Smart rings represent a promising frontier in wearables for pervasive physiological health monitoring. They offer convenient and unobtrusive wear and reliable and firm contact with the finger for higher sensing fidelity. However, their use for monitoring complex physiological parameters with clinical use-cases, such as blood pressure (BP), still remains a challenge. While existing approaches including optical (e.g., PPG) and electrical (e.g., bio-impedance) modalities offer great promise, the direction to measure small mechanical movements of the skin due to blood volumetric changes and convert this to BP provides an exciting and complementary direction. Our Hall effect ring sensor can capture variations in magnetic flux due to the underlying pulsatile activity at digital arteries. We provide an end-to-end analysis to assess the performance of our smart rings in estimating BP. Our results indicate that we achieved a mean absolute error (MAE) of 4.79 mmHg and 2.61 mmHg for systolic and diastolic blood pressure, respectively, demonstrating the potential of this innovative design for clinical applications.

Inter-beat interval (IBI) measurement enables estimation of heart-tare variability (HRV) which, in turn, can provide early indication of potential cardiovascular diseases (CVDs). However, extracting IBIs from noisy signals is challenging since the morphology of the signal gets distorted in the presence of noise. Electrocardiogram (ECG) of a person in heavy motion is highly corrupted with noise, known as motion-artifact, and IBI extracted from it is inaccurate. As a part of remote health monitoring and wearable system development, denoising ECG signals and estimating IBIs correctly from them have become an emerging topic among signal-processing researchers. Apart from conventional methods, deep-learning techniques have been successfully used in signal denoising recently, and diagnosis process has become easier, leading to accuracy levels that were previously unachievable. We propose a deep-learning approach leveraging tiramisu autoencoder model to suppress motion-artifact noise and make the R-peaks of the ECG signal prominent even in the presence of high-intensity motion. After denoising, IBIs are estimated more accurately expediting diagnosis tasks. Results illustrate that our method enables IBI estimation from noisy ECG signals with SNR up to -30 dB with average root mean square error (RMSE) of 13 milliseconds for estimated IBIs. At this noise level, our error percentage remains below 8% and outperforms other state-of-the-art techniques.

Accurate estimation of physiological biomarkers using raw waveform data from non-invasive wearable devices requires extensive data preprocessing. An automatic noise detection method in time-series data would offer significant utility for various domains. As data labeling is onerous, having a minimally supervised abnormality detection method for input data, as well as an estimation of the severity of the signal corruptness, is essential. We propose a model-free, time-series biomedical waveform noise detection framework using a Variational Autoencoder coupled with Gaussian Mixture Models, which can detect a range of waveform abnormalities without annotation, providing a confidence metric for each segment. Our technique operates on biomedical signals that exhibit periodicity of heart activities. This framework can be applied to any machine learning or deep learning model as an initial signal validator component. Moreover, the confidence score generated by the proposed framework can be incorporated into different models' optimization to construct confidence-aware modeling. We conduct experiments using dynamic time warping (DTW) distance of segments to validated cardiac cycle morphology. The result confirms that our approach removes noisy cardiac cycles and the remaining signals, classified as clean, exhibit a 59.92% reduction in the standard deviation of DTW distances. Using a dataset of bio-impedance data of 97885 cardiac cycles, we further demonstrate a significant improvement in the downstream task of cuffless blood pressure estimation, with an average reduction of 2.67 mmHg root mean square error (RMSE) of Diastolic Blood pressure and 2.13 mmHg RMSE of systolic blood pressure, with increases of average Pearson correlation of 0.28 and 0.08, with a statistically significant improvement of signal-to-noise ratio respectively in the presence of different synthetic noise sources. This enables burden-free validation of wearable sensor data for downstream biomedical applications.

We propose our Confidence-Aware Particle Filter (CAPF) framework that analyzes a series of estimated changes in blood pressure (BP) to provide several true state hypotheses for a given instance. Particularly, our novel confidence-awareness mechanism assigns likelihood scores to each hypothesis in an effort to discard potentially erroneous measurements - based on the agreement amongst a series of estimated changes and the physiological plausibility when considering DBP/SBP pairs. The particle filter formulation (or sequential Monte Carlo method) can jointly consider the hypotheses and their probabilities over time to provide a stable trend of estimated BP measurements. In this study, we evaluate BP trend estimation from an emerging bio-impedance (Bio-Z) prototype wearable modality although it is applicable to all types of physiological modalities. Each subject in the evaluation cohort underwent a hand-gripper exercise, a cold pressor test, and a recovery state to increase the variation to the captured BP ranges. Experiments show that CAPF yields superior continuous pulse pressure (PP), diastolic blood pressure (DBP), and systolic blood pressure (SBP) estimation performance compared to ten baseline approaches. Furthermore, CAPF performs on track to comply with AAMI and BHS standards for achieving a performance classification of Grade A, with mean error accuracies of -0.16 ± 3.75 mmHg for PP (r = 0.81), 0.42 ± 4.39 mmHg for DBP (r = 0.92), and -0.09 ± 6.51 mmHg for SBP (r = 0.92) from more than test 3500 data points.

Bioimpedance is a powerful modality to continuously and non-invasively monitor cardiovascular and respiratory health parameters through the wearable operation. However, for bioimpedance sensors to be utilized in medical-grade settings, the reliability and robustness of the system should be improved. Previous studies provide limited fundamental analyses of the factors involved in the system that impact the sensitivity and the specificity of the modality in capturing the hemodynamics. This study provides a parametric model of the human wrist that involves different tissue layers (i.e., skin, fat, artery, muscle, bone) with complex dielectric properties built based on the human wrist anatomy. We run a frequency domain electrical field simulation using finite element analysis to map electric current distribution within the wrist to find the optimum operating frequency and electrode placement that provide the highest sensitivity and specificity to the blood flow. Our results suggest using an operating frequency between 10-100 kHz range with minimal electrode separation to capture the pulsatile activity with high accuracy.

Modality translation grants diagnostic value to wearable devices by translating signals collected from low-power sensors to their highly-interpretable counterparts that are more familiar to healthcare providers. For instance, bio-impedance (Bio-Z) is a conveniently collected modality for measuring physiological parameters but is not highly interpretable. Thus, translating it to a well-known modality such as electrocardiogram (ECG) improves the usability of Bio-Z in wearables. Deep learning solutions are well-suited for this task given complex relationships between modalities generated by distinct processes. However, current algorithms usually train a single model for all users that results in ignoring cross-user variations. Retraining for new users usually requires collecting abundant labeled data, which is challenging in healthcare applications. In this paper, we build a modality translation framework to translate Bio-Z to ECG by learning personalized user information without training several independent architectures. Furthermore, our framework is able to adapt to new users in testing using very few samples. We design a meta-learning framework that contains shared and user-specific parameters to account for user differences while learning from the similarity amongst user signals. In this model, a meta-learner approximated by a neural network learns how to learn user-specific parameters and can efficiently update them in testing. Our experiments show that the proposed model reduces the percentage root mean square difference (PRD) by 41% compared to training a single model for all users and by 36% compared to training independent models for each user. When adapting the model to new users, our model outperforms fine-tuning a pre-trained model through back-propagation by 40% using as few as two new samples in testing.

Estimating physiological parameters - such as blood pressure (BP) - from raw sensor data captured by noninvasive, wearable devices rely on either burdensome manual feature extraction designed by domain experts to identify key waveform characteristics and phases, or deep learning (DL) models that require extensive data collection. We propose the Data-Driven Guided Attention (DDGA) framework to optimize DL models to learn features supported by the underlying physiology and physics of the captured waveforms, with minimal expert annotation. With only a single template waveform cardiac cycle and its labelled fiducial points, we leverage dynamic time warping (DTW) to annotate all other training samples. DL models are trained to first identify them before estimating BP to inform them which regions of the input represent key phases of the cardiac cycle, yet we still grant the flexibility for DL to determine the optimal feature set from them. In this study, we evaluate DDGA's improvements to a BP estimation task for three prominent DL-based architectures with two datasets: 1) the MIMIC-III waveform dataset with ample training data and 2) a bio-impedance (Bio-Z) dataset with less than abundant training data. Experiments show that DDGA improves personalized BP estimation models by an average 8.14% in root mean square error (RMSE) when there is an imbalanced distribution of target values in a training set and improves model generalizability by an average 4.92% in RMSE when testing estimation of BP value ranges not previously seen in training.

This article describes a novel methodology to balance computational power and estimation accuracy for the application of robust heartrate (HR) monitoring through leveraging particle filters. We formulate the power-accuracy trade-off as the number of particles in the particle filter framework, where a higher number of particles leads to higher computational resolution and accuracy at the cost of higher computational power. Our particle filter-based HR monitoring technique can be applied to a variety of physiological signals including but not limited to electrocardiogram (ECG) and photoplethysmogram (PPG).

Wearable systems have unlocked new sensing paradigms in various applications such as human activity recognition, which can enhance effectiveness of mobile health applications. Current systems using wearables are not capable of understanding their surroundings, which limits their sensing capabilities. For instance, distinguishing certain activities such as attending a meeting or class, which have similar motion patterns but happen in different contexts, is challenging by merely using wearable motion sensors. This article focuses on understanding user's surroundings, i.e., environmental context, to enhance capability of wearables, with focus on detecting complex activities of daily living (ADL). We develop a methodology to automatically detect the context using passively observable information broadcasted by devices in users’ locale. This system does not require specific infrastructure or additional hardware. We develop a pattern extraction algorithm and probabilistic mapping between the context and activities to reduce the set of probable outcomes. The proposed system contains a general ADL classifier working with motion sensors, learns personalized context, and uses that to reduce the search space of activities to those that occur within a certain context. We collected real-world data of complex ADLs and by narrowing the search space with context, we improve average F1-score from 0.72 to 0.80.

Cardiovascular disorders cause nearly one in three deaths in the United States. Short- and long-term care for these disorders is often determined in short-term settings. However, these decisions are made with minimal longitudinal and long-term data. To overcome this bias towards data from acute care settings, improved longitudinal monitoring for cardiovascular patients is needed. Longitudinal monitoring provides a more comprehensive picture of patient health, allowing for informed decision making. This work surveys sensing and machine learning in the field of remote health monitoring for cardiovascular disorders. We highlight three needs in the design of new smart health technologies: (1) need for sensing technologies that track longitudinal trends of the cardiovascular disorder despite infrequent, noisy, or missing data measurements; (2) need for new analytic techniques designed in a longitudinal, continual fashion to aid in the development of new risk prediction techniques and in tracking disease progression; and (3) need for personalized and interpretable machine learning techniques, allowing for advancements in clinical decision making. We highlight these needs based upon the current state of the art in smart health technologies and analytics. We then discuss opportunities in addressing these needs for development of smart health technologies for the field of cardiovascular disorders and care.

Annotating activities of daily living (ADL) is vital for developing machine learning models for activity recognition. In addition, it is critical for self-reporting purposes such as in assisted living where the users are asked to log their ADLs. However, data annotation becomes extremely challenging in real-world data collection scenarios, where the users have to provide annotations and labels on their own. Methods such as self-reports that rely on users’ memory and compliance are prone to human errors and become burdensome since they increase users’ cognitive load. In this article, we propose a light yet effective context-aware change point detection algorithm that is implemented and run on a smartwatch for facilitating data annotation for high-level ADLs. The proposed system detects the moments of transition from one to another activity and prompts the users to annotate their data. We leverage freely available Bluetooth low energy (BLE) information broadcasted by various devices to detect changes in environmental context. This contextual information is combined with a motion-based change point detection algorithm, which utilizes data from wearable motion sensors, to reduce the false positives and enhance the system's accuracy. Through real-world experiments, we show that the proposed system improves the quality and quantity of labels collected from users by reducing human errors while eliminating users’ cognitive load and facilitating the data annotation process.

Novel wearable sensor modalities, such as bio-impedance (Bio-Z), are being introduced and often provide various advantages over current state-of-the-art in terms of accuracy, sensing coverage, or convenience of wear. The principal challenge, however, lies in the ability to interpret the sensor reading by healthcare providers. In this work, we propose a two-stage deep learning framework that leverages a novel attention mechanism to translate Bio-Z signals to highly interpretable electrocardiogram (ECG) waveforms while also predicting translation uncertainty. Our experiments indicate a 66% improvement in accuracy for 1D-CNN based models to perform competitively with more sophisticated hybrid CNN-LSTM based models in a fraction of the training time while also providing a valid uncertainty measurement.

Personal tracking algorithms for health monitoring are critical for understanding an individual's life-style and personal choices in natural environments (NE). In order to train such tracking algorithms in NE, however, annotated data is needed, particularly when tracking a variety of activities of daily living. These algorithms are often trained in laboratory settings, with expectations that they will perform equally well in NE, which is often not the case; they must be trained on annotated data collected in NE and wearable computers provide opportunities to collect such data, though the process is burdensome. Therefore, we propose an intelligent scoring algorithm that limits the number of user annotation requests through the confidence of predictions generated by the tracking algorithm and automatically annotating data with high confidence. We enhance our scoring algorithm by providing improvements in our tracking algorithm by obtaining context data from nearable sensors. Each specific context of a user bounds the set of activities that can likely occur, which in turn improves the tracking algorithm and confidence. Finally, we propose a hierarchical annotation approach, where repeated use allows us to ask for detailed annotations that differentiate fine-grained differences in ways individuals perform activities. We validate our approach in a diet monitoring case study. We vary the number of annotations requested per day to evaluate model accuracy; we improve accuracy in NE by 8% when restricting requests to 20 per day and improve F1-score of activities by 11% with hierarchical annotations, while discussing implementation, accuracy, and power consumption in real-time use.

We introduce a novel approach for robust estimation of physiological parameters such as interbeat interval (IBI) and heart rate variability (HRV) from cardiac signals captured with wearable sensors in the presence of motion artifacts. Motion artifact due to physical exercise is known as a major source of noise that contributes to a significant decline in the performance of IBI and HRV estimation techniques for cardiac monitoring in free-living environments. Therefore, developing robust estimation algorithms is essential for utilization of wearable sensors in daily life situations. The proposed approach includes two algorithmic components. First, we propose a combinatorial technique to select characteristic points that define heartbeats in noisy signals in time domain. The heartbeat detection problem is defined as a shortest path search problem on a direct acyclic graph that leverages morphological features of the cardiac signals by taking advantage of the time-continuity of heartbeats - each heartbeat ends with the starting point of the next heartbeat. The graph is constructed with vertices and edges representing candidate morphological features and IBIs, respectively. Second, we propose a fusion technique to combine physiological parameters estimated from different morphological features using the shortest path algorithm to obtain more accurate IBI/HRV estimations. We evaluate our techniques on motion-corrupted photoplethysmogram and electrocardiogram signals. Our results indicate that the estimated IBIs are highly correlated with the ground truth (r = 0.89) and detected HRV parameters indicate high correlation with the true HRV parameters. Furthermore, our findings demonstrate that the developed fusion technique, which utilizes different morphological features, achieves a correlation coefficient that is at least 3% higher than that obtained using single physiological characteristic.

Recognizing activities of daily living (ADL) in-the-wild, while users follow their daily routine, is challenging due to the presence of various activities that do not belong to the set of desired activities in which the system is interested (i.e., NULL class). In this paper, we propose a framework for ADL recognition via wearable motion sensors with the ability to detect NULL class. Existing ADL recognition systems either ignore the NULL class or use some training data to train a model for recognizing it. However, our framework uses only samples of the desired activities in the training phase and learns to detect the NULL samples based on a modified variational autoencoder model that outputs reconstruction probability. Experimental results show that in detecting six ADL with accelerometer data, our system achieves 14% higher F1-score compared to the models that use training samples of NULL activities.

Wearable sensors provide enormous opportunities to identify activities and events of interest for various applications. However, a major limitation of the current systems is the fact that machine learning algorithms trained on particular sensors need to be retrained upon any changes in configuration of the system, such as adding a new sensor. In this paper, we aim to seamlessly train machine learning algorithms for the new sensors to identify activities and observations that are detectable by the pre-existing sensors. We create a domain adaptation method to expand training algorithms from known wearable sensors to new sensors, eliminating the need for manual training of machine learning algorithms. Specifically, our proposed approach eliminates the need for capturing substantial amount of data on new sensors. We propose the concept of stochastic features for human activity recognition, and design a new architecture of deep neural network to approximate the posterior distribution of the features. This approximation aligns the feature space of the new and old sensors by using limited, unlabeled data from the new sensor so that the previously defined classifier can be used with the new sensor. The experimental results show that (i) stochastic features are more robust against additive noise compared to typical convolutional neural networks based on deterministic features (ii) our framework outperforms the state-of-the-art domain adaptation algorithms. It can also achieve 10% improvement when training new sensors with limited unlabeled training data compared to training a model from scratch for the new sensor.

Activity/gesture recognition using wearable motion sensors, also known as inertial measurement units (IMUs), provides an important context for many ubiquitous sensing applications. The design of the activity/gesture recognition algorithms typically requires information about the placement and orientation of the IMUs on the body, and the signal processing is often designed to work with a known orientation and placement of the sensors. However, sensors could be worn or held differently. Therefore, signal processing algorithms may not perform as well as expected. In this paper, we present an orientation independent activity/gesture recognition approach by exploring a novel feature set that functions irrespective of how the sensors are oriented. We also propose a template refinement technique to determine the best representative segment of each gesture thus improving the recognition accuracy. We evaluated our approach in the context of two applications: 1) activity of daily living recognition and 2) hand gesture recognition. The experimental results show that our approach achieves 98.2% and 95.6% average accuracies for subject dependent testing of activities of daily living and gestures, respectively.

The integration knowledge of wearable sensors, wireless communications, and artificial intelligence have brought forth the smart health systems, which empower the consumer’s to make a difference to their well-being by connecting data to personalized analysis to timely insights. Therefore, the real-time data obtained directly reflects the personal status of interest and can be used in a variety of healthcare applications in the Internet of Things (IoT), from preventive treatment to diagnostics and rehabilitation, as well as in virtual and augmented reality environments.

As the Internet of Things (IoT) continues to expand into our daily lives, consumers are finding a growing catalogue of smart devices to boost the intelligence of their homes. Currently, the user must manage a proprietary user interface (UI) for each device, and each application comes with its own UI, creating a cumbersome app environment. Clearly, a single UI that can control all of these devices would be preferable. This interface should be accessible using forms of communication that feel natural, for example, speech, body language, and facial expressions, to name a few. In this paper, we propose a framework for multimodal UI using a flexible, slotted command ontology and decision-level Bayesian fusion. Our case study explores command recognition for device control with a wearable system accessed via speech and gestures, using a wrist-mounted inertial measurement unit (IMU) for hand gesture recognition. We achieve an accuracy of 94.82% on a set of 17 commands.

This paper describes a novel methodology leveraging particle filters for the application of robust heart rate monitoring in the presence of motion artifacts. Motion is a key source of noise that confounds traditional heart rate estimation algorithms for wearable sensors due to the introduction of spurious artifacts in the signals. In contrast to previous particle filtering approaches, we formulate the heart rate itself as the only state to be estimated, and do not rely on multiple specific signal features. Instead, we design observation mechanisms to leverage the known steady, consistent nature of heart rate variations to meet the objective of continuous monitoring of heart rate using wearable sensors. Furthermore, this independence from specific signal features also allows us to fuse information from multiple sensors and signal modalities to further improve estimation accuracy. The signal processing methods described in this work were tested on real motion artifact affected electrocardiogram and photoplethysmogram data with concurrent accelerometer readings. Results show promising average error rates less than 2 beats/min for data collected during intense running activities. Furthermore, a comparison with contemporary signal processing techniques for the same objective shows how the proposed implementation is also computationally more efficient for comparable performance.

The aim of this paper is to propose a robust, accurate and portable system for human body motion measurement. The system includes Inertial Measurement Units (IMU) and a Kinect or vision sensors. Since Kinect sampling rate is low (30 Hz per second) and it suffers from occlusion, it cannot individually measure human body motion accurately and robustly. On the other hand, IMU does not suffer from these problems, but it suffers from drift in particular with long-term motion monitoring and other types of errors (e.g., high acceleration motions, temperature and voltage variations). Thus, in this study, IMU and Kinect data were fused using a context enhanced extended Kalman filter. Rules were generated based on the context of motion in order to adjust Kalman filter parameters. In addition, an automated approach is introduced to estimate the variance of the noise of the sensors during the operation. Considering motion context and automatic noise detection, the robustness of monitoring is enhanced against errors related to motion context (i.e., high acceleration and long-term motions); furthermore, offline calibration is no longer required to set the parameters of the filter. The system was tested on leg and arm motions. The root mean square error of our fusion method was 6.08° lower than using only gyroscope, 16.98° lower than using only accelerometer, 2.49° lower than using only the Kinect and 8.99° lower than using simple EKF fusion method, which does not consider motion context and automatic noise estimation.

The cognitive states of students in a lecture can give good indications of student concentration and learning, and therefore, modeling them would have a positive impact on their quality of education by enabling the intervention of instructors. In a traditional class, the instructor would assess the students' level of attention. However, the assessment may not be accurate for a variety of reasons. Additionally, this creates a burden for the instructors. Wearable sensors and signal processing techniques could provide opportunities to assist teachers with this assessment. In this paper, we propose a methodology to model students' cognitive states by leveraging hand motion and heart activity captured with smart watches. Following the application of a sequence of signal processing techniques to the raw data, we generate features, which describe characteristics of the hand motion and heart activity in a group of students. The most prominent features are selected for machine learning algorithms. By applying cross validation, the results of experiments on 30 students in two lectures offer accuracies of 98.99% and 95.78% for predictions of `interest level' and `perception of difficulty' on the topics covered during the lectures.

Gait analysis is important in diagnosing and quantifying the severity of Parkinson's disease. Different motion tracking systems such as inertial measurement units (IMU) are widely used to detect gait parameters associated with the severity of Parkinson's disease. Although these systems are accurate enough to measure different gait parameters, they utilize a predefined model of human gait to measure these parameters. Model- based signal processing, that takes into account the kinematics of human body, enforces that sensors be placed in a certain configuration in terms of orientation and location which introduces a burden at the signal processing development phase. In addition, it affects the accuracy and robustness of the system when the user does not place the sensors at their pre-defined locations and with a pre-define orientation. In this paper, we introduce a set of model-free features to estimate gait parameters for the applications of diagnosing and quantifying the severity of Parkinson's disease. A model-free signal processing technique does not limit sensor placement, in addition, it does not require the knowledge on the kinematics of the users and the human subjects. We show that our proposed features, using a model-free signal processing technique, are highly correlated (R-value up to 0.96 for suitable locations) with gait parameters obtained from model-based sophisticated algorithms. Therefore, these simple model-free features may be suitable for ongoing assessment of Parkinson's disease and they can be an alternative for conventional gait parameters used for rapid application development.

Ambulatory blood pressure monitoring (ABPM) is a term used to describe measurement of blood pressure (BP) at regular intervals and can be an important diagnostic tool, especially for hypertensive patients. Since traditional cuff-based measurement of blood pressure is not convenient for a continuous, ABPM measurement, one approach that is being explored is pulse wave velocity (PWV), which is known to be correlated with blood pressure. However, an easily wearable and reliable system has not been realized to date. In this work, we examine the feasibility and design requirements for measuring PWV using two photoplethysmography sensors placed 4 cm apart on the arm. Measurements of PWV in vivo were made with our system and we showed that PWV measurements changed in accord with induced variations in blood pressure of the subject. Furthermore, we examined the minimum requirements for the sampling rate and bit resolution for the analog-to-digital converter (ADC) in our system with error in PWV measurement as the criterion. Our results show the feasibility of measuring PWV at small distances and outlined the design requirements for an ABPM device.

Sensors and actuators are finding their way into our lives and our surroundings at a very fast pace. These heterogeneous sensors deployed in the environment can prove to be useful in providing insights into the behavior and trends of the environment. In this work, we capture a part of that knowledge and propose a novel concept called Urban Heartbeat using data captured by various sensors that essentially identify periodic activities in the environment. The Urban Heartbeat can be leveraged to identify when an unexpected event has occurred or is about to occur to more effectively prepare the citizens. We first develop techniques to find couplings between sensors using multiple operators, in cases when direct measurement of a parameter is not possible. Next, we define an algorithm that can be used to find quasi-periodic patterns from time series data that has spatiotemporal deviations. We then introduce the notion of Urban Heartbeat, which leverages data from heterogeneous sensors to identify the normal heartbeat of the environment. The Urban Heartbeat can be used not only to differentiate between normal and abnormal trends thereby giving us the ability to detect anomalies but also in making predictions about the user or the environment behavior. We also show how we build heartbeat for a lab environment, learn useful information about the users and offer predictions about their behavior in the lab.

The Internet of Things (IoT) is fueled by the growth of sensors, actuators, and services that collect and process raw sensor data. Wearable and environmental sensors will be a major component of the IoT and provide context about people and activities that are occurring. It is imperative that sensors in the IoT are synchronized, which increases the usefulness and value of the sensor data and allows data from multiple sources to be combined and compared. Due to the heterogeneous nature of sensors (e.g., synchronization protocols, communication channels, etc.), synchronization can be difficult. In this article, we present novel techniques for synchronizing data from multi-sensor environments based on the events and interactions measured by the sensors. We present methods to determine which interactions can likely be used for synchronization and methods to improve synchronization by removing erroneous synchronization points. We validate our technique through experiments with wearable and environmental sensors in a laboratory environment. Experiments resulted in median drift error reduction from 66% to 98% for sensors synchronized through physical interactions.

Wearable inertial devices are being widely used in the applications of activity tracking, health care, and professional sports, and their usage is on a rapid rise. Signal processing algorithms for these devices are often designed to work with a known location of the wearable sensor on the body. However, in reality, the wearable sensor may be worn at different body locations due to the user's preference or unintentional misplacement. The calibration of the sensor location is important to ensure that the algorithms operate correctly. In this article, we propose an auto-calibration technique for determining the location of wearables on the body by fusing the 3-axis accelerometer data from the devices and three-dimensional camera (i.e., Kinect) information obtained from the environment. The automatic calibration is achieved by a cascade decision-tree-based classifier on top of the minimum least-squares errors obtained by solving Wahba's problem, operating on heterogeneous sensors. The core contribution of our work is that there is no extra burden on the user as a result of this technique. The calibration is done seamlessly, leveraging sensor fusion in an Internet-of-Things setting opportunistically when the user is present in front of an environmental camera performing arbitrary movements. Our approach is evaluated with two different types of movements: simple actions (e.g., sit-to-stand or picking up phone) and complicated tasks (e.g., cooking or playing basketball), yielding 100% and 82.56% recall for simple actions and for complicated tasks, respectively, in determining the correct location of sensors.

This paper presents a human action recognition approach by the simultaneous deployment of a second generation Kinect depth sensor and a wearable inertial sensor. Three data modalities consisting of depth images, skeleton joint positions, and inertial signals are fused by utilizing three collaborative representation classifiers. A database consisting of 10 actions performed by 6 subjects is put together to carry out two types of testing of the developed fusion approach: subject-generic and subject-specific. The overall recognition rates obtained from both types of testing indicate recognition improvements when fusing all the data modalities compared to the situations when data modalities are used individually.

Noninvasive continuous blood pressure (BP) monitoring is not yet practically available for daily use. Challenges include making the system easily wearable, reducing noise level and improving accuracy. Variations in each person's physical characteristics, as well as the possibility of different postures, increase the complexity of continuous BP monitoring, especially outside the hospital. This study attempts to provide an easily wearable solution and proposes training to specific posture and individual for further improving accuracy. The wrist watch-based system we developed can measure electrocardiogram and photoplethysmogram. From these two signals, we measure pulse transit time through which we can obtain systolic and diastolic blood pressure through regression techniques. In this study, we investigate various functions to perform the training to obtain blood pressure. We validate measurements on different postures and subjects, and show the value of training the device to each posture and each subject. We observed that the average RMSE between the measured actual systolic BP and calculated systolic BP is between 7.83 to 9.37 mmHg across 11 subjects. The corresponding range of error for diastolic BP is 5.77 to 6.90 mmHg. The system can also automatically detect the arm position of the user using an accelerometer with an average accuracy of 98%, to make sure that the sensor is kept at the proper height. This system, called BioWatch, can potentially be a unified solution for heart rate, SPO2 and continuous BP monitoring.

The papers in this special section focus on the topic of sensor informatics for mental health applications. The papers provide novel insights on advances in detection, sensing, analysis, and modeling of central and/or autonomic correlates useful in psychophysiological states assessment.

A sign language recognition system translates signs performed by deaf individuals into text/speech in real time. Inertial measurement unit and surface electromyography (sEMG) are both useful modalities to detect hand/arm gestures. They are able to capture signs and the fusion of these two complementary sensor modalities will enhance system performance. In this paper, a wearable system for recognizing American Sign Language (ASL) in real time is proposed, fusing information from an inertial sensor and sEMG sensors. An information gain-based feature selection scheme is used to select the best subset of features from a broad range of well-established features. Four popular classification algorithms are evaluated for 80 commonly used ASL signs on four subjects. The experimental results show 96.16% and 85.24% average accuracies for intra-subject and intra-subject cross session evaluation, respectively, with the selected feature subset and a support vector machine classifier. The significance of adding sEMG for ASL recognition is explored and the best channel of sEMG is highlighted.

Electroencephalography (EEG) is the recording of electrical activity produced by the firing of neurons within the brain. These activities can be decoded by signal processing techniques. However, EEG recordings are always contaminated with artifacts which hinder the decoding process. Therefore, identifying and removing artifacts is an important step. Researchers often clean EEG recordings with assistance from independent component analysis (ICA), since it can decompose EEG recordings into a number of artifact-related and event-related potential (ERP)-related independent components. However, existing ICA-based artifact identification strategies mostly restrict themselves to a subset of artifacts, e.g., identifying eye movement artifacts only, and have not been shown to reliably identify artifacts caused by nonbiological origins like high-impedance electrodes. In this paper, we propose an automatic algorithm for the identification of general artifacts. The proposed algorithm consists of two parts: 1) an event-related feature-based clustering algorithm used to identify artifacts which have physiological origins; and 2) the electrode-scalp impedance information employed for identifying nonbiological artifacts. The results on EEG data collected from ten subjects show that our algorithm can effectively detect, separate, and remove both physiological and nonbiological artifacts. Qualitative evaluation of the reconstructed EEG signals demonstrates that our proposed method can effectively enhance the signal quality, especially the quality of ERPs, even for those that barely display ERPs in the raw EEG. The performance results also show that our proposed method can effectively identify artifacts and subsequently enhance the classification accuracies compared to four commonly used automatic artifact removal methods.

Wearables are being widely utilized in health and wellness applications, primarily due to the recent advances in sensor and wireless communication, which enhance the promise of wearable systems in providing continuous and real-time monitoring and interventions. Wearables are generally composed of hardware/software components for collection, processing, and communication of physiological data. Practical implementation of wearable monitoring in real-life applications is currently limited due to notable obstacles. The wearability and form factor are dominated by the amount of energy needed for sensing, processing, and communication. In this article, we propose an ultra-low-power granular decision-making architecture, also called screening classifier, which can be viewed as a tiered wake-up circuitry, consuming three orders of magnitude-less power than the state-of-the-art low-power microcontrollers. This processing model operates based on computationally simple template matching modules, based on coarse- to fine-grained analysis of the signals with on-demand and gradually increasing the processing power consumption. Initial template matching rejects signals that are clearly not of interest from the signal processing chain, keeping the rest of processing blocks idle. If the signal is likely of interest, the sensitivity and the power of the template matching modules are gradually increased, and ultimately, the main processing unit is activated. We pose optimization techniques to efficiently split a full template into smaller bins, called mini-templates, and activate only a subset of bins during each classification decision. Our experimental results on real data show that this signal screening model reduces power consumption of the processing architecture by a factor of 70% while the sensitivity of detection remains at least 80%.

Context aware systems like smart homes and offices will benefit from determining human-object and human-human interactions. In this paper, we explore interaction detection methods using only wearable Inertial Measurement Units (IMUs). The interactions we explore involve two actors - the primary person and a secondary object or person. We explore how several commonly used time domain signal processing operators can be utilized to detect the similar movements in the interactions and thus the interactions themselves. We also utilize a well-known boosting algorithm to potentially increase the accuracy of the operator results. The techniques operate on the magnitudes of the acceleration and gyroscope readings to keep the analysis independent of the orientation of the sensors. The detection accuracy for six interactions using the approach presented in the paper range from 84.2% to 69.6%.

A Sign Language Recognition (SLR) system enables communication between hearing disabled individuals and those who can hear and speak. With the prevalence of the wearable computers, this technology is becoming an important human computer interface capable of reading hand gestures and inferring user;s intent. In this paper, we propose a real-time American SLR system leveraging fusion of surface electromyography (sEMG) and a wrist-worn inertial sensor at the feature level. A feature selection is provided for 40 most commonly used words and for four subjects. The experimental results show that after feature selection and conditioning, our system achieves 95.94% recognition rate. The results also illustrate the fusion of two modalities perform better than using only the inertial sensor. We observed that only one channel of sEMG (out of four) located on the wrist and under the wrist-watch is sufficient.

Human action recognition has a wide range of applications including biometrics, surveillance, and human computer interaction. The use of multimodal sensors for human action recognition is steadily increasing. However, there are limited publicly available datasets where depth camera and inertial sensor data are captured at the same time. This paper describes a freely available dataset, named UTD-MHAD, which consists of four temporally synchronized data modalities. These modalities include RGB videos, depth videos, skeleton positions, and inertial signals from a Kinect camera and a wearable inertial sensor for a comprehensive set of 27 human actions. Experimental results are provided to show how this database can be used to study fusion approaches that involve using both depth camera data and inertial sensor data. This public domain dataset is of benefit to multimodality research activities being conducted for human action recognition by various research groups.

This paper presents a computationally efficient method for action recognition from depth video sequences. It employs the so called depth motion maps (DMMs) from three projection views (front, side and top) to capture motion cues and uses local binary patterns (LBPs) to gain a compact feature representation. Two types of fusion consisting of feature-level fusion and decision-level fusion are considered. In the feature-level fusion, LBP features from three DMMs are merged before classification while in the decision-level fusion, a soft decision-fusion rule is used to combine the classification outcomes. The introduced method is evaluated on two standard datasets and is also compared with the existing methods. The results indicate that it outperforms the existing methods and is able to process depth video sequences in real-time.

Wireless health systems enable remote and continuous monitoring of individuals, with applications in elderly care support, chronic disease management, and preventive care. The underlying sensing platform provides constructs that consider the quality of information driven from the system and ensure the reliability/validity of the outcomes to support the decision-making processes. In this paper, we present an approach to integrate contextual information within the data processing flow in order to improve the quality of measurements. We focus on a pilot application that uses wearable motion sensors to calculate metabolic equivalent of task (MET) of exergaming movements. Exergames need to show energy expenditure values, often using accelerometer approximations applied to general activities. We focus on two contextual factors, namely “activity type” and “sensor location,” and demonstrate how these factors can be used to enhance the measured values, since allocating larger weights to more informative sensors can improve the final measurements. Further, designing regression models for each activity provides better results than any generalized model. Indeed, the averaged R2 value for the movements using simple sensor location improve from a general 0.71 to as high as 0.84 for an individual activity type. The different methods present a range of R2 value averages across activity type from 0.64 for sensor location to 0.89 for multidimensional regression, with an average game play MET value of 7.93. Finally, in a leaveone-subject-out cross validation, a mean absolute error of 2.231 METs is found when predicting the activity levels using the best models.

The articles in this special section are focused on two major aspects of Internet of things (IoT) technologies for smart and connected health services (SCH): 1) monitoring and assisting individuals by means of smart systems including sensors, devices, and robotics; and 2) creating interoperable digital health information infrastructures to increase medical/health information availability and use. The papers published in this SI provide evidence that SCH tools that rely upon IoT technologiescould significantly improve clinical outcomes and thequality of life of individuals undergoing monitoring.

This paper presents a fusion approach for improving human action recognition based on two differing modality sensors consisting of a depth camera and an inertial body sensor. Computationally efficient action features are extracted from depth images provided by the depth camera and from accelerometer signals provided by the inertial body sensor. These features consist of depth motion maps and statistical signal attributes. For action recognition, both feature-level fusion and decision-level fusion are examined by using a collaborative representation classifier. In the feature-level fusion, features generated from the two differing modality sensors are merged before classification, while in the decision-level fusion, the Dempster-Shafer theory is used to combine the classification outcomes from two classifiers, each corresponding to one sensor. The introduced fusion framework is evaluated using the Berkeley multimodal human action database. The results indicate that because of the complementary aspect of the data from these sensors, the introduced fusion approaches lead to 2% to 23% recognition rate improvements depending on the action over the situations when each sensor is used individually.

Inertial measurement units (IMUs) including accelerometers and gyroscopes are becoming very common and can be found in cell phones, fitness trackers, and other wearable devices. With the growth in wearable computing and body sensor networks, IMUs are also becoming more prevalent researchenvironments for estimation and tracking of human motion. Wedemonstrate that the accelerometer angle estimate is inaccurate for typical motions and present a method using a kinematic model to correct the accelerometer angle estimate and improve overall orientation estimates using an extended Kalman filter (EKF). Our method improves upon the raw accelerometer orientation estimation method and a gyroscope based EKF method for sensor orientation during motion and at rest.

Due to the low signal-to-noise ratio of electroencephalographic (EEG) recordings, the quality of the electrode-scalp contact is an important factor in EEG-based brain-computer interfaces (BCIs). For this reason, the impedance between each individual electrode and the scalp is measured prior to each EEG recording session. In order to obtain high quality EEG signals and accurate performance, the impedance has to be low (below 5K Ohms). Typically, researchers have reduced the electrode-scalp impedance by performing time-consuming electrode adjustments prior to the data acquisition stage. In this paper, we utilize the electrode-scalp impedance information to remove the EEG artifacts caused by high impedance electrodes in order to enhance the signal quality during the signal processing stage. Our proposed method is based on the independent component analysis (ICA) algorithm, which is used to decompose the EEG signals into independent components. The electrode-scalp impedance is employed to automatically distinguish irrelevant components from event-related components. The experimental results show that our method can effectively remove artifacts and enhance the BCI performance compared to the scenario where no artifacts were removed, and the scenario in which irrelevant independent components were removed manually based on prior knowledge.

Monitoring human movements using wireless wearable sensors finds applications in a variety of domains including healthcare and wellness. In these systems, sensory devices are tightly integrated with the human body and infer status of the user through signal and information processing. Typically, highly accurate observations can be made at the cost of deploying a sufficiently large number of sensors, which in turn results in increased energy consumption of the system and reduced adherence to using the system. Therefore, optimizing power consumption of the system while maintaining acceptable accuracy plays a crucial role in realizing these stringent resource constraint systems. In this paper, we present an activity monitoring approach that minimizes power consumption of the system subject to a lower bound on the classification accuracy. The system utilizes computationally simple template-matching blocks that perform classifications on individual sensor nodes. The system further employs a boosting approach to enhance accuracy of the distributed classifier by selecting a subset of sensors optimized in terms of power consumption and capable of achieving a given lower bound accuracy criterion. A proof-of-concept evaluation with three participants performing 14 transitional actions was conducted, where collected signals were segmented and labeled manually for each action. The results indicated that the proposed approach provides more than a 65% reduction in the power consumption of the signal processing, while maintaining 80% sensitivity in classifying human movements.

Summary form only given, as follows. Wireless Health brings to fruition many opportunities to continuously monitor human body with sensors placed on body or implanted in the body. These platforms will revolutionize many application domains including health care and wellness. They provide new avenues to continuously monitor individuals, whether it is intended to detect an early onset of a disease or to assess the effectiveness of the treatment. In the past few years, the community has observed a large number of wireless health applications that have been developed using wearable computers. Yet, not many have been deployed in a large scale. There are still several challenges that need to be addressed before realizing the ubiquitous use of wireless health systems. In this talk, we will highlight several applications of the wireless health and wearable computers. We will describe components of the wireless health computing systems and will outline challenges associated with their ubiquitous deployment. We will highlight current research efforts toward creating application specific self-powered architecture for Wireless Health and will discuss future directions.

Movement monitoring using wearable computers has been widely used in healthcare and wellness applications. To reduce the form factor of wearable nodes which is dominated by battery size, ultra-low power signal processing is crucial. In this paper, we propose an architecture that can be viewed as a hardware accelerator and employs dynamic time warping (DTW) in a hierarchical fashion. The proposed architecture removes events that are not of interest from the signal processing chain as early as possible, deactivating all remaining modules. We consider tunable parameters such as sampling frequency and bit resolution of the incoming sensor readings for DTW to balance the power consumption and classification precision trade-off. We formulate a methodology for determining the optimal set of tunable parameters and provide a solution using Active-set algorithm. We synthesized the architecture using 45nm CMOS and illustrated that a three-tiered module achieves 98% accuracy with a power budget of 1.23µW, while a single level DTW consumes 6.3µW with the same accuracy. We furthermore propose a fast approximation methodology that runs 3200 times faster while introducing less than 3% error over the original optimization for determining the total power consumption.

The primary aim of a Brain-Computer Interface (BCI) is to provide communication capabilities through brain signals recorded from the scalp for those with brain disorders to be able to interact with the outside world. In order to properly decode the electroencephalographic (EEG) brain signals, the BCI needs to adapt to the subject via calibration to ensure stable performance. One of the major challenges in realization of the EEG signals is the long calibration time required since they show significant variations between recording sessions even for the same subject within the same experimental condition. This paper proposes a score-based adaptive training algorithm that maximally utilizes relevant information from prior recording sessions and significantly shortens the calibration time. Also the proposed method is suitable to develop real-time, wearable, and low-power BCI embedded devices. The BCI developed in this work is based on the P300 word speller application introduced by Farwell and Donchin in 1988. The experimental results show that by employing few letters for calibration, the proposed adaptive training algorithm can achieve 100% classification accuracy.

A new delay-bounded asynchronous logic technique aimed at maximizing reliability at very low voltages is proposed. Compared to previous asynchronous logic approaches, the area and nominal delay overheads are small. Conventional standard cell libraries and conventional logic synthesis tools are used. The bounding delay elements used by the asynchronous controller feature programmable delays that are initially set based on static timing analysis. However, the delay elements are updated on-the-fly during actual operation of the circuit, resulting in strong resiliency even at low voltages and with extreme variations. Several benchmark circuits were implemented with the new asynchronous design flow using the 45nm TI process. Monte Carlo analysis demonstrates the expected resiliency. Compared to the equivalent synchronous circuits, the asynchronous versions have area overheads averaging 40%, although much smaller for large circuits. Nominal delay overheads average about 10%.

Advances in technology have led to the development of wearable sensing, computing, and communication devices that can be woven into the physical environment of our daily lives, enabling a large variety of new applications in several domains, including wellness and health care. Despite their tremendous potential to impact our lives, wearable health monitoring systems face a number of hurdles to become a reality. The enabling processors and architectures demand a large amount of energy, requiring sizable batteries. In this article, we propose a granular decision-making architecture for physical movement monitoring applications. The module can be viewed as a tiered wake-up circuitry. This decision-making module, in combination with a low-power microcontroller, allows for significant power saving through an ultra low-power processing architecture. The significant power saving is achieved by performing a preliminary ultra low-power signal processing, and hence, keeping the microcontroller off when the incoming signal is not of interest. The preliminary signal processing is performed by a set of special-purpose functional units, also called screening blocks, that implement template matching functions. We formulate and solve an optimization problem for selecting screening blocks such that the accuracy requirements of the signal processing are accommodated while the total power is minimized. Our experimental results on real data from wearable motion sensors show that the proposed algorithm achieves 63.2% energy saving while maintaining a sensitivity of 94.3% in recognizing transitional actions.

No abstract available.

Wireless body sensor networks (BSNs) possess enormous potential for changing people's daily lives. They can enhance many human-centered application domains such as m-Health, sport and wellness, and human-centered applications that involve physical/virtual social interactions. However, there are still challenging issues that limit their wide diffusion in real life: primarily, the programming complexity of these systems, due to the lack of high-level software abstractions, and the hardware constraints of wearable devices. In contrast with low-level programming and general-purpose middleware, domain-specific frameworks are an emerging programming paradigm designed to fulfill the lack of suitable BSN programming support with proper abstraction layers. This paper analyzes the most important requirements for an effective BSN-specific software framework, enabling efficient signal-processing applications. Specifically, we present signal processing in node environment (SPINE), an open-source programming framework, designed to support rapid and flexible prototyping and management of BSN applications. We describe how SPINE efficiently addresses the identified requirements while providing performance analysis on the most common hardware/software sensor platforms. We also report a few high-impact BSN applications that have been entirely implemented using SPINE to demonstrate practical examples of its effectiveness and flexibility. This development experience has notably led to the definition of a SPINE-based design methodology for BSN applications. Finally, lessons learned from the development of such applications and from feedback received by the SPINE community are discussed.

Brain Computer Interface (BCI) is gaining popularity due to recent advances in developing small and compact electronic technology and electrodes. Miniaturization and form factor reduction in particular are the key objectives for Body Sensor Networks (BSNs) and wearable systems that implement BCIs. More complex signal processing techniques have been developed in the past few years for BCI which create further challenges for form factor reduction. In this paper, we perform a computational profiling on signal processing tasks for a typical BCI system. We employ several common feature extraction techniques. We define a cost function based on the computational complexity for each feature dimension and present a sequential feature selection to explore the complexity versus the accuracy. We discuss the trade-offs between the computational cost and the accuracy of the system. This will be useful for emerging mobile, wearable and power-aware BCI systems where the computational complexity, the form factor, the size of the battery and the power consumption are of significant importance. We investigate adaptive algorithms that will adjust the computational complexity of the signal processing based on the amount of energy available, while guaranteeing that the accuracy is minimally compromised. We perform an analysis on a standard inhibition (Go/NoGo) task. We demonstrate while classification accuracy is reduced by 2%, compared to the best classification accuracy obtained, the computational complexity of the system can be reduced by more than 60%. Furthermore, we investigate the performance of our technique on real-time EEG signals provided by an eMotiv® device for a Push/No Push task.

Electroencephalography (EEG) is the recording of electrical activity along the scalp produced by the firing of neurons within the brain. These activities can be decoded by signal processing techniques, however, they are typically influenced by extraneous interference, like muscle movements, eye blinks, eye movements, background noise, etc. Therefore, a preprocessing step to remove artifacts is extremely important. This paper presents an effective artifact removal algorithm, based on Independent Component Analysis (ICA) and Hierarchical Clustering. Our technique utilizes general temporal and spectral features and particular information about target Event-Related Potentials (ERPs) (e.g. the timing of N200 and P300 on inhibition task or the specific electrodes contributing to the ERPs) to separate ERPs and artifact activities. Our method considers templates for desired ERPs to select event-related components for signal reconstruction. In our experimental study, we show that our proposed method can effectively enhance the ERPs for all fifteen subjects in the study, even for those that barely display ERPs in the raw recordings.

We present an immersive gaming demonstration using the minimum amount of wearable sensors. The game demonstrated is two-player tennis. We combine a virtual environment with real 3D representations of physical objects like the players and the tennis racquet (if available). The main objective of the game is to provide as real an experience of tennis as possible, while also being as less intrusive as possible. The game is played across a network, and this opens the possibility of two remote players playing a game together on a single virtual tennis pitch. The Microsoft Kinect sensors are used to obtain a 3D point cloud and a skeletal map representation of the player. This 3D point cloud is mapped on to the virtual tennis pitch. We also use a wireless wearable Attitude and Heading Reference System (AHRS) mote, which is strapped onto the wrist of the players. This mote gives us precise information about the movement (swing, rotation etc.) of the playing arm. This information along with the skeletal map is used to implement the physics of the game. Using this game we demonstrate our solutions for simultaneous data acquisition, 3D point-cloud mapping in a virtual space, use of the Kinect and AHRS sensors to calibrate real and virtual objects and for interaction of virtual objects with a 3D point cloud.

Recent years have witnessed a large influx of applications in the field of cyber-physical systems. An important class of these systems is body sensor networks (BSNs) where lightweight embedded processors and communication systems are tightly coupled with the human body. BSNs can provide researchers, care providers and clinicians access to tremendously valuable information extracted from data that are collected in users' natural environment. With this information, one can monitor the progression of a disease, identify its early onset, or simply assess user's wellness. One major obstacle is managing repositories that store the large amount of sensing data. To address this issue, we propose a data mining approach inspired by the experience in the areas of text and natural language processing. We represent sensor readings with a sequence of characters, called motion transcripts. Transcripts reduce complexity of the data significantly while maintaining morphological and structural properties of the physiological signals. To further take advantage of the physiological signal's structure, our data mining technique focuses on the characteristic transitions in the signals. These transitions are efficiently captured using the concept ofn-grams. To facilitate a lightweight and fast mining approach, we reduce the overwhelmingly large number ofn-grams via information gain (IG) feature selection. We report the effectiveness of the proposed approach in terms of the speed of mining while maintaining an acceptable accuracy in terms of the F-score combining both precision and recall.

One important application of body sensor networks is action recognition. Action recognition often implicitly requires partitioning sensor data into intervals, then labeling the partitions according to the action that each represents or as a non-action. The temporal partitioning stage is called segmentation, and the labeling is called classification. While many effective methods exist for classification, segmentation remains problematic. We present a technique inspired by continuous speech recognition that combines segmentation and classification using hidden Markov models. This technique is distributed across several sensor nodes. We show the results of this technique and the bandwidth savings over full data transmission.

A major objective in design of wearable and light-weight embedded systems is reducing the power consumption. This leads to reduction of the battery size and enhances the wear ability of the system. In this paper, we propose an ultra low power tiered wake-up architecture with signal processing capability. The signal processing is based on template matching and normalized cross correlation. The template matching at the beginning is performed with low sensitivity (with fewer bits and samples) but at very low power. Initial template matching removes signals that are obviously not of interest. If the signal is likely to be of interest, the sensitivity and the power consumption of the template matching blocks are gradually increased, until the signal of interest is detected with a reasonable confidence. Consequently, a microcontroller is activated for additional processing. The tunable parameters for template matching include the number of samples, the size of template (window size) and the number of bits per sample. The proposed architecture can enable the next generation of ultra low power or even battery less wearable and implantable computers due to tremendously reducing the power consumption of the signal processing. We estimate that the power consumption of the proposed tiered wake-up circuitry will be three to six orders of magnitude smaller than state-of-the-art low power microcontrollers, depending on the complexity of the template matching. Further, the proposed architecture provides high level of programmability which is lacking in ASIC architectures custom built for applications.

Enhancing the wear ability and reducing the form factor often are among the major objectives in design of wearable platforms. Power optimization techniques will significantly reduce the form factor and/or will prolong the time intervals between recharges. In this paper, we propose an ultra low power programmable architecture based on Dynamic Time Warping specifically designed for wearable inertial sensors. The low power architecture performs the signal processing merely as fast as the production rate for the inertial sensors, and further considers the minimum bit resolution and the number of samples that are just enough to detect the movement of interest. Our results show that the power consumption for inertial based monitoring systems can be reduced by at least three orders of magnitude using our proposed architecture compared to the state-of-the-art low power microcontrollers.

Wearable sensor nodes are highly constrained in terms of size, and, as a result, battery size and capacity. During a real time data collection, sensor nodes can communicate data continuously, however, this may reduce the system lifetime. Hence, we suggest an intelligent data collection algorithm that screens the sensor data and transmits only the segments of sensor data that might be of interest. Additionally, the proposed approach does not require extensive system training, since it is based on a flexible Body Sensor Network(BSN) data repository. User can select movements of interest in a repository, and load the sensor nodes with the relevant meta data. Based on the meta data, sensor nodes can decide whether a specific part of the collected sensor data needs to be transmitted. We extend the idea by exploring the idea of dynamically associating the relevance of the data to the amount of energy available at the sensor node.

Advances in technology have led to development of wearable sensing, computing and communication devices that can be woven into the physical environment of our daily lives, enabling a large variety of new applications in several domains including wellness and health care. Despite their tremendous potential to impact our lives, wearable health monitoring systems face a number of hurdles to become a reality. The enabling processors and architectures demand a large amount of energy, requiring sizable batteries. In this paper, we propose a granular decision making architecture that can be viewed as a tiered wake up circuitry. This module, in combination with a low-power microcontroller, enables an ultra low-power architecture. The significant power saving is achieved by performing a preliminary ultra low-power signal processing and hence, keeping the microcontroller off when the incoming signal is not of interest. The preliminary signal processing is performed by a set of special purpose functional units, also called screening blocks, that implements template matching functions. We formulate and solve an optimization problem to select screening blocks such that the accuracy requirements of the signal processing are accommodated while the total power is minimized. Our experimental results on real data from wearable motion sensors show that the proposed algorithm achieves 65.2% energy saving while maintaining 92.7% sensitivity in recognizing human movements.

Monitoring human activities using wearable sensor nodes has the potential to enable many useful applications for everyday situations. Limited computation, battery lifetime and communication bandwidth make efficient use of these platforms crucial. In this paper, we introduce a novel classification model that identifies physical movements from body-worn inertial sensors while taking collaborative nature and limited resources of the system into consideration. Our action recognition model uses a decision tree structure to minimize the number of nodes involved in classification of each action. The decision tree is constructed based on the quality of action recognition in individual nodes. A clustering technique is employed to group similar actions and measure quality of per-node identifications. We pose an optimization problem for finding a minimal set of sensor nodes contributing to the action recognition. We then prove that this problem is NP-hard and provide fast greedy algorithms to approximate the solution. Finally, we demonstrate the effectiveness of our distributed algorithm on data collected from five healthy subjects. In particular, our system achieves a 72.4% reduction in the number of active nodes while maintaining 93.3% classification accuracy.

Body sensor networks are emerging as a promising platform for remote human monitoring. With the aim of extracting bio-kinematic parameters from distributed body-worn sensors, these systems require collaboration of sensor nodes to obtain relevant information from an overwhelmingly large volume of data. Clearly, efficient data reduction techniques and distributed signal processing algorithms are needed. In this paper, we present a data processing technique that constructs motion transcripts from inertial sensors and identifies human movements by taking collaboration between the nodes into consideration. Transcripts of basic motions, called primitives, are built to reduce the complexity of the sensor data. This model leads to a distributed algorithm for segmentation and action recognition. We demonstrate the effectiveness of our framework using data collected from five normal subjects performing ten transitional movements. The results clearly illustrate the effectiveness of our framework. In particular, we obtain a classification accuracy of 84.13% with only one sensor node involved in the classification process.

Body sensor networks (BSNs) have proved their viability to greatly improve quality of medical care by providing continuous and in-home monitoring solutions. Highly constrained nature of the platform demands a design that efficiently utilizes limited resources of the system. Energy optimization techniques are especially desirable as the system lifetime is constrained by small batteries that power sensor nodes in a BSN. In this paper, we introduce a novel data-centering routing model to minimize communication energy, taking collaborative nature of signal processing for healthcare applications into consideration. Transmission energy for a path is determined as a compromise between the path length and the amount of data being transmitted along the path. Data produced by different nodes are aggregated to form packets of large size that consume smaller energy per bit. We formulate the problem as a minimum concave cost multicommodity flow problem and propose two approaches to find both optimal and approximate solutions. We evaluate performance of our energy minimization techniques on a variety of synthesized signal processing task graphs, as well as a real application for evaluating human postural control system. The results show an average of 35% energy saving with our proposed routing against a simple shortest path approach.

Monitoring human movements using wireless sensory devices promises to revolutionize the delivery of healthcare services. Such platforms use inertial information of their subjects for motion analysis. Potentially, each action or disease can be discovered by collaborative processing of sensor data from multiple locations on the body. This functionality is provided by a Body Sensor Network (BSN), which consists of several wireless sensor nodes positioned on different parts of the body. In spite of the revolutionary potential of this platform, power requirements and wearability have limited the commercialization of these systems. In this paper, we present an energy-efficient communication model for BSN applications which uses buffers to limit communication to short bursts, decreasing power usage and simplifying the communication. We formulate an optimization problem to reduce transmissions among sensor nodes and present an ILP-based solution and a fast greedy heuristic algorithm. We show that despite the decreased transmission efficiency, our greedy algorithm can be adopted for fast allocation of buffers in real-time. We experimentally compare the performance of both of the proposed approaches to the performance of an unbuffered system. Our results demonstrate that ILP and greedy solutions can reduce the amount of transmissions by an average factor of 70 and 41, respectively.

The evaluation of the postural control system (PCS) has applications in rehabilitation, sports medicine, gait analysis, fall detection, and diagnosis of many diseases associated with a reduction in balance ability. Standing involves significant muscle use to maintain balance, making standing balance a good indicator of the health of the PCS. Inertial sensor systems have been used to quantify standing balance by assessing displacement of the center of mass, resulting in several standardized measures. Electromyogram (EMG) sensors directly measure the muscle control signals. Despite strong evidence of the potential of muscle activity for balance evaluation, less study has been done on extracting unique features from EMG data that express balance abnormalities. In this paper, we present machine learning and statistical techniques to extract parameters from EMG sensors placed on the tibialis anterior and gastrocnemius muscles, which show a strong correlation to the standard parameters extracted from accelerometer data. This novel interpretation of the neuromuscular system provides a unique method of assessing human balance based on EMG signals. In order to verify the effectiveness of the introduced features in measuring postural sway, we conduct several classification tests that operate on the EMG features and predict significance of different balance measures.

Mobile sensor-based systems are emerging as promising platforms for healthcare monitoring. An important goal of these systems is to extract physiological information about the subject wearing the network. Such information can be used for life logging, quality of life measures, fall detection, extraction of contextual information, and many other applications. Data collected by these sensor nodes are overwhelming, and hence, an efficient data processing technique is essential. In this paper, we present a system using inexpensive, off-the-shelf inertial sensor nodes that constructs motion transcripts from biomedical signals and identifies movements by taking collaboration between the nodes into consideration. Transcripts are built of motion primitives and aim to reduce the complexity of the original data. We then label each primitive with a unique symbol and generate a sequence of symbols, known as motion template, representing a particular action. This model leads to a distributed algorithm for action recognition using edit distance with respect to motion templates. The algorithm reduces the number of active nodes during every classification decision. We present our results using data collected from five normal subjects performing transitional movements. The results clearly illustrate the effectiveness of our framework. In particular, we obtain a classification accuracy of 84.13% with only one sensor node involved in the classification process.

Body sensor networks are emerging as a promising platform for healthcare monitoring. These systems are composed of battery-operated embedded devices which process physiological data. The reduction in the power consumption is an important factor to increase the lifetime for such systems and to enhance their wearability through reducing the size of the battery. In this paper, we develop an energy-efficient communication scheme that uses buffers to reduce the number of transmissions among the sensor nodes constrained to limited hardware resources. A direct acyclic graph is used to model the information flow. We define a communication optimization problem and solve it using convex optimization techniques. We present results that support the efficiency of the proposed technique.

Advances in technology have led to the development of various light-weight sensor devices that can be woven into the physical environment of our daily lives. Such systems enable on-body and mobile health-care monitoring. Our interest particularly lies in the area of movement-monitoring platforms that operate with inertial sensors. In this paper, we introduce the notion of compatibility graphs and describe how they can be utilized for power optimization. We first formulate an action coverage problem that will consider the sensing coverage from a collaborative signal processing perspective. Our solution is capable of eliminating redundant sensor nodes while maintaining the quality of service. The problem we outline can be transformed into an NP-hard problem. Therefore, we propose an ILP formulation to attain a lower bound on the solution and a fast greedy technique. Moreover, we present a system for dynamically activating and deactivating sensor nodes in real time. We then use our graph representation to develop an efficient formulation for maximum lifetime. This formulation provides sufficient information for finding activation duties for each sensor node. Finally, we demonstrate the effectiveness of our techniques on data collected from several subjects.

This article presents an efficient technique for placement and routing of sensors/actuators and processing units in a grid network. The driver application that we present is a medical jacket, which requires an extremely high level of robustness and fault tolerance. The power consumption of such jacket is another key technological constraint. Our proposed interconnection network is a mesh of wires. A jacket made of fabric and wires would be susceptible to accidental damage via tears. By modeling the tears, we evaluate the probability of having failures on every segment of wires in our mesh interconnection network. Then, we study two problems of placement and routing in the sensor networks such that the fault tolerance is maximized while the power consumption is minimized. We develop efficient integer linear programming (ILP) formulations to address these problems and perform both placement and routing, simultaneously. This ensures that the solution is a lower bound for both problems. We evaluate the effectiveness of our proposed techniques on a variety of benchmarks.

Human movement models often divide movements into parts. In walking, the stride can be segmented into four different parts, and in golf and other sports, the swing is divided into sections based on the primary direction of motion. These parts are often divided based on key events, also called temporal parameters. When analyzing a movement, it is important to correctly locate these key events, and so automated techniques are needed. There exist many methods for dividing specific actions using data from specific sensors, but for new sensors or sensing positions, new techniques must be developed. We introduce a generic method for temporal parameter extraction called the hidden Markov event model based on hidden Markov models. Our method constrains the state structure to facilitate precise location of key events. This method can be quickly adapted to new movements and new sensors/sensor placements. Furthermore, it generalizes well to subjects not used for training. A multiobjective optimization technique using genetic algorithms is applied to decrease error and increase cross-subject generalizability. Further, collaborative techniques are explored. We validate this method on a walking dataset by using inertial sensors placed on various locations on a human body. Our technique is designed to be computationally complex for training, but computationally simple at runtime to allow deployment on resource-constrained sensor nodes.

Advances in technology have led to the development of various light-weight sensory devices that can be woven into the physical environment of our daily lives. Such systems enable on-body and mobile health-care monitoring. Our interest particularly lies in the area of movement monitoring platforms that operate with inertial sensors. In this paper, we propose a power optimization technique that will consider the sensing coverage problem from a collaborative signal processing perspective. We introduce compatibility graphs and describe how they can be utilized for power optimization. The problem we outline can be transformed into an NP-hard problem. Therefore, we propose an ILP formulation to attain a lower bound on the solution and a fast greedy technique. Along side this, we introduce a system for dynamically activating and deactivating sensor nodes in real-time. Finally, we elucidate the effectiveness of our techniques on data collected from several subjects.

Body sensor networks (BSNs) create an enormous opportunity to revolutionalize the way we learn, work, entertain and live today. A particularly promising application of BSNs is in health monitoring. Research indicates that various disorders in aging ranging from mild cognitive impairment to dementia and Alzheimer’s could be diagnosed early based on the study of locomotion. Sensor platforms integrated into clothing provide the possibility of reliable locomotion monitoring. In this work, we demonstrate a real-time wireless sensor system that quantitatively measures some of the factors involved in locomotion.

Monitoring human activities using wearable wireless sensor nodes has the potential to enable many useful applications for everyday situations. The deployment of a compact and computationally efficient grammatical representation of actions reduces the complexities involved in the detection and recognition of human behaviors in a distributed system. In this paper, we introduce a road map to a linguistic framework for the symbolic representation of inertial information for physical movement monitoring. Our method for creating phonetic descriptions consists of constructing primitives across the network and assigning certain primitives to each movement. Our technique exploits the notion of a decision tree to identify atomic actions corresponding to every given movement. We pose an optimization problem for the fast identification of primitives. We then prove that this problem is NP-Complete and provide a fast greedy algorithm to approximate the solution. Finally, we demonstrate the effectiveness of our phonetic model on data collected from three subjects.

Light-weight embedded systems are now gaining more popularity due to the recent technological advances in fabrication that have resulted in more powerful tiny processors with greater communication capabilities that pose various scientific challenges for researchers. Perhaps the most significant challenge is the energy consumption concern and reliability, mainly due to the small size of batteries. In this tutorial, we portray a brief description of low-power, light-weight embedded systems, depict several power profiling studies previously conducted, and present several research challenges that require low-power consumption in embedded systems. For each challenge, we highlight how low-power designs may enhance the overall performance of the system. Finally, we present a several techniques that minimize the power consumption in such systems.

Register sharing for high-level synthesis of programs represented in static single assignment (SSA) form is proven to have a polynomial-time solution. Register sharing is modeled as a graph-coloring problem. Although graph coloring is NP-Complete in the general case, an interference graph constructed for a program in SSA form probably belongs to the class of chordal graphs that have an optimal O(|V|+|E|) time algorithm. Chordal graph coloring reduces the number of registers allocated to the program by as much as 86% and 64.93% on average compared to linear scan register allocation.

This paper presents a theoretical framework that solves optimally and in polynomial time many open problems in time budgeting. The approach unifies a large class of existing time-management paradigms. Examples include time budgeting for maximizing total weighted delay relaxation, minimizing the maximum relaxation, and min-skew time budget distribution. The authors develop a combinatorial framework through which we prove that many of the time-management problems can be transformed into a min-cost flow problem instance. The methodology is applied to intellectual-property-based datapath synthesis targeting field-programmable gate arrays. The synthesis flow maps the input operations to parameterized library modules during which different time budgeting policies have been applied. The techniques always improve the area requirement of the implemented test benches and consistently outperform a widely used competitor. The experiments verify that combining fairness and maximization objectives improves the results further as compared with pure maximum budgeting. The combined fairness and maximization objective improves the area by 25.8% and 28.7% in slice and LUT counts, respectively.

Tiny embedded systems have not been an ideal outfit for high performance computing due to their constrained resources-limitations in processing power, battery life, communication bandwidth, and memory constrain the applicability of existing complex medical analysis algorithms such as the electrocardiogram (ECG) analysis. Among various limitations, battery lifetime has been a major key technological constraint. In this paper, we address the issue of partitioning such a complex algorithm while the energy consumption due to wireless transmission is minimized. ECG analysis algorithms normally consist of preprocessing, pattern recognition, and classification. Considering the orientation of the ECG leads, we devise a technique to perform preprocessing and pattern recognition locally in small embedded systems attached to the leads. The features detected in the pattern recognition phase are considered for the classification. Ideally, if the features detected for each heartbeat reside in a single processing node, the transmission will be unnecessary. Otherwise, to perform classification, the features must be gathered on a local node and, thus, the communication is inevitable. We perform such a feature grouping by modeling the problem as a hypergraph and applying partitioning schemes which yield a significant power saving in wireless communications. Furthermore, we utilize dynamic reconfiguration by software module migration. This technique, with respect to partitioning, enhances the overall power saving in such systems. Moreover, it adaptively alters the system configuration in various environments and on different patients. We evaluate the effectiveness of our proposed techniques on MIT/BIH benchmarks and, on average, achieve 70 percent energy saving

Unlike their hard real-time counterparts, soft real-time applications are only expected to guarantee their "expected delay" over input data space. This paradigm shift calls for customized statistical design techniques to replace the conventional pessimistic worst case analysis methodologies. We present a novel statistical time-budgeting algorithm to translate the application expected delay constraint into its components' local delay constraints. We utilize the mathematical properties of the problem to quickly calculate the system expected delay and incrementally estimate the component utility variation with its timing relaxation. Our algorithm determines the optimal maximum weighted timing relaxation of an application under expected delay constraint. Experimental results on core-based synthesis of several multimedia applications targeting field-programmable gate arrays show that our technique always improves the design area. Furthermore, it consistently outperforms optimal time budgeting under hard real-time constraint, which is the best existing competitor. Design area improvements were up to 26% and averaged about 17% on several MediaBench applications.

We propose a platform for health monitoring using wireless sensor networks. Our platform is a new architecture called CustoMed that will reduce the customization and reconfiguration time for medical systems that use reconfigurable embedded systems. This architecture is a network enabled system that supports various wearable sensors and contains on-board general computing capabilities for executing individually tailored event detection, alerts, and network communication with various medical informatics services. The customization of such system with a large number of "med nodes" is extremely fast even by non-engineering staff. In this paper, we present the architecture of such device along with experimental analysis that evaluates the performance of such system.

This paper presents an efficient technique for placement and routing of sensors/actuators and processing units in a grid network. The driver application that we present is a medical jacket which requires an extremely high level of robustness and fault tolerance. The power consumption of such jacket is another key technological constraint. Our proposed interconnection network is a mesh of wires. A jacket made of fabric and wires would be susceptible to accidental damage via tears. By modeling the tears, we evaluate the probability of having failures on every segment of wires in our mesh interconnection network. Then we study two problems of placement and routing in the sensor networks such that the fault tolerance is maximized while the power consumption is minimized. We develop efficient integer linear programming (ILP) formulations to address these problems and perform both placement and routing simultaneously. This ensures that the solution is a lower bound for both problems. We evaluate the effectiveness of our proposed techniques on a variety of benchmarks.

We propose the use of multiple gateways to significantly reduce latency and energy consumption in multi-hop wireless sensor networks during data aggregation. We have derived efficient integer linear programming formulations as well as a novel negative selection statistically-tuned heuristics. The heuristics are based on newly developed relaxation based lower bounds that are also used to quantify the effectiveness of the proposed heuristics. Our simulation study indicates that the use of gateways can often reduce latency and energy consumption by several times.

In this paper we discuss the global resource sharing problem during synthesis of control data flow graphs for FPGAs. We first define the Global Resource Sharing (GRS) problem. Then, we introduce the Global Inter Basic Block Resource Sharing (GIBBS) technique to solve the GRS problem. We developed five heuristics to solve the GRS problem. The first tries to minimize the number of connections between modules, the second considers the area gain, the third uses the criticality of operations assigned to resources as a measure for deciding on merging any given pair of resources, the fourth tries to capture common resource chains and overlap those to minimize both area and delay, and the fifth is the combination of these heuristics. While applying resource sharing, we also consider the execution frequency of the basic blocks. Using our techniques we synthesized several CDFGs representing applications from MediaBench suite. Our results show that, we can reduce the total area requirement by 44% on average (up to 59%) while increasing the execution time by 6% on average.