves throughput over impulse or narrowband communication systems. Throughput improvements of 300% are demonstrated over current narrowband low-frequency techniques, and show improved spectral efficiency over high-frequency techniques found in the literature.