Recent Advances in Wearable Sensing Technologies

Download Recent Advances in Wearable Sensing Technologies

Preview text

Recent Advances in Wearable Sensing Technologies
Alfredo J. Perez 1,*,† and Sherali Zeadally 2,*,†
1 TSYS School of Computer Science, Columbus State University, Columbus, GA 31909, USA 2 College of Communication and Information, University of Kentucky, Lexington, KY 40506, USA * Correspondence: [email protected] (A.J.P.); [email protected] (S.Z.) † These authors contributed equally to this work.
Abstract: Wearable sensing technologies are having a worldwide impact on the creation of novel business opportunities and application services that are benefiting the common citizen. By using these technologies, people have transformed the way they live, interact with each other and their surroundings, their daily routines, and how they monitor their health conditions. We review recent advances in the area of wearable sensing technologies, focusing on aspects such as sensor technologies, communication infrastructures, service infrastructures, security, and privacy. We also review the use of consumer wearables during the coronavirus disease 19 (COVID-19) pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and we discuss open challenges that must be addressed to further improve the efficacy of wearable sensing systems in the future.
Keywords: wearables; smartphones; sensing; fitness; mobile payments; financial technology; m-health; crowdsensing; Internet of Things; security; privacy; energy; COVID-19; SARS-CoV-2

Citation: Perez, A.J.; Zeadally, S. Recent Advances in Wearable Sensing Technologies . Sensors 2021, 21, 6828.
Academic Editor: Paolo Visconti
Received: 19 September 2021 Accepted: 6 October 2021 Published: 14 October 2021

1. Introduction
Wearable sensing technologies continue to improve rapidly with advances in sensors, communication technologies, and artificial intelligence (AI) in the past decade. Research and development in wearable sensing technologies are fueling a revolution in the creation of novel services in gaming, fitness, entertainment, and specialized applications in industries such as healthcare, security, and defense, among others. In 2020, the market for wearable devices was USD 80 billion, which has tripled in terms of annual revenue since 2014 and it is expected to reach USD138 billion by 2025 [1]. In the consumer wearables market, in 2019, smartwatches and wristbands dominated the market with a combined market share of 51%; as of 2021, the leading wearables are ear-worn wearables with a market share of 48%, followed by a 37% combined market share of smartwatches and wristbands [2]. Figure 1 presents the consumer wearable devices’ market share by device type (2019–2022).

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// 4.0/).

Figure 1. Consumer wearables device market share (2019–2022).

Sensors 2021, 21, 6828.

Sensors 2021, 21, 6828

2 of 34
Ear-worn wearables, in special hearables such as true wireless stereo (TWS) wearables, have surged from almost zero market share to a significant share of the wearable device market [1] since the introduction of Apple AirPods in 2016, and have significantly increased during the coronavirus disease (COVID-19) pandemic [3], caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), as many people have worked and studied from their homes worldwide. During the pandemic, smart reusable masks that can detect SARSCoV-2 and self-sterilize have become an active area of research and development [4–8]. The COVID-19 pandemic has also positively impacted the adoption of other consumer wearable technologies for mobile payment systems, patient tracking, contact tracing, and remote patient monitoring and treatment [9–12]. Combined fitness/medical-connected services was the leading market for wearable sensing technologies as of 2020 [13–15]. Other markets such as industrial wearables services, entertainment/gaming (i.e., augmented reality (AR) games and devices), and wearables for defense and security are also surging in popularity with recent technological advances in wearable technologies.
According to latest market research analysis, by 2025, the wearable payments services market (around USD 72 billion by 2025) is expected to be larger than the combined fitness/medical wearables services market by approximately USD10 billion [13–15]. The wearable payments market has grown due to the adoption of near-field communication (NFC) in smartphones by manufacturers supporting financial payment standards [16–18], and the incorporation in the near future of NFC in new generations of smartwatches, fitness trackers, and other wearables such as smart rings [19]. However, by 2028, it is forecasted that the wearable fitness market will be approximately USD 138.7 billion [20], while the wearable payment services will remain around USD 80 billion [21]. Figure 2 presents the market value of wearable services for the years 2018–2020, and a projection for 2025 based on available market research data [13–15,22].

Figure 2. Wearable services market value.
Consumer wearable sensing systems were initially researched with cellphones and smartphones during the second half of the 2000s. During that time, the widespread adoption of cellular communication in the world, the mobile Internet, and the embedding of sensors in cellphones such as location sensors, accelerometers, and cameras paved the way to the development of sensing applications (in particular applications related to humancentric activities) in urban environments at a low cost compared with the deployment of static wireless sensor networks (WSNs) to achieve the same human-centric sensing goals [23]. The research in this area led to the development of many applications in the context of participatory and crowdsensing systems [24,25] using not only embedded cellphone sensors, but using external sensors connected via Bluetooth. Table 1 presents a summary of related works in mobile and wearable sensing during the past decade.

Sensors 2021, 21, 6828

3 of 34

Table 1. Summary of survey works in mobile and wearable sensing.







A survey of mobile phone sensing

Review of applications and architectures for smartphone sensing in human-centric and participatory sensing systems. No mention of wearables

A survey on privacy in mobile participatory sensing Review of privacy mechanisms for




smartphone-based crowdsensing systems. No

mention of wearables.

A survey on human activity recognition using

Review of machine learning (ML) models to



wearable sensors

classify activities using wearables. Review does

not include deep learning (DL) systems.



Mobile phone sensing systems: A survey

Review of mobile-smartphone-based sensing applications in participatory/crowdsensing settings. Mentions two systems that, as of 2012,used electrocardiogram (ECG) sensors.



Mobile sensing systems

Review of mobile sensing systems based on smartphones and their communication architectures. Provides short review on security.

Wearables: Fundamentals, advancements, and a

Review of wearable technology as of 2014 with a



roadmap for the future

focus on sensors and applications. Does not

review security or privacy issues.

Review of monetary and nonmonetary

A survey of incentive techniques for mobile crowd

incentives mechanisms for mobile crowdsensing




systems based on smartphones. Incentives are

important in crowdsensing to recruit

participants to collect data.

Reviews energy-aware security mechanisms for



A survey on energy-aware security mechanisms

WSNs, mobile devices (focus on smartphones),

and network nodes as of 2015. Review does not

mention wearables.

Pervasive eHealth services a security and privacy risk Presents risk awareness and perception for



awareness survey

eHealth wearables using Amazon Mechanical


Review of application-specific and



Incentive mechanisms for participatory sensing:

general-purpose incentive mechanisms for

Survey and research challenges

mobile crowdsensing systems based on




Deep, convolutional, and recurrent models for human Reviews and evaluates of deep learning methods

activity recognition using wearables

for human activity recognition.



A survey of wearable devices and challenges

Review focuses on consumer wearables available as of 2017. Work also addresses security, power, task offloading, and machine learning. Work does not address privacy issues.



The use of wearables in healthcare–challenges and

Reviews applications of wearables in healthcare


from the application perspective.

Review of wearables available as of 2017 in the



A survey on smart wearables in the application of

context of fitness. Work does not address


security, privacy, power, or ML in wearable




Mobile payment systems: secure network

architectures and protocols

Describes architectures and protocols to enable mobile payments. From the device perspective, it focuses on mobile phones. No mention of wearables.

Sensors 2021, 21, 6828

4 of 34

References [38] [39] [40] [41] [42] [43] [44] [45] [46] [9] [47] [48]

Year 2018 2018 2018 2019 2019 2020 2020 2020 2020 2020 2021 2021

Table 1. Cont.

Privacy issues and solutions for consumer wearables
A critical review of consumer wearables, mobile applications, and equipment for providing biofeedback, monitoring stress, and sleep in physically active populations
Wearables and the medical revolution
Demystifying IoT security: An exhaustive survey on IoT vulnerabilities and a first empirical look on Internet-scale IoT exploitations Buddy’s wearable is not your buddy: Privacy implications of pet wearables
Design architectures for energy harvesting in the Internet of Things
A comprehensive overview of smart wearables: The state of the art literature, recent advances, and future challenges Use of wearable sensor technology in gait, balance, and range of motion analysis Wearables and the Internet of Things (IoT), applications, opportunities, and challenges: A Survey
A survey of COVID-19 contact tracing apps
Wearables for Industrial Work Safety: A Survey
A survey on wearable technology: History, state-of-the-art and current challenges

Review of privacy issues in consumer wearables. Work does not address power or machine learning.
Review of the utilization of consumer wearables for stress and sleep monitoring. No privacy or security issues mentioned in the paper.
Reviews the utilization of wearables for medical use (m-Health). No privacy or security issues reviewed in the paper.
Review of security issues and solutions in Internet of Things (IoT) systems. Review does not mention wearables.
Review of privacy issues and possible privacy violations or privacy leakages to owners of pets (pet parents) by having their pets use wearables.
Reviews power and energy harvesting techniques for Internet of Things (IoT) devices including wearable devices.
Bibliographic review of published works related to wearable devices. This work reviews published works from 2010 to 2019 (before the COVID-19 pandemic).
Review of wearables and ML systems with a focus on gait analysis.
Review of sensors and applications of wearables before the COVID-19 pandemic. This work does not review security, privacy, or ML.
Reviews contact tracing apps developed during the COVID-19 pandemic.
Review of wearables in the context of industrial settings. Work focuses on applications of wearables for industry.
Review provides a comprehensive historical review of wearables devices. Reports on applications and some aspects of security and privacy.

Most of the works cited in Table 1 addressed specific aspects of mobile and wearable sensing systems, with many works focusing on smartphone-based sensing/crowdsensing systems in the past decade. In this work, we present a comprehensive review to provide the reader with not only a summary of past works but also new opportunities in wearables. Moreover, the unexpected COVID-19 pandemic has brought to the spotlight the use of wearable sensing technologies, and has positively shifted the perception and adoption of wearable technologies despite their privacy and security issues. Thus, while recent advancements in wearable sensing technologies have paved the way for the emergence of a plethora of services we are currently using in our lives, there are several areas still in need of further research. In this work, we describe current advances in wearable sensing technologies and services, and their use and opportunities to continue moving the field forward. The main contributions of this paper are as follows:

Sensors 2021, 21, 6828

5 of 34
• We present a comprehensive review of current advances in wearable sensing technologies;
• We describe recent developments in communication, services, security, and privacy technologies for wearables;
• We discuss some research opportunities and challenges that we need to address in the future for wearable sensing technologies.
We organize the rest of the paper as follows: In Section 2, we review the hardware architecture of wearable sensing devices. Section 3 presents communication technologies for wearable sensing. In Section 4, we discuss remote services for wearable sensing. Section 5 reviews security and privacy challenges and solutions for wearable sensing devices. In Section 6, we present challenges and opportunities in wearable sensing. Finally, Section 7 presents concluding remarks. Figure 3 presents the organization of this work.

Figure 3. Paper organization.
2. Wearable Sensing Technologies
A wearable sensing device is a device that consists of sensors, actuators/output devices, a power generating unit, and an embedded computer, which may be implanted, worn, or carried around by a user [29,38]. This user may be a person or, in the case of some wearables, worn by animals. Depending on the characteristics and sensing goals of a wearable sensing device, it may be connected to external systems using the Internet either through a cellular network and/or wireless local area network (WLAN). External systems can store and conduct analysis using artificial intelligence (AI) techniques and may provide feedback to the user of the device. While the ubiquity of wireless sensing technologies has dramatically increased in recent years, early utilization of wearable sensing devices dates back decades ago [29,48]. As of 2021, there are at least 266 companies producing at least 430 wearable sensing devices [49] that can be categorized in a taxonomy based on three layers that include: market type, intrusiveness, and body location.
The first layer (market type) determines the ease with which a general user can typically access a wearable sensor device. Based on the market, devices can be grouped into consumer wearable sensing devices/systems and specialized wearables. Consumer wearable sensing devices can be further categorized into fitness, entertainment/gaming, security, or pet use [42,50]. Specialized wearables can only be acquired through specialized vendors, and they comply with special standards or may be regulated by laws that specify who may acquire and/or use them and the specific purposes for which each is designed. Thus, we can categorize specialized wearable sensing devices into industrial, healthcare/medical, security/defense, and research fields.
The second layer (intrusiveness level) determines whether the wearable can be implanted/placed into the body of a living organism (implantable), placed on/worn by a living organism (non-implantable), or carried by a user, for example, on a backpack (external). Under this classification, ingestibles would be classified as an implantable device [51]. The difference between a nonimplantable and an external wearable is whether the device

Sensors 2021, 21, 6828

6 of 34
is directly in contact with the body of the user (nonimplantable) or not (external). Figure 4 illustrates sensors based on intrusiveness level.

Figure 4. Wearable sensors based on intrusiveness level.
The third layer (body location) determines the placement of the wearable sensing device, which can be the head, trunk, arm, or leg. It is worth noting that these are general positions on the body of a user, so when we refer to the head, this location may include the neck, ears, or eyes. Thus, an example is a consumer, nonimplantable wearable device that can be worn on a wrist (e.g., a fitness wristband).
A wearable sensor device may be composed of the following components depending on its objectives and functionalities (as shown in Figure 5):
• A power unit. This component of the wearable sensor device provides the energy used by the wearable sensor device to operate. Some wearable devices may include rechargeable or nonrechargeable batteries and energy-harvesting technologies [43]. Table 2 presents different types of power-generating units that can be used.
• Sensors. These are electronic and microelectro-mechanical systems (MEMS) components that measure a physical quantity on the user (human-centric) or their surrounding environment (environmental). These sensors may be intrusive to the user (e.g., implanted in the body), with part of the wearable device worn by the user (e.g., smart fabrics [52] and photoplethysmography (PPG) sensors [53]), or carried around/worn by the user (e.g., location trackers [54]). Figure 6 shows a Venn diagram with wearable sensors grouped by type, and Table 3 describes each sensor.
• Processing/control unit. Based on the capabilities and/or design/objectives of the wearable, this component may perform basic calculations, filter data, or execute AI algorithms or control algorithms.
• Embedded storage media. Some wearable sensing devices have a flash-type storage media that stores sensor data for further analysis.
• Network interfaces. Using communication interfaces, a wearable sensor device may create a personal area network (PAN) with other wearable sensors, to communicate with a more powerful device such as a smart phone, or to directly forward data to a remote service.
• Actuators. Actuator components produce vibrations, sound, and visual cues (e.g., lights, screens, or heads-up displays) to notify the user about the device’s status. Some wearable sensing devices may not send data to a remote server/service, but they may provide automated feedback or execute an intrusive action on the user (e.g., an automatic defibrillator [55] and wearables for automated medication delivery [56,57] using microneedles) without the need for external systems, and some wearables provide information on a smartphone screen.

Sensors 2021, 21, 6828

7 of 34
If smartphones are classified as external wearable sensing devices based on intrusiveness, as of 2021, the most-used wearable sensors (from those presented in Table 3) were the sensors embedded in most smartphones. These sensors are the microphone, location sensors, CMOS/CCD camera sensors, accelerometers, gyroscopes, and, to a lesser extent, the NFC interface as a contactless payment sensor. According to the 2021 Ericsson mobility report [58], as there are 5.5 billion smartphones in the world, there are 5.5 billion microphones, 5.5 billion location sensors, 5.5 billion CMOS/CCD cameras, and 5.5 billion accelerometers and gyroscopes collecting data in the world. If considered a wearable, the smartphone would be the most-used wearable during the COVID-19 pandemic caused by SARS-CoV-2. As these sensors are commonly available in most if not all smartphones, the NFC interface with 2.2 billion NFC-enabled smartphones/smartphone-like devices (e.g., tablets) [59] is next.
If the smartphone is not considered a wearable, then the most-available wearable sensors are accelerometers embedded in 708 million smartwatches and activity tracker units shipped between 2018 and 2021 (projection) [60]. However, if it is assumed that all true wireless stereo (TWS) hearables have microphones, then there would be 709 million microphones shipped as part of the TWSs sold between 2018 and 2021 (projection) [61]. After these sensors, the most commonly used sensor is the photoplethysmography(PPG) sensor available in many smartwatches, activity trackers, and pulse oximeters [62].

Figure 5. Typical components of a wearable sensing device. The red dotted line indicates possible connection.

Figure 6. Typical sensors available in wearable devices grouped by type of collected data.

Sensors 2021, 21, 6828

8 of 34

Table 2. Energy sources for wearable sensing devices.

Energy Source Nonrechargeable batteries Rechargeable batteries Solar-powered Radiofrequency (RF)
Movement and mechanical waves
Thermoelectric generators

Use of standard-size small or specialized-size batteries that power a wearable sensing device
Lithium ion batteries that may be connected to an external power source to be recharged
Use of photovoltaic (PV) cells to recharge a battery that powers a wearable
Use of antennas that extract energy from radio signals to recharge a battery or to power directly a wearable sensor
Use of piezoelectric devices to extract energy from human movements [68] or mechanical waves such as wind or ultrasound to recharge a battery or to power a device [69]
Use of body heat to generate power to recharge a battery or to power directly a wearable sensor [71]

Examples of Wearable Sensing Devices
Insulin pumps, cochlear implants/devices, implantable cardioverter defibrillators
Smart watches, smart phones, heart trackers, insulin pumps, digital stethoscopes [63], portable handheld ultrasound diagnostic devices [64]
Smart bracelets [65], smart watches, external wearables such as tracking devices, smart fabrics
Radiofrequency identification (RFID) implants [66], bioelectronic stickers/tattoos [67]
Implantable medical devices [69], wrist wearables [70]
Biometric wearables and smart t-shirts for electrocardiogram monitoring [72]

Table 3. Sensor technologies for wearable sensing devices.

Sensor Type
Smart fabrics (e-textiles)
Electrocardiogram (ECG) sensor
Near-field communication (NFC) Galvanic skin response (GSR) sensor

Fabrics developed from traditional materials (e.g., cotton, polyester, nylon) combined with materials possessing electrical conductivity, or that can be embedded/uses to carry other sensors/electronic components. Some smart fabrics can detect the presence of chemical substances [73]
Measures the electrical impulses of the heart muscle. Usually placed in contact with the skin. May be used in conjunction with implantable cardioverter defibrillators. Provides heart pulse data
Enables communication at short distances (less than 10 cm). Used as a wearable payment sensor [74,75]. Can be used to detect proximity and infer location, and for multiple-factor authentication methods [76].
Measures skin conductivity. Used in wearables to recognize stress levels/emotional state of an individual [77].

Wearable Device Examples
Zephyr compression shirt, Nadi X smart yoga pants
Shimmer3 ECG chest unit, Apple Watch Series 6
NFC Ring, many smartphones, smartwaches
Empatica E4 wristband

Type of Collected Data Human-centric
Human-centric Human-centric

Sensors 2021, 21, 6828

9 of 34

Table 3. Cont.

Sensor Type
Photoplethysmography (PPG) sensor
Electroencephalography (EEG) sensors
Glucose monitors Infrared (IR) sensor Accelerometer/gyroscope Microphone
Location sensor
Complementary Metal-Oxide Semiconductor (CMOS)/CCD imaging sensor
Radiofrequency identification (RFID) tags

Measures blood volume changes. These sensors illuminate the skin of a wearer and measure light absorption to determine human body variables including heart rate [78,79], blood oxygenation levels [80], and blood pressure when used in conjunction with an ECG sensor [81].
Measure electrical activity in the scalp of a user. These devices can be used to diagnose abnormal brain activity when used in healthcare applications [82] or to control devices through brain–computer interfaces (BCIs) [83].
Monitor blood glucose levels for people with diabetes. Devices can monitor glucose levels continuously or at a single moment in time [84].
Measures skin or ambient temperature. Temperature can be used to predict ovulation in female mammals.
Detects sudden accelerationmovement. Accelerometers can be used to detect and characterize human activities [85].
Detects sound. They can be used to detect health conditions, ambient sounds, activity, location contexts (e.g., being in a restaurant, hospital, home) [86].
Tracks the locations/places where a user carrying a device with location may be [87]. Location sensors may be outdoor location or indoor location sensors and include technologies such as a global positioning system (GPS; United States), Galileo (European Space Agency), GLONASS (Russia), BeiDou (China) receivers, or the Navigation with Indian Constellation (India) systems. Indoor location technologies/sensors may include sonar-based, dead reckoning, Bluetooth low energy (BLE) beacons, among others [88].
Takes photographs. When combined with AI, it may be used to detect objects and possibly recognize people’s identities without consent [89]. May be used to detect emotions in humans.
Store information about its wearer. RFID can be active or passive and can be used to track assets [90]. RFID can be used for location-based systems and to estimate crowd size in crowd-management systems [91].

Wearable Device Examples
Wellvue O2 Ring, pulse oximeters, most fitness bands and smart watches
Emotiv EpocX
Dexcom G6 CGM
Ava fertility tracker
Shimmer3 IMU, Samsung Galaxy Watch 3, activity trackers, smartphones Eko CORE family of stethoscopes/stethoscope attachments
Game Golf GPS receiver, Jiobit, Pet tracker, smartphones, most smartwatches
Iristick, Ray-Ban/Facebook Stories smart glasses, H1 head-mounted smart glasses, Microsoft HoloLens, Axon Body 2 body cameras, smartphones,
3M RFID tags, ARDES Injection needle with RFID chip for cats and dogs, smartphones

Type of Collected Data
Human-centric Humancentric/environmental Human-centric/ Environmental Humancentric/Environmental

Sensors 2021, 21, 6828
Sensor Type Laser emitter
Ultrasound sensor Air quality sensor Spectrometer
Radiation sensor Barometric pressure sensor Compass

10 of 34

Table 3. Cont.

Laser emitters are used to measure distances through light detection and ranging (LiDAR) and there are plans to be integrate them in future augmented reality (AR) glasses and smartphones [92]. A laser emitter can also be used for both acute and chronic pain management [93]. Detects objects in the proximity of a user/device [94]. Used also as an imaging sensor in handheld healthcare medical devices [64]. Detects harmful gas concentrations/volatile components [95]. Separates and measures the spectral components reflected by a material. The light spectrum can be used to determine the components of the material [96].
Tracks ionizing [97] and nonionizing [98] radiation in the proximity of its wearer.
Detects barometric (atmospheric) pressure. Can be used to detect movement, activity [99,100], and altitude. Determines orientation and used for navigation

Wearable Device Examples
CuraviPlus Laser Therapy Belt for Lower Back Pain, future smart AR glasses and smartphones
WeWALK smart cane, UltraCane, SonoQue, and Clarius portable handheld ultrasound devices Atmotube PRO, TZOA, Flow 2 by plume labs
GoyaLab IndiGo modular visible spectrometer
Instadose 2 Personal Radiation (X-ray) badge, Landauer RaySafe i3 Real-time Personal, Radiation Dosimetry, Landauee Tactical RadWatch
Garmin Fenix 5X
Most smartwatches

Type of Collected Data
Humancentric/Environmental Environmental Environmental
Environmental Environmental

3. Communication Technologies for Wearable Sensing
Advances in communication technologies support the current generation of wearable sensing services. From improvements in intrabody, body area networks (BANs), and personal area networks (PANs) to worldwide deployments of broadband wireless network connectivity, and computing paradigms such as cloud computing and blockchain, communication technologies are supporting many services that use wearable sensing technologies to deliver and provide value-added services to their users.
Figure 7 shows a general architecture for a wearable sensing system. In this architecture, wearable sensor devices collect data and conduct filtering or execute basic data analysis and/or models trained using machine learning (ML) algorithms [23]. Some wearable sensing devices may connect to other sensors (intrabody area networks) or to a smartphone using BANs or PANs. At some point, and based on the design or features of the wearable sensing device, the latter may forward the data collected to a remote service using a cellular network or WLAN either directly connected to the Internet or via a smartphone or communication hub that serves as a gateway device for the wearable device. Depending of the application, specialized networks such as tactical communication networks and satellite communication may be used.
Today, technological advances in communication technologies are found in intrabody networks, BANs, and PANs. These networking technologies are used in wearable sensing to connect wearable sensor devices amongst themselves and to other devices such as a smartphone, a communication hub, or actuator devices over a short distance [101].

Preparing to load PDF file. please wait...

0 of 0
Recent Advances in Wearable Sensing Technologies