Enabling Technologies for Wireless Body Area … · A wireless body area network (WBAN) is a radio...

IEEE Communications Magazine • December 200984 0163-6804/09/$25.00 © 2009 IEEE

1 There is no strict differ-ence between WBAN andWPAN in their defini-tions. In this articleWBAN refers to a networkof wireless devices in oron a human body, whileWPAN refers to a networkof wireless peripherals inproximity to a person.

INTRODUCTION

With the growing needs in ubiquitous communi-cations and recent advances in very-low-powerwireless technologies, there has been consider-able interest in the development and applicationof wireless networks around humans. A wirelessbody area network (WBAN) is a radio frequen-cy (RF)-based wireless networking technologythat interconnects tiny nodes with sensor oractuator capabilities in, on, or around a humanbody. Typically, the transmissions of these nodescover a short range of about 2 m. Complement-ing wireless personal area networks (WPANs),1in which radio coverage is usually about 10 m,WBANs target diverse applications includinghealthcare, athletic training, workplace safety,consumer electronics, secure authentication,and safeguarding of uniformed personnel. AWBAN can also be connected to local and widearea networks by various wired and wirelesscommunication technologies, as illustrated inFig. 1.

WBANs will play an important role inenabling ubiquitous communications, creating ahuge potential market. In the area of healthcare,according to the World Health Organization’sstatistics, millions of people suffer from obesityor chronic diseases every day, while the agingpopulation is becoming a significant problem.Both the current situation and future trend callfor new technologies such as WBANs to facili-

tate first-hand health monitoring and medicalcare (point of care). From the consumer elec-tronics perspective, short-range wireless tech-nologies for human-computer interaction (HCI)and entertainment are booming. Take BluetoothLow Energy technology as an example; a recentreport predicts the initial market volume ofthose ultra-low-power products to be in the bil-lions.

Unlike conventional wireless sensor networks(WSNs), WBANs have their own characteristics,as discussed below, which distinguish them fromWSNs and also create new technical challenges.

Architecture: A WBAN consists of two cate-gories of nodes: sensors/actuators in or on ahuman body, and router nodes around WBANwearers or second-tier radio devices equippedon the wearers, functioning as an infrastructurefor relaying data. In WSNs, however, every nodefunctions as a sensor node as well as a routernode.

Density: The number of sensors/actuatorsdeployed on the wearer depends on use cases.Typically, they are not deployed with high redun-dancy to tolerate node failures as in convention-al WSNs, and thus do not require high nodedensity.

Data rate: Most WSNs are applied for event-based monitoring, where events can happenirregularly. In contrast, WBANs are employedfor monitoring human physiological activities,which vary in a more periodic manner. As aresult, the application data streams exhibit rela-tively stable rates. Typical WBAN sensors aresummarized later.

Latency: For both healthcare and consumerapplications, latency resulting from the underly-ing network such as a WBAN should be mini-mized. While power saving is definitelybeneficial, replacement of batteries in WBANnodes is much easier than in WSNs, in whichnodes may be physically unreachable afterdeployment. Therefore, it may be necessary tomaximize battery life in a WSN at the expense ofhigher latency.

Mobility: Wearers of WBANs may movearound. WBAN nodes affiliated with the samewearer move together and in the same direction.In contrast, WSN nodes are usually consideredto be stationary, and any node mobility does notoccur in groups.

ABSTRACT

A wireless body area network is a radio-fre-quency-based wireless networking technologythat interconnects tiny nodes with sensor oractuator capabilities in, on, or around a humanbody. In a civilian networking environment,WBANs provide ubiquitous networking func-tionalities for applications varying from health-care to safeguarding of uniformed personnel.This article surveys pioneer WBAN researchprojects and enabling technologies. It exploresapplication scenarios, sensor/actuator devices,radio systems, and interconnection of WBANs toprovide perspective on the trade-offs betweendata rate, power consumption, and network cov-erage. Finally, a number of open research issuesare discussed.

CONSUMER COMMUNICATIONS AND NETWORKING

Huasong Cao and Victor Leung, University of British Columbia

Cupid Chow and Henry Chan, The Hong Kong Polytechnic University

Enabling Technologies for Wireless BodyArea Networks: A Survey and Outlook

CHAN LAYOUT 11/18/09 2:02 PM Page 84

IEEE Communications Magazine • December 2009 85

APPLICATIONS OF WBANSWBAN applications can be categorized based onthe type of sensors/actuators, radio systems, net-work topologies, and use cases. We enumeratehere several pioneer healthcare WBAN researchprojects, as well as platforms for HCI applications.

WBANS FOR HEALTHCAREWBANs extend conventional bedside monitoringto ambulatory monitoring, providing a point ofcare to patients, the elderly, and infants in bothhospital-based and home-based scenarios. Moni-toring, autonomous diagnostic, alarm, and emer-gency services, as well as management ofelectronic patient record databases can all beintegrated into one system to better serve people.

The CodeBlue project at Harvard University[1] considers a hospital environment where mul-tiple router nodes can be deployed on the wall.All nodes use the same ZigBee radio.Patients/caregivers can publish/subscribe to themesh network by multicasting; there is no cen-tralized or distributed server or database forcontrol and storage. Localization functionality isprovided by MoteTrack with an accuracy of 1 m,based on the same radio. As a result of mobilityand multihop transmissions, the system experi-ences considerable packet loss and is limited to40 kb/s aggregate bandwidth per receiver.

Based on the CodeBlue architecture, theAdvanced Health and Disaster Aid Network(AID-N) is being developed at Johns HopkinsUniversity [2] for mass casualty incidents whereelectronic triage tags can be deployed on victims.Additional wireless capabilities (e.g., Wi-Fi andcellular networks) are introduced to facilitate

communications between personal servers andthe central server where data are stored. Fur-thermore, a web portal is provided to multipletypes of users, including emergency departmentpersonnel, incident commanders, and medicalspecialists. A Global Positioning System (GPS)module is employed for outdoor localization,while a MoteTrack system is designed for track-ing indoors. However, patients have mobilityconstraints due to the lack of routers in the net-work, and a very limited number of sensor nodescan be put on each patient because of the limit-ed bandwidth.

The Wearable Health Monitoring Systems(WHMS) is being developed at the University ofAlabama [3] and targets a larger-scaletelemedicine system for ambulatory health statusmonitoring. Unlike CodeBlue and AID-N,WHMS has a star-topology network for eachpatient, which is connected via Wi-Fi or a cellu-lar network to a healthcare provider. The per-sonal server, implemented on a personal digitalassistant (PDA), cell phone, or personal comput-er (PC), coordinates the data collection fromsensor nodes using a time-division multipleaccess (TDMA) mechanism, provides an inter-face to users, and transfers data to a remote cen-tral server. Physicians can access data via theInternet, and alerts can be created by an agentrunning on the server. However, the power con-sumption and cost associated with long-termdata uploading can hamper system realization.

WBANS FOR HCITraditional computer interfaces, like keyboards,mice, joysticks, and touch screens, are allreplaceable by potential WBAN devices capable

Figure 1. Interconnection of WBAN, WPAN, (W)LAN, and wide area networks.

WLAN: Wireless Local Area NetworkWMAN: Wireless Metropolitan Area Network

WBAN WPAN (W)LAN

To Internet

To WMAN

Bloodpressure

ECG

Watch

EEG

EMG

Motion

Glucose


IEEE Communications Magazine • December 200986

of automatically recognizing human motions,gestures, and activities. Disabled people canbenefit from novel WBAN platforms based on aseries of miniature sensors. The intra-body com-munications (IBC) applications proposed in [4]can be used to assist handicapped people. Forexample, an IBC enabled sensor embeddedinside the shoes of a blind person can be used tosend voice information such as the current loca-tion to him/her by an IBC enabled facility, suchas a doorway or crosswalk. IBC enabled eye-glasses that can display texts, working with IBCenabled speakers, can help deaf people compre-hend audio broadcast announcements.

Early research efforts at MIT Media Labhave produced MITHril [5], a wearable comput-ing platform that includes electrocardiography(ECG), skin temperature, and galvanic skinresponse sensors for wearable sensing and con-text-aware interaction. MITHril is not a realWBAN in that multiple sensors are wired to asingle processor. A later version of this platform,MITHril 2003, extends MITHril to a multi-userwireless distributed wearable computing plat-form by utilizing Wi-Fi function available onPDAs (i.e., a PDA acts as a personal server andrelays data of each person to a central station).

The Microsystems Platform for Mobile Ser-vices and Applications (MIMOSA) [6] is aresearch project involving 15 partners from eightdifferent European countries to create ambientintelligence. MIMOSA’s approach is similar toWHMS while it exclusively employs a mobilephone as the user-carried interface device.Wibree, later renamed Bluetooth Low Energytechnology, and radio frequency identification(RFID) tags are used for connecting local sensornodes. NanoIP and Simple Sensor Interface(SSI) protocols are integrated into MIMOSA toprovide an application programming interface(API) for local connectivity and facilitate sensorreadings.

The Wireless Sensor Node for a Motion Cap-ture System with Accelerometers (WiMoCA) [7]project at several Italian universities is con-cerned with the design and implementation of adistributed gesture recognition system. The sys-tem has a star topology with all sensing nodessending data to a non-sensing coordinator nodeusing a TDMA-like approach, and the coordina-tor in turn relays the data to an external process-ing unit using Bluetooth. The sensing modules,each made up of a tri-axial accelerometer, canbe put on multiple parts of the body for motiondetection. The radio modules of all nodes workin the 868 MHz European license-exempt band,with up to 100 kb/s data rate. A Java-basedgraphical user interface (GUI) at the processingunit side interprets the data stream for posturerecognition.

SENSOR DEVICESSensors are the key components of a WBAN, asthey bridge the physical world and electronic sys-tems. Generally, they can be classified intochemical, thermal, mechanical, and acoustic sen-sors. Previous studies have shown that the fre-quency and amplitude range of humanphysiological signals are comparatively low; thus,

a low sampling frequency and low data transmis-sion rate would be sufficient. However, whatkind of and how many sensors a WBAN systememploys depend largely on the application sce-nario and the system infrastructure. To bettermonitor a human’s vital signals, behavior, andsurrounding environment, a wide range of com-mercially available sensors can be deployed, suchas accelerometer and gyroscope, ECG, elec-tromyography (EMG), and electroencephalogra-phy (EEG) electrodes, pulse oximetry,respiration, carbon dioxide (CO2), blood pres-sure, blood sugar, humidity, and temperaturesensors (Table 1). Commonly used sensordevices for WBANs are surveyed below.

Accelerometers are widely used for motioncapture. They measure the acceleration relativeto freefall in three axes. With an accelerometermounted on a certain part of a human body, thesystem can effectively register the subject’smovement. As an addition or alternative to anaccelerometer, a gyroscope can be used forapplications such as ambulatory gait monitoringand analysis. For distinguishing motions, anaccelerometer/gyroscope array with tens of sen-sors can be deployed. This raises questions forpositioning and noise reduction techniques.Proper positioning reduces the number ofaccelerometers/gyroscopes needed and theresulting data rate. Decreasing the number ofsensors reduces the motion detection signal-to-noise ratio due to lower redundancy, thus requir-ing the sensors to be deployed at the plannedlocations with higher accuracy.

ECG/EMG/EEG sensors measure potentialdifferences across electrodes attached to corre-sponding parts of the body. The electric currentappearing on the skin is a result of heart, mus-cle, or brain activities and conductivity of thehuman body. Therefore, temporal graphsobtained from these electrodes provide indirectways of analyzing and diagnosing certain humanphysiological conditions. For bedside monitor-ing, disposable electrodes are usually used. How-ever, long-term usage of these types of sensorsmay cause artifacts as well as skin problems. Analternate solution is textile-structured electrodes,which are ECG sensors embedded inside gar-ments, such as fiber, yarn, and fabric structures[8]. These textile-structure electrodes, possiblywoven into clothes, are more comfortable andsuitable for long-term monitoring. They aremuch more flexible than the disposable elec-trodes since the shape can change with humanmovement, and they are also free of skin prob-lems. Again, noise reduction is the crucial prob-lem for these monitoring devices. Ideally, thesecircuits are implanted to have direct contact withwhatever they are monitoring, but that could betoo obtrusive. While a particular type of sensoris monitoring a specific physiological signal (e.g.,ECG), other physiological signals such as EMGare contributing to the overall noise and need tobe suppressed.

With advances in micro-electromechanicalsystems (MEMS), sensor devices are gettingeven tinier in size and changing the traditionalway of measuring human physiological parame-ters. Accelerometers and gyroscopes are goodexamples. It has been reported that MEMS elec-

Traditional computer

interfaces, like

keyboards, mice,

joysticks, and touch

screens are all

replaceable by

potential WBAN

devices capable of

automatically

recognizing human

motions, gestures,

and activities. Dis-

abled people can

benefit from novel

WBAN platforms

based on a series of

miniature sensors.



trodes for ECG/EEG acquisition have beendeveloped by fabricating little spikes on siliconor polymers.

Figure 2 shows a typical sensor node withsensor, radio, and memory modules. The sensormodule consists of a sensor, a filter, and an ana-log-to-digital converter (ADC). The sensor con-verts some form of energy to analog electricsignals, which are bandpass filtered and thendigitalized by the ADC for transmissions or fur-ther processing. We discuss radio systems forWBANs and WPANs used for transmissions ofsensed data in the next section.

RADIO TECHNOLOGIES

RADIO PROPAGATIONIn the past few years, researchers have madeconsiderable progress in characterizing the bodyarea propagation environment through bothmeasurement-based and simulation-based stud-ies in order to support prediction of link levelperformance in alternative sensor deploymentconfigurations, and development of more effec-tive antennas with, say, lower specific absorptionand better coupling to the dominant propagationmodes. These works have been conducted inboth the industrial, scientific, and medical (ISM)bands between 400 MHz and 2.45 GHz, and theultra-wideband (UWB) frequency allocationbetween 3.1 and 10.6 GHz [9]. In each of thefrequency bands, intra-body, on-body, and off-body channels have been studied. Figure 3 showsan example of the path loss measurements forseveral body locations and frequency bandsbased on [10]. The intra-body propagation chan-nel can be described using an appropriate model.For example, the Ricean distribution can beused for modeling the intra-body propagationchannel based on the K-factor, which is the ratiobetween the average powers of the direct and

reflected paths [4], and indicates the channelquality. Significant progress has also been madetoward:• Identification of the propagation mecha-

nisms that affect signal transmissionsbetween nodes

• Assessment of the effects of multipathreflections from the external environmentto signal transmissions between nodes

• Characterization of the fading statistics onbody links that occur with body motion andchange of body position in both sparse andrich scattering environments

• Development of standard UWB channelimpulse response models and evaluation oftypical modulation schemes utilizing them[11]Following is a comparative study of emerging

and existing standards for WBANs and WPANs,including Bluetooth Low Energy, UWB, Blue-tooth 3.0, and ZigBee. They are listed in Table 2for comparison. Also summarized in the tableare proprietary and open technologies likeInsteon, Z-Wave, ANT, RuBee, and RFID.Insteon and Z-Wave are both proprietary meshnetworking technologies for home automation.Z-Wave works in the 2.4 GHz ISM band, whileInsteon makes use of both power lines and the900 MHz ISM band. ANT is another proprietarysensor networking technology, featuring a sim-pler protocol stack and lower power consump-tion. ANT has been embedded in some Nikeshoes to collect workout data and is able to talkto iPod products. RuBee and RFID are bothused for asset management and tracking. Theyare complimentary to each other in terms of fre-quency bands, battery life, and application sce-narios. These technologies have all beenimplemented on silicon chips and are being soldin comparable volumes each year. With advancesin very large-scale integration (VLSI), dual and

Table 1. Sensors commonly employed in WBAN systems and their typical data rates.

Sensor How it works Data rate

Accelerometer Measures the acceleration relative to freefall in three axes High

Gyroscope Measures the orientation, based on the principles of angular momentum High

ECG/EEG/EMG Measures potential difference across electrodes put on corresponding parts of the body High

Pulse oximetry Measures ratio of changing absorbance of the red and infrared light passing from one side to theother of a thin part of the body's anatomy Low

Respiration Uses two electrodes, cathode and anode covered by a thin membrane to measure the oxygen dis-solved in a liquid Low

Carbon dioxide Uses the infrared light and measures the absorption of the gas presented Low

Blood pressure Measures the systolic pressure (peak pressure) and diastolic pressure (minimum pressure) Low

Blood sugar Traditionally analyzes drops of blood from a finger tip, recently, uses non-invasive method including anear infrared spectroscopy, ultrasound, optical measurement at the eye, and the use of breath analysis Low

Humidity Measures the conductivity changes of the level of humidity Very low

Temperature Uses a silicon integrated circuit to detect the temperature changes by measuring the resistance Very low



multiple-standard radios can be integrated into asingle chip, greatly reducing the cost and powerconsumption, while fostering combining as wellas merging of technologies.

BLUETOOTH LOW ENERGY TECHNOLOGYBluetooth Low Energy technology, formerlyknown as Bluetooth Low End Extension (LEE)and later Wibree, provides ultra-low power con-sumption and cost while minimizing the differ-ence between Bluetooth and itself. Introduced in2004 by Nokia, Bluetooth LEE was designed towirelessly connect small devices to mobile termi-nals. Those devices are often too tiny to bear thepower consumption as well as cost associatedwith a standard Bluetooth radio, but are idealchoices for the health monitoring applicationspreviously discussed. Bluetooth LEE was said tobe a hardware-optimized radio, which means itsmajor difference from Bluetooth resides in theradio transceiver, baseband digital signal pro-cessing, and data packet format. After furtherdevelopment under the project MIMOSA, whichtargets use cases including both WBANs andWPANs, LEE was released to the public withthe name Wibree in 2006. One year later, anagreement was reached to include it in futureBluetooth specifications as Bluetooth Low Ener-gy technology.

Bluetooth Low Energy technology is expectedto provide a data rate of up to 1 Mb/s. Usingfewer channels for pairing devices, synchroniza-tion can be done in a few milliseconds comparedto Bluetooth’s seconds. This benefits latency-critical WBAN applications (e.g., alarm genera-

tion and emergency response) and enhancespower saving. Bluetooth Low Energy productscan be categorized into two groups: dual-modechips and standalone chips. As the names indi-cate, standalone chips are intended to beequipped with sensors/actuators and talk to eachother only, while dual-mode chips are to beequipped with a personal server (e.g., smartphone) and also be able to connect to traditionalBluetooth devices.

Similar to Bluetooth, Bluetooth Low Energytechnology will likely operate using a simplerprotocol stack and focus on short-range star-con-figured networks without complicated routingalgorithms. This suits WBANs configured in startopology such as WHMS, and provides bettermobility support for them. Inter-WBAN commu-nications can be realized through a second radioor using a dual-mode chip; however, the trade-off is higher power consumption.

UWBAccording to the Federal Communications Com-mission (FCC), UWB refers to any radio tech-nology having a transmission bandwidthexceeding the lesser of 500 MHz or 20 percentof the arithmetic center frequency. FCC alsoregulates license-free use of UWB in the3.1–10.6 GHz band to have a relatively lowpower spectral density emission. This leads tothe suitability of UWB applications in short-range and indoor environments, and environ-ments sensitive to RF emissions (e.g., in ahospital). Commercial products based on UWBprovide extremely high data rates; for example,

Figure 2. Typical modules on a sensor node.

Sensor node

Sensor node

Sensor node

Bus

Data flow Sensor node

WBAN

RAM: Random access memoryROM: Read-only memory

Radio module

WPAN Senso ADC

Sensor module

Filter RAM Flash

Memory module

ROM

Microprocessor module

Bluetooth LEE was

designed to

wirelessly connect

small devices to

mobile terminals.

Those devices are

often too tiny to

bear the power

consumption as well

as cost associated

with a standard

Bluetooth radio, but

are ideal choices for

health-monitoring

applications.



certified wireless USB devices work at up to 480Mb/s, enabling short-range wireless multimediaapplications, such as wireless monitors, wirelessdigital audio and video players, and other HCIuse cases. These multimedia devices can beeither wirelessly connected with WBANs orthemselves portable as part of a WBAN. UWBis also an ideal technology for precise localiza-tion, which complements GPS in the indoorenvironment for WBAN tracking.

An emerging WBAN standard, IEEE 802.15.6— Body Area Networks (BANs), will likelyemploy UWB, according to recent proposals andmeeting minutes. The standard intends to endowfuture generation electronics in close proximityto or inside the human body. However, whenthis standard and any electronics that utilize itwill become available remains unknown.

BLUETOOTH 3.0 + HIGH SPEEDBluetooth technology was designed as a replace-ment of RS232 cables, and later evolved tobecome a widely accepted wireless alternativefor connecting a variety of personal devices. Itdiffers from others in separately supportingaudio and data traffic streams. This is probablywhy Bluetooth headsets are seen everywhere.The newly adopted standard, Bluetooth 3.0 +HS, introduces the 802.11 protocol adaptationlayer (PAL) into the protocol stack, and increas-es data rate support from 3 Mb/s to 24 Mb/s,supporting applications like transferring bulkdata files. Together with its Low Energy exten-sion, Bluetooth accommodates applications withdifferent data rate, power consumption, and net-work coverage requirements.

Limitations of Bluetooth include the smallnumber of active slaves (seven) that each piconetsupports and indirect communications betweenslaves. Although one slave can participate inmore than one piconet, it is not an efficient wayof connecting nodes. A WBAN project, Mobi-Health, developed in the early 2000s employedBluetooth for transmitting sensor data from afront-end device to a mobile phone or PDA [12].However, the emergence of Bluetooth LowEnergy presents a more suitable alternative.Bluetooth is most suitable for short-term high-data-rate communications, connecting two peerdevices in an ad hoc way, such as exchangingdata between two personal servers in twoWBANs, or between a WBAN and a PC. Blue-tooth has already been widely adopted in themobile phone industry to connect headsets, PCs,and other mobile computing platforms together.Commercial WPAN products employing Blue-tooth include carkits, printers, digital cameras,Nintendo’s Wii, and Sony’s PlayStation 3.

ZIGBEEZigBee/IEEE 802.15.4 targets low-data-rate andlow-power-consumption applications. Specifical-ly, the ZigBee Alliance has been working onsolutions for smart energy, and home, building,and industrial automation. The recently complet-ed ZigBee Health Care public application pro-file provides a flexible framework to meetContinua Health Alliance requirements forremote health and fitness monitoring. Thesesolutions better suit WBAN deployment scenar-ios in a limited area (e.g., a hospital or a house),as in the cases of CodeBlue [1] and AID-N [2].

Table 2. A comparison of WBAN and WPAN technologies. (Only most commonly acknowledged and/or applied parameters are listedhere due to space limitation.)

Technology Frequency band Data rate (b/s) Multiple access method Coverage area(meter)

Networktopology

Bluetooth Low Energy 2.4 GHz ISM 1 M FH + TDMA 10 Star

UWB (ECMA-368) 3.1~10.6 GHz 480 M CSMA/TDMA <10 Star

Bluetooth 3.0 + HighSpeed 2.4 GHz ISM 3~24 M FH + TDMA/CSMA (Wi-Fi) 10 star

ZigBee (IEEE 802.15.4) ISM 250 k CSMA 30~100 Star/mesh

Insteon131.65 KHz(powerline)902~924 MHz

13 k Unknown Home area Mesh

Z-Wave 900 MHz ISM 9.6 k Unknown 30 Mesh

ANT 2.4 GHz ISM 1 M TDMA Local area Star/mesh

RuBee (IEEE 1902.1) 131 KHz 9.6 k unknown 30 Peer-to-peer

RFID (ISO/IEC 18000-6) 860~960 MHz 10~100 k Slotted-Aloha/binary tree 1~100 Peer-to-peer

FH: Frequency hoppingTDMA: Time-division multiple accessCSMA: Carrier sense multiple access


Compared to Bluetooth and UWB, ZigBee/IEEE 802.15.4 devices can operate in three ISMbands, with data rates from 20 kb/s to 250 kb/s.ZigBee supports three types of topologies: star,cluster tree, and mesh. In the star topology, acoordinator initiates and controls the network(i.e., similar to a piconet in Bluetooth), but thereis no need for synchronization. The major advan-tage of ZigBee is its capability of providing mul-tihop routing in a cluster tree or mesh topology.As a result, WBAN network coverage can beexpanded to a WPAN using the same radio. AZigBee mesh network may include both full-function devices (FFDs) and reduced-functiondevices (RFDs), where an RFD is equivalent toa standalone chip in Bluetooth Low Energy andcan only act as an end device, while an FFD isequivalent to a dual-mode chip and can also actas a coordinator or router.

There have been many academic researchprojects uti l izing ZigBee for transportinghealth-related data. For example, a wirelessECG Plaster for WBAN developed at theNational University of Singapore uses an ECGfront-end developed in-house and a TI CC2430chip to collect ECG streams [13]. Most proto-types mentioned earlier, however, are based onIEEE 802.15.4 chips that do not employ thehigher-layer ZigBee protocol stack, becauseeither networking capability is not a must, or

researchers are interested in devising moreappropriate protocols. In our view, ZigBee mayhave a better chance to be adopted in the areaof home automation and industrial automationand control, while in the area of connectinglow-power peripheral devices around thehuman body (e.g., watches, health-related mon-itors, and sports sensors), Bluetooth Low Ener-gy technology has greater potential to be widelyemployed, due to its association with Bluetoothas well as lower cost and lower power con-sumption.

CONNECTING WBANSAND THE WORLD

While the coverage of a WBAN is limited toabout 2 m, it may interwork with other wirelessnetworks to largely extend its coverage area,facilitating connectivity between those sensorydevices and the outside world (Fig. 4). Thisenables emergency alarms to be generated bothlocally and remotely, and monitoring, data stor-age, and management capabilities to be support-ed in a more capable computing platform at adistant location. These services can be providedbased on push or pull strategies.

A global trend for interconnection of datanetworks is to use IP. WBAN packets can be


Figure 3. Path loss values for different body locations and frequency bands (based on [10]).

Measurements for 820 MHz

Chest

10

Path

loss

in d

B

0Rightwrist

Leftwrist

Rightankle

Leftankle

Back

(Receiver at right hip)

20

30

40

50

60

40

80Measurements for 3.1–3.5 GHz

Leftear

10

Path

loss

in d

B

0

(Transmitter at left waist)

Rightear

Leftwrist

Rightwrist

Rightwaist

Leftankle

Rightankle

30

20

40

50

60

70

80

90

Measurements for 2.36 GHz

Chest

10

Path

loss

in d

B

0Rightwrist

Leftwrist

Rightankle

Leftankle

Back

(Receiver at right hip)

20

30

40

50

60

40

80Measurements for 7.25–8.5 GHz

Leftear

10

Path

loss

in d

B

0

(Transmitter at left waist)

Rightear

Leftwrist

Rightwrist

Rightwaist

Leftankle

Rightankle

30

20

40

50

60

70

80

90StandingWalkingRunning

StandingSitting

StandingWalkingRunning

StandingSitting



translated into IP datagrams by a gateway atthe edge of a WBAN, as in the AID-N,WHMS, and MIMOSA platforms. In particu-lar, such a gateway can be a smart phoneequipped with multiple network interfaces,which enables the owner to interact withhis/her WBAN and forward data anywhere inthe world. Existing communications technolo-gies such as short message services (SMS), gen-eral packet radio services (GPRS), and emailservices can also be used to speed up or assistthe data transfer. Another approach is tonatively integrate IP into WBAN packets; as aresult, the underlying network infrastructurewill be transparent to applications. There areongoing projects aimed at this goal, such asBluetooth Personal Area Networking Profile,IPv6 over Low Power Wireless Personal AreaNetworks (6LoWPAN), and IP for SmartObjects (IPSO). The ZigBee Alliance recentlyalso announced its decision to incorporatestandards from the Internet Engineering TaskForce (IETF) into its specifications.

The ubiquitous access and connectivity ofWBANs into the global network requires notonly network infrastructure support, but alsolow-power and low-footprint software implemen-tations for routing, flow/error control, remoteprocedure calls, database management, and userinterface. Recently, user interfaces are increas-ingly provided through Web 2.0 portals, consid-ering the easy access to its services and stronginteractive characteristics.

OPEN RESEARCH ISSUES

While WBANs will undoubtedly play an impor-tant role in enabling ubiquitous communications,many issues remain to be addressed beforeWBAN technologies are widely applied, as sum-marized below.

PHYSICAL CHARACTERISTICS OFSENSOR/ACTUATOR MATERIALS AND

ELECTRONIC CIRCUITS

As sensors/actuators are going to be put onhuman bodies or even implanted, their size,form factor, and physical compatibility to humantissues are crucial. This motivates the search forand synthesis of novel materials. At the sametime, concerns regarding electronic and magnet-ic energy absorbed by human tissues from RFcircuits placed in close proximity to humansmean that WBAN devices need to employ lowtransmission power and low transmission dutycycles. In this regard UWB outperforms conven-tional transmission methods and thus attractsmuch attention.

DEVELOPMENT AND EVALUATION OF IMPROVEDPROPAGATION AND CHANNEL MODELS

As discussed earlier, body area propagationenvironment has been characterized extensivelyat link level. There is still a need for accuratemodels that help researchers predict the impact

Figure 4. Connecting WBANs to the world and global data storage, management and sharing.

WBANs

WBANs

WBANs

WBANs

Link to other networks

Metropolitan area

WBANs



of realistic channels on network level perfor-mance. Taking into account factors such as reli-ability, latency, mutual interference, energyconsumption, and mobility effects in such amodel, more effective network architecture androuting algorithms for WBANs can be devised.Recent years have seen growing interest inusing UWB channel models for WBANs. Forexample, in [14] an experiment was done on ahuman body over 3.1–10.6 GHz in the indoorand anechoic chamber to study the path lossexponent under various conditions. Anotherrelated issue is performance evaluation. Forexample, when the WBAN signal is transmittedbetween two sensors, the signal propagationthrough the body is affected by the diffractionaround the body and the reflections from thebody or other objects. Path loss and delayspread will affect the performance of the sys-tem, especially when the sensors are placed ondifferent sides of a body. According to onestudy, the packet error rate should generally bekept less than 1 percent [15].

NETWORKING ANDRESOURCE MANAGEMENT SCHEMES

As the application scenarios of WBANs aredifferent from traditional sensor networks,problems like power management, sensor cali-bration, and context-aware network configura-tion need to be revisited as well. Sensor nodescan join/leave the network at any time, andthus impose the requirements of configuringthe devices on the fly. Dynamic managementof resources, including both sensor functionali-ties and communication bandwidth, is alsonecessary.

SECURITY, AUTHENTICATION, ANDPRIVACY ISSUES

Privacy requires effective and efficient authenti-cation techniques in WBANs. Multimodalauthentication schemes based on such things ashuman faces, hand features, and EEG signals areactively being developed in both academia andindustry. Complex but distinguishable humanbody characteristics provide an ideal way ofauthenticating users, but they also create otherchallenges (e.g., protecting the privacy of users).Different levels of security should be identified,and appropriate mechanisms shall be developedto distinguish life-threatening requests fromother applications with various security prioritiesand appropriate privacy protection measures.

POWER SUPPLY ISSUESAs all WBAN devices require an energy sourcefor data collection, processing, and transmission,development of suitable power supplies becomesparamount. Most WBAN devices are poweredby batteries, which may not even be replaceablein cases where the devices are implanted in thehuman body; thus, techniques like remote bat-tery recharging are important. In addition toenergy harvesting methods (e.g., based on bodymovements) many researchers are studying,recently researchers at MIT have reported wire-less energy transmission to power electronic

devices over a short range (i.e., several meters)using evanescent waves [16].

RULES OF ENGAGEMENTEfforts have been put into the interoperabilityof desktop telemedicine systems and bedsidedevices (e.g., the development of Health Level7 and ISO/IEEE 11073 [17]). However, intelli-gent monitoring and treatment systems employ-ing WBANs require standardized rules ofengagement in ambulatory environments, pro-viding point of care without limitation of thewearer’s location/mobility while protecting thepatient’s privacy. Interoperability protocols atthe application or domain level (e.g., samplerate, data precision, association/disassociation,device descriptions, and nomenclature) shouldall be addressed, and vendor-independentattributes and user interfaces shall be madeavailable.

CONCLUSIONAs a complement to existing wireless technolo-gies, the WBAN plays a very important role inubiquitous healthcare applications and enjoys ahuge potential market in the area of consumerelectronics. Its advancements have been theresult of interdisciplinary research and develop-ment. In this article we have provided a compre-hensive review and outlook of this promisingfield through a survey of pioneer WBANresearch projects and enabling technologies,including application scenarios, sensor/actuatordevices, radio systems, and interconnection ofWBANs. While WBAN technologies provide apromising platform to enable ubiquitous com-munications, several open issues still need to beaddressed. In particular, for life-saving applica-tions, thorough studies and tests should be con-ducted before WBANs can be widely applied tohumans.

ACKNOWLEDGMENTSThis work is supported in part by the CanadianNatural Sciences and Engineering ResearchCouncil under grant STPGP 365208-08 and bythe Department of Computing, The Hong KongPolytechnic University.

REFERENCES[1] V. Shnayder et al., “Sensor Networks for Medical Care,”

Harvard Univ. tech. rep. TR-08-05, Apr. 2005.[2] T. Gao et al., “The Advanced Health and Disaster Aid

Network: A Lightweight Wireless Medical System forTriage,” IEEE Trans. Biomedical Circuits and Sys., vol. 1,no. 3, Sept. 2007, pp. 203–16.

[3] A. Milenkovic, C. Otto, and E. Jovanov, “Wireless SensorNetworks for Personal Health Monitoring: Issues and anImplementation,” Comp. Commun., vol. 29, no. 13–14,Aug. 2006, pp. 2521–33.

[4] J. A. Ruiz and S. Shimamoto, “Novel CommunicationServices Based on Human Body and Environment Inter-action: Applications inside Trains and Applications forHandicapped People,” Proc. IEEE WCNC 2006, LasVegas, NV, 2006.

[5] S. Pentland, “Healthwear: Medical Technology BecomesWearable,” Computer, vol. 37, no. 5, May 2004, pp.42–49.

[6] I. Jantunen et al., “Smart Sensor Architecture forMobile-Terminal-Centric Ambient Intelligence,” Sensorsand Actuators A: Physical, vol. 142, no.1, Mar. 2004,pp. 352–60.

While WBAN

technologies provide

a promising platform

to enable ubiquitous

communications,

several open issues

still need to be

addressed.

In particular, for

those life-saving

applications,

thorough studies and

tests should be

conducted before

WBANs can be

widely applied to

humans.

CHAN LAYOUT 11/18/09 2:02 PM 2


[7] E. Farella et al., “Interfacing Human and Computer withWireless Body Area Sensor Networks: The WiMoCASolution,” Multimedia Tools App., vol. 38, no. 3, July2008, pp. 337–63.

[8] P. J. Xu, H. Zhang, and X. M. Tao, “Textile-StructuredElectrodes for Electrocardiogram,” Textile Progress, vol.40, no. 4, Dec. 2008, pp. 183–213.

[9] P. S. Hall and Y. Hao, Antennas and Propagation forBody-Centric Wireless Communications, Artech House,2006.

[10] K. Y. Yazdandoost et al., “Channel Model for BodyArea Network (BAN),” IEEE P802.15-08-0780-08-0006,Apr. 2009.

[11] K. Takizawa, T. Aoyagi, and R. Kohno, “Channel Mod-eling and Performance Evaluation of UWB-based Wire-less Body Area Networks,” Proc. IEEE ICC 2009 ,Dresden, Germany, 2009.

[12] R. Istepanian, S. Laxminarayan, and C. S. Pattichis, M-Health: Emerging Mobile Health Systems, Springer,2005.

[13] M. C. Munshi et al., “Wireless ECG Plaster for BodySensor Network,” Proc. 5th Int’l. Wksp. Wearable andImplantable Body Sensor Net., Hong Kong, China,2008.

[14] Y. P. Zhang, L. Bin, and C. Qi, “Characterization of On-Human-Body UWB Radio Propagation Channel,”Microwave Optical Tech. Lett., vol. 49, no. 6, pp.1365–71.

[15] J.-Y. Yu, W.-C. Liao, and C.-Y. Lee, “A MT-CDMA basedWireless Body Area Network for Ubiquitous HealthcareMonitoring,” Proc. BioCAS 2006, Nov. 2006, pp.98–101.

[16] A. Kurs et al., “Wireless Power Transfer via StronglyCoupled Magnetic Resonances,” Science, vol. 317, no.5834, July 2007, pp. 83–86.

[17] S. Warren and E. Jovanov, “The Need for Rules ofEngagement Applied to Wireless Body Area Networks,”Proc. IEEE CCNC 2006, Las Vegas, NV, 2006.

BIOGRAPHIESHUASONG CAO [S] ([email protected]) received hisB.Eng. degree in electrical engineering from Wuhan Univer-sity, Wuhan, P.R. China, in 2007. He is currently an M.A.Sc.student in electrical and computer engineering at the Uni-versity of British Columbia. His research interests includewireless networks in general and specifically wireless bodyarea networks.

CUPID CHOW ([email protected]) is a laboratoryofficer at the Department of Computing, The Hong KongPolytechnic University. She received her M.A.Sc. degree inelectrical engineering from the University of BritishColumbia, Canada. She worked as a research engineer inthe Nokia Research Center, Finland, to develop the adap-tive modulation and coding (AMC) scheme for high-speeddownlink Packet Access (HSDPA). She has been working onvarious wireless technologies, such as RFID, Bluetooth,WCDMA, and Wi-Fi.

HENRY C. B. CHAN [M] ([email protected])received his B.A. and M.A. degrees from the University ofCambridge, England, and his Ph.D. degree from the Univer-sity of British Columbia, Canada. He is an associate profes-sor in the Department of Computing, The Hong KongPolytechnic University. His research interests include net-working/communications, electronic commerce, and Inter-net technologies.

VICTOR C. M. LEUNG [F] ([email protected]) received hisB.A.Sc. and Ph.D. degrees, both in electrical engineering,from the University of British Columbia in 1977 and 1981,respectively. He is a professor and holder of the TELUSMobility Research Chair in the Department of Electrical andComputer Engineering of the same university. His researchinterests are in wireless networks and mobile systems. Heis an editor of IEEE Transactions on Computers.

Latest Resources for Wireless Engineering Professionals

A Guide to the Wireless Engineering Body of Knowledge

(WEBOK)

www.wiley.comUS $69.95

IEEE Communications Society member order with

15% discount with Promo Code: 18493

+1 877 762 2974 (US) +1 800 567 4797 (Canada) +1 44 1243 843294 (world)

ISBN: 978-0-470-43366-9Paper, 272pp, 2009

WCET Area 2:Wireless Access Technologies by Javan Erfanian

WCET Area 3: Network and Service Architecturesby Daniel Wong

More online tutorials offered at www.comsoc.org/tutorialsnow

Recommended Resources for IEEE WCET Certification Program

www.ieee-wcet.org


Enabling Technologies for Wireless Body Area … · A wireless body area network (WBAN) is a radio...

Documents

Transcript of Enabling Technologies for Wireless Body Area … · A wireless body area network (WBAN) is a radio...