A Comprehensive Survey of Wireless Body Area Netwo
A Comprehensive Survey of Wireless Body Area Netwo
A Comprehensive Survey of Wireless Body Area Netwo
net/publication/45709672
CITATIONS READS
788 5,759
9 authors, including:
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Ingrid Moerman on 11 August 2014.
recorded over a longer period of time, improving the – And finally the devices are often very heterogeneous,
quality of the measured data [9]. may have very different demands or may require
Generally speaking, two types of devices can be dis- different resources of the network in terms of data
tinguished: sensors and actuators. The sensors are used rates, power consumption and reliability.
to measure certain parameters of the human body, ei- When referring to a WBAN where each node com-
ther externally or internally. Examples include mea- prises a biosensor or a medical device with sensing unit,
suring the heartbeat, body temperature or recording some researchers use the name Body Area Sensor Net-
a prolonged electrocardiogram (ECG). The actuators work (BASN) or in short Body Sensor Network (BSN)
(or actors) on the other hand take some specific ac- instead of WBAN [12]. These networks are very similar
tions according to the data they receive from the sensors to each other and share the same challenges and prop-
or through interaction with the user. E.g., an actuator erties. In the following, we will use the term WBAN
equipped with a built-in reservoir and pump adminis- which is also the one used by the IEEE [13].
ters the correct dose of insulin to give to diabetics based In this article we present a survey of the state of
on the glucose level measurements. Interaction with the the art in Wireless Body Area Networks. Our aim is to
user or other persons is usually handled by a personal provide a better understanding of the current research
device, e.g. a PDA or a smart phone which acts as a issues in this emerging field. The remainder of this pa-
sink for data of the wireless devices. per is organized as follows. First, the patient monitoring
In order to realize communication between these de- application is discussed in Section 2. Next, the char-
vices, techniques from Wireless Sensor Networks (WSNs) acteristics of the communication and the positioning
and ad hoc networks could be used. However, because of WBANs amongst other wireless technologies is dis-
of the typical properties of a WBAN, current proto- cussed in Section 4. Section 5 gives an overview of the
cols designed for these networks are not always well properties of the physical layer and the issues of com-
suited to support a WBAN. The following illustrates municating near or in the body. Existing protocols for
the differences between a Wireless Sensor Network and the MAC-layer and network layer are discussed in Sec-
a Wireless Body Area Network: tion 6 and Section 7 respectively. Section 8 deals with
cross-layer protocols available for WBANs. The Quality
– The devices used have limited energy resources avail- of Service and possible security mechanisms are treated
able as they have a very small form factor (often less in Section 9 and 10. An overview of existing projects
than 1 cm3 [10]). Furthermore, for most devices it is given in Section 11. Finally, the open research issues
is impossible to recharge or change the batteries al- are discussed in Section 12 and Section 13 concludes
though a long lifetime of the device is wanted (up the paper.
to several years or even decades for implanted devi-
ces). Hence, the energy resources and consequently
the computational power and available memory of 2 Patient Monitoring
such devices will be limited;
– All devices are equally important and devices are The main cause of death in the world is CardioVascular
only added when they are needed for an application Disease (CVD), representing 30% of all global deaths.
(i.e. no redundant devices are available); According to the World Health Organization, world-
– An extremely low transmit power per node is needed wide about 17.5 million people die of heart attacks or
to minimize interference and to cope with health strokes each year; in 2015, almost 20 million people will
concerns [11]; die from CVD. These deaths can often be prevented
– The propagation of the waves takes place in or on a with proper health care [14]. Worldwide, more than 246
(very) lossy medium, the human body. As a result, million people suffer from diabetes, a number that is
the waves are attenuated considerably before they expected to rise to 380 million by 2025 [15]. Frequent
reach the receiver; monitoring enables proper dosing and reduces the risk
– The devices are located on the human body that can of fainting and in later life blindness, loss of circulation
be in motion. WBANs should therefore be robust and other complications [15].
against frequent changes in the network topology; These two examples already illustrate the need for
– The data mostly consists of medical information. continuous monitoring and the usefulness of WBANs.
Hence, high reliability and low delay is required; Numerous other examples of diseases would benefit from
– Stringent security mechanisms are required in order continuous or prolonged monitoring, such as hyperten-
to ensure the strictly private and confidential char- sion, asthma, Alzheimer’s disease, Parkinson’s disease,
acter of the medical data; renal failure, post-operative monitoring, stress-monitoring,
3
3.1 Types of Devices Table 1 Examples of medical WBAN applications [21, 25–27]
sensor’s: actuator hardware (e.g. hardware for medi- Cochlear implant 100 kbps – –
cine administration, including a reservoir to hold the Artificial retina 50-700 kbps – –
medicine), a power unit, a processor, memory and Audio 1 Mbps – –
a receiver or transceiver. Voice 50-100 kbps – –
(Wireless) Personal Device (PD):
A device that gathers all the information acquired
by the sensors and actuators and informs the user is a higher than the raw bit rate of most existing low
(i.e. the patient, a nurse, a GP etc.) via an exter- power radios.
nal gateway, an actuator or a display/LEDS on the The reliability of the data transmission is provided
device. The components are a power unit, a (large) in terms of the necessary bit error rate (BER) which is
processor, memory and a transceiver. This device is used as a measure for the number of lost packets. For a
also called a Body Control Unit (BCU) [4], body- medical device, the reliability depends on the data rate.
gateway or a sink. In some implementations, a Per- Low data rate devices can cope with a high BER (e.g.
sonal Digital Assistant (PDA) or smart phone is 10−4 ), while devices with a higher data rate require
used. a lower BER (e.g. 10−10 ). The required BER is also
dependent on the criticalness of the data.
Many different types of sensors and actuators are
used in a WBAN. The main use of all these devices is
to be found in the area of health applications. In the
following, the term nodes refers to both the sensor as 3.3 Energy
actuator nodes.
Energy consumption can be divided into three domains:
The number of nodes in a WBAN is limited by na-
sensing, (wireless) communication and data process-
ture of the network. It is expected that the number of
ing [23]. The wireless communication is likely to be
nodes will be in the range of 20–50 [6, 24].
the most power consuming. The power available in the
nodes is often restricted. The size of the battery used
to store the needed energy is in most cases the largest
3.2 Data Rates contributor to the sensor device in terms of both di-
mensions and weight. Batteries are, as a consequence,
Due to the strong heterogeneity of the applications, kept small and energy consumption of the devices needs
data rates will vary strongly, ranging from simple data to be reduced. In some applications, a WBAN’s sensor-
at a few kbit/s to video streams of several Mbit/s. Data /actuator node should operate while supporting a bat-
can also be sent in bursts, which means that it is sent tery life time of months or even years without interven-
at higher rate during the bursts. tion. For example, a pacemaker or a glucose monitor
The data rates for the different applications are given would require a lifetime lasting more than 5 years. Es-
in in Table 1 and are calculated by means of the sam- pecially for implanted devices, the lifetime is crucial.
pling rate, the range and the desired accuracy of the The need for replacement or recharging induces a cost
measurements [25, 26]. Overall, it can be seen that the and convenience penalty which is undesirable not only
application data rates are not high. However, if one has for implanted devices, but also for larger ones.
a WBAN with several of these devices (i.e. a dozen mo- The lifetime of a node for a given battery capacity
tion sensors, ECG, EMG, glucose monitoring etc.) the can be enhanced by scavenging energy during the op-
aggregated data rate easily reaches a few Mbps, which eration of the system. If the scavenged energy is larger
5
than the average consumed energy, such systems could intervention. The self-organizing aspect also includes
run eternally. However, energy scavenging will only de- the problem of addressing the nodes. An address can
liver small amounts of energy [5, 28]. A combination be configured at manufacturing time (e.g. the MAC-
of lower energy consumption and energy scavenging is address) or at setup time by the network itself. Fur-
the optimal solution for achieving autonomous Wireless ther, the network should be quickly reconfigurable, for
Body Area Networks. For a WBAN, energy scavenging adding new services. When a route fails, a back up path
from on-body sources such as body heat and body vi- should be set up.
bration seems very well suited. In the former, a thermo- The devices may be scattered over and in the whole
electric generator (TEG) is used to transform the tem- body. The exact location of a device will depend on the
perature difference between the environment and the application, e.g. a heart sensor obviously must be placed
human body into electrical energy [27]. The latter uses in the neighborhood of the heart, a temperature sen-
for example the human gait as energy source [29]. sor can be placed almost anywhere. Researchers seem
During communication the devices produce heat which to disagree on the ideal body location for some sensor
is absorbed by the surrounding tissue and increases the nodes, i.e. motion sensors, as the interpretation of the
temperature of the body. In order to limit this temper- measured data is not always the same [32]. The net-
ature rise and in addition to save the battery resources, work should not be regarded as a static one. The body
the energy consumption should be restricted to a min- may be in motion (e.g. walking, running, twisting etc.)
imum. The amount of power absorbed by the tissue is which induces channel fading and shadowing effects.
expressed by the specific absorption rate (SAR). Since The nodes should have a small form factor consis-
the device may be in close proximity to, or inside, a tent with wearable and implanted applications. This
human body, the localized SAR could be quite large. will make WBANs invisible and unobtrusive.
The localized SAR into the body must be minimized
and needs to comply with international and local SAR
regulations. The regulation for transmitting near the 3.6 Security and Privacy
human body is similar to the one for mobile phones,
with strict transmit power requirements [11, 30] The communication of health related information be-
tween sensors in a WBAN and over the Internet to
servers is strictly private and confidential [33] and should
3.4 Quality of Service and Reliability
be encrypted to protect the patient’s privacy. The med-
ical staff collecting the data needs to be confident that
Proper quality of service (QoS) handling is an impor-
the data is not tampered with and indeed originates
tant part in the framework of risk management of med-
from that patient. Further, it can not be expected that
ical applications. A crucial issue is the reliability of the
an average person or the medical staff is capable of set-
transmission in order to guarantee that the monitored
ting up and managing authentication and authorization
data is received correctly by the health care profession-
processes. Moreover the network should be accessible
als. The reliability can be considered either end-to-end
when the user is not capable of giving the password (e.g.
or on a per link base. Examples of reliability include
to guarantee accessibility by paramedics in trauma sit-
the guaranteed delivery of data (i.e. packet delivery ra-
uations). Security and privacy protection mechanisms
tio), in-order-delivery, . . . Moreover, messages should
use a significant part of the available energy and should
be delivered in reasonable time. The reliability of the
therefor be energy efficient and lightweight.
network directly affects the quality of patient monitor-
ing and in a worst case scenario it can be fatal when a
life threatening event has gone undetected [31].
4 Positioning WBANs
Intra-body communication
WAN
WMAN
WLAN
WPAN
AN
WB
WBAN
Medical Server BAN MAN
Sensor
Internet
Sensor
Emergency
Personal Device
Personal Device
Extra-body communication Communication Distance
Physician Wireless communication link
Fig. 2 ExampleWBAN
of intra-body and extra-body communication in Fig. 3 Positioning of a Wireless Body Area Network in the realm
a WBAN.
a BAN communicatie of wireless networks.
shown on Figure 2. The former controls the informa- values around 1-2 meters. While a WBAN is devoted
tion handling on the body between the sensors or actu- to interconnection of one person’s wearable devices, a
Medische Server
ators and the personal device [34–37], the latter ensures WPAN is a network in the environment around the
Sensor
communication between the personal device and an ex- person. The communication range can reach up to 10
ternal network [32, 38–40]. Doing so, the medical data meters for high data rate applications and up to sev-
from the patient at home can be consulted by a Spoed physi- eral dozens of meters for low data rate applications. A
Internet
cian or stored in a medical database. This segmentation WLAN has a typical communication range up to hun-
is similar to the one defined in [40] where a multi-tiered dreds of meters. Each type of network has its typical
telemedicine system is presented. Tier 1 encompasses enabling technology, defined by the IEEE. A WPAN
Arts
the intra-body communication, tier 2 the extra-body uses IEEE 802.15.1 (Bluetooth) or IEEE 802.15.4 (Zig-
communication between the personal device and the Bee), a WLAN uses IEEE 802.11 (WiFi) and a WMAN
Internet and tier 3 represents the extra-body commu- IEEE 802.16 (WiMax). The communication in a WAN
nication from the Internet to the medical server. The can be established via satellite links.
combination of intra-body and extra-body communica-
In several papers, Wireless Body Area Networks
tion can be seen as an enabler for ubiquitous health
are considered as a special type of a Wireless Sensor
care service provisioning. An example can be found
Network or a Wireless Sensor and Actuator Network
in [41] where Utility Grid Computing is combined with
(WSAN) with its own requirements1 . However, tradi-
a WBAN. Doing so, the data extracted from the WBAN
tional sensor networks do not tackle the specific chal-
is sent to the grid that provides access to appropriate
lenges associated with human body monitoring. The
computational services with high bandwidth and to a
human body consists of a complicated internal envi-
large collection of distributed time-varying resources.
ronment that responds to and interacts with its exter-
To date, development has been mainly focused on nal surroundings, but is in a way separate and self-
building the system architecture and service platform contained. The human body environment not only has
for extra-body communication. Much of these imple- a smaller scale, but also requires a different type and
mentations focus on the repackaging of traditional sen- frequency of monitoring, with different challenges than
sors (e.g. ECG, heart rate) with existing wireless de- those faced by WSNs. The monitoring of medical data
vices. They consider a very limited WBAN consist- results in an increased demand for reliability. The ease
ing of only a few sensors that are directly and wire- of use of sensors placed on the body leads to a small
lessly connected to a personal device. Further they use form factor that includes the battery and antenna part,
transceivers with a large form factor and large antennas resulting in a higher need for energy efficiency. Sensor
that are not adapted for use on a body. nodes can move with regard to each other, for example
In Figure 3, a WBAN is compared with other types a sensor node placed on the wrist moves in relation to a
of wireless networks, such as Wireless Personal (WPAN), sensor node attached to the hip. This requires mobility
Wireless Local (WLAN), Wireless Metropolitan (WMAN) support. In brief, although challenges faced by WBANs
and Wide Area Networks (WAN) [42]. A WBAN is op-
erated close to the human body and its communication 1In the following, we will not make a distinction between a
range will be restricted to a few meters, with typical WSAN and a WSN although they have significant differences [43].
7
Fig. 4 Characteristics of a Wireless Body Area Network com- The propagation of electromagnetic (EM) waves in the
pared with Wireless Sensor Networks (WSN) and Wireless Local human body has been investigated in [49,50]. The body
Area Network (WLAN). Based on [44]. acts as a communication channel where losses are mainly
due to absorption of power in the tissue, which is dissi-
pated as heat. As the tissue is lossy and mostly consists
are in many ways similar to WSNs, there are intrin-
of water, the EM-waves are attenuated considerably be-
sic differences between the two, requiring special atten-
fore they reach the receiver. In order to determine the
tion. An overview of some of these differences is given
amount of power lost due to heat dissipation, a stan-
in Table 2. A schematic overview of the challenges in a
dard measure of how much power is absorbed in tissue is
WBAN and a comparison with WSNs and WLANs is
used: the specific absorption rate (SAR). It is concluded
given in Figure 4.
that the path loss is very high and that, compared to
the free space propagation, an additional 30-35 dB at
small distances is noticed. A simplified temperature in-
5 Physical layer
crease prediction scheme based on SAR is presented
The characteristics of the physical layer are different in [50]. It is argued that considering energy consump-
for a WBAN compared to a regular sensor network tion is not enough and that the tissue is sensitive to
or an ad-hoc network due to the proximity of the hu- temperature increase. The influence of a patient’s body
man body. Tests with TelosB motes (using the CC2420 shape and position on the radiation pattern from an
transceiver) showed lack of communications between implanted radio transmitter has been studied in [51]. It
nodes located on the chest and nodes located on the is concluded that the difference between body shapes
back of the patient [46]. This was accentuated when (i.e. male, female and child) are at least as large as the
the transmit power was set to a minimum for energy impact of a patient’s arm movements.
savings reasons. Similar conclusions where drawn with
a CC2420 transceiver in [47]: when a person was sitting 5.1.2 Along the Body
on a sofa, no communication was possible between the
chest and the ankle. Better results were obtained when Most of the devices used in a WBAN however are at-
the antenna was placed 1 cm above the body. As the tached on the body. The propagation along the human
devices get smaller and more ubiquitous, a direct con- body can be divided into line of sight (LOS) and non-
nection to the personal device will no longer be possible line of sight (NLOS) situations. In the former, the cur-
and more complex network topologies will be needed. vature effects of the body are not taken into account as
In this section, we will discuss the characteristics of the simulations are performed on a flat phantom or exper-
propagation of radio waves in a WBAN and other types iments are done at one side of the body. In the latter,
of communication. the effect of propagation from the front of the body to
the side or back are evaluated.
The channel model for line of sight (LOS) propaga-
5.1 RF communication tion along the human body was studied in [24, 52–55],
both by simulations and experiments. The studies were
Several researchers have been investigating the path done for both narrowband and UWB signals. However,
loss along and inside the human body either using nar- the results can be compared as the studies for UWB
rowband radio signals or Ultra Wide Band (UWB). All signals were performed in a band between 3 to 6 GHz
8
Table 2 Schematic overview of differences between Wireless Sensor Networks and Wireless Body Area Networks, based on [45].
Challenges Wireless Sensor Network Wireless Body Area Network
Scale Monitored environment (meters / kilometers) Human body (centimeters / meters)
Node Number Many redundant nodes for wide area coverage Fewer, limited in space
Result accuracy Through node redundancy Through node accuracy and robustness
Node Tasks Node performs a dedicated task Node performs multiple tasks
Node Size Small is preferred, but not important Small is essential
Network Topology Very likely to be fixed or static More variable due to body movement
Data Rates Most often homogeneous Most often heterogeneous
Node Replacement Performed easily, nodes even disposable Replacement of implanted nodes difficult
Node Lifetime Several years / months Several years / months, smaller battery capac-
ity
Power Supply Accessible and likely to be replaced more easily Inaccessible and difficult to replaced in an im-
and frequently plantable setting
Power Demand Likely to be large, energy supply easier Likely to be lower, energy supply more difficult
Energy Scavenging Source Most likely solar and wind power Most likely motion (vibration) and thermal
(body heat)
Biocompatibility Not a consideration in most applications A must for implants and some external sensors
Security Level Lower Higher, to protect patient information
Impact of Data Loss Likely to be compensated by redundant nodes More significant, may require additional mea-
sures to ensure QoS and real-time data delivery.
Wireless Technology Bluetooth, ZigBee, GPRS, WLAN, . . . Low power technology required
5.2 Movement of the Body sor Networks has some points in common with network-
ing in WBANs, it is useful to consider the research in
The movement of the body plays an important role in MAC-protocols designed for WSNs. An overview can be
the strength of the received signal. In [58] it is shown found in [66, 67]. Two major categories are contention-
that arm motions to the front and side of the body can based and schedule-based. For the former, CSMA/CA
have a small impact on the received power. More sig- is a typical example, while TDMA is a typical scheme
nificant variations are found when the arms are moved for the latter. The advantages of contention-based ap-
so that they block the line of sight between the two proaches are the simplicity, its infrastructure-free ad
antennas. In [59] a preliminary system model for gait hoc feature and good adaptability to traffic fluctuation,
analysis has been proposed. It is concluded that signifi- especially for low load. Schedule-based approaches on
cant attenuation can occur (up to 20 dB) when a body the other hand are free of idle listening, overhearing and
limb is moved in between the Tx and Rx antenna. Ac- packet collisions because of the lack of medium com-
cording to [60] the movement of the limbs can induce petition, but require tight time synchronization. The
an attenuation of 30 dB or more. A similar conclusion most commonly used technique for reducing energy con-
was found in an actual implementation [37] where the sumption in contention-based protocols is controlling
sensors communicate directly with the personal device the power and duty cycle of the radio.
using an RF-radio operating at 868 MHz. Loss rates Some implementations of WBANs use Bluetooth
of more than 50% where found when the body was in (IEEE 802.15.1) [68]. This was developed as a cable
motion. replacement and does not support (or only very lim-
ited) multi-hop communication. It has a complex pro-
tocol stack and a high energy consumption compared
5.3 Non-RF Communication to IEEE 802.15.4. It is therefore not suited to be used
in a WBAN.
Next to the propagation of radio waves, several re-
Most current implementations of WBANs use IEEE
searchers have investigated the possibility to transfer
802.15.4 [69] or ZigBee [70] as enabling technology. As
electronic data by capacitive and galvanic coupling, also
most of the radios used in a WBAN are based on an
called body-coupled communication (BCC). These ra-
IEEE 802.15.4 compliant chip set, some researchers have
dios work at low frequencies (ranging from 10 kHz to
adapted the IEEE 802.15.4 MAC-protocol to make it
10 MHz). Zimmerman [61] first showed the potential
more suitable for WBANs. We will therefore first dis-
of interference-free ultra low power data communica-
cuss the usefulness of IEEE 802.15.4 for medical net-
tion through the human body. High variations of the
working. In a second part, other MAC-protocols for
transmission attenuation have been observed at differ-
WBANs will discussed. An overview is given in Table 3.
ent locations of the body. Galvanic coupling promises
It can be noticed that all proposed MAC-protocols use
to be a potential communication technology for sensor
slotted communication and assume a star topology by
application on the thorax and for short distances on the
using master-slave communication. However, in the pre-
limbs [62]. This technology can also be used to exchange
vious section it has been shown that single-hop commu-
data from one body to another by for example shaking
nication is not always possible.
hands [63]. In [64] OsteoConduct is presented, where
the human musculoskeletal system is used to trans-
mit data and information in a low-power, secure, non-
intrusive fashion. Although this research looks promis- 6.1 IEEE 802.15.4
ing, only very low data rates can be achieved (5 bits/s).
The idea of BCC is further exploited by [65] for In [71] the star network configuration of the IEEE 802.15.4
bootstrapping WBANs. They argue to equip the nodes standard at 2.4 GHz was considered for a WBAN. The
with both RF and BCC capabilities. As a BCC is re- analysis considers quite extensively a very low data rate
stricted to a person’s body, the BCC can be used to dis- star network with 10 body implanted sensors transmit-
cover and identify sensor nodes on the same body and ting data 1 to 40 times per hour. The analysis focuses
for waking up RF radios from low-power sleep mode. on the effect of crystal tolerance, frame size and the
usage of IEEE 802.15.4 Guaranteed Time Slots (GTS)
on a node lifetime. The main consideration in this work
6 MAC layer was the long-term power consumption of devices. The
results show that IEEE 802.15.4 provides a limited an-
The number of MAC-protocols specifically developed swer for medical sensor networking when configured in
for WBANs is limited. As networking in Wireless Sen- non-beacon mode with low data rate asymmetric traf-
10
MAC-protocol IEEE 802.15.4 TDMA based CSMA based Star topology Time Synchronization available
based (master/slave) in the protocol
√ √ √
Timmons [71]
√ √
BSN-MAC [72] mixed
√ √ √
Lamprinos [73]
√ √
Omeni [74]
√ √ √
H-MAC [75]
fic. Beacon mode can also be used, but with even more at one of the nodes, the node can be assigned an extra
severe restrictions on data rate and crystal tolerance. slot for direct communication. The protocol has been
Another adaptation is BSN-MAC [72]. The coordi- evaluated on a Sensium platform. The H-MAC proto-
nator controls the communication by varying the super- col [75] uses the human heartbeat rhythm information
frame structure of IEEE 802.15.4. This divides the time to perform time synchronization for TDMA. The bio-
axis in a contention-free and contention-based period. sensors can thus achieve time synchronization without
The sensors provide real-time feedback to a BSN co- having to turn on their radio. The algorithm is veri-
ordinator with application-specific and sensor-specific fied with real world data but assumes a certain buffer.
information. Hence, based on the feedback the BSN co- The simulations do not show the energy gain and the
ordinator can make dynamic adjustments for the length protocol is designed for a star-topology WBAN only.
of the contention-free and contention-based period to
achieve better performance in energy efficiency and la-
tency.
Both [76] and [77] come to the conclusion that al- 6.3 IEEE 802.15.6
though 802.15.4 can provide QoS, the technology is not
scalable in terms of power consumption and can not be Started as a Study Group in 2006 and motivated by the
used as a single solution for all WBAN applications. increasing research and industry interest in WBANs,
It can be concluded that IEEE 802.15.4 is not the the IEEE Standards Association decided to form the
best solution for supporting communication in WBANs. IEEE 802.15 Task Group 6 in November 2007. It de-
Although it can be used for a quick (and easy) imple- scribes itself as follows: The IEEE 802.15 Task Group
mentation, the results are rather poor. IEEE 802.15.4 6 (BAN) is developing a communication standard op-
was not designed to support WBANs. Specialized MAC timized for low power devices and operation on, in or
protocols are needed. around the human body (but not limited to humans) to
serve a variety of applications including medical, con-
sumer electronics / personal entertainment and other
6.2 WBAN Specific Protocols [13].
Project Authorization Request (PAR) 07-0575 presents
One of the few MAC-protocols for WBANs was pro- an extended description of the task group [79]. It stresses
posed by Lamprinos et al. [73]. They use a master-slave the fact that current WPANs do not meet medical com-
architecture and, to avoid idle listening, all slaves are munication guidelines, because of the proximity to hu-
locked in the Rx-slot of the master and go in standby man tissue. Moreover, WPAN technology is said not to
at the same time. The main drawback of this protocol support Quality of Service, low power operation and
is that some slaves will have a low duty cycle whereas noninterference, all required to support WBAN appli-
the nodes that are serviced later have a higher duty cy- cations. Based on the responses to the Call for Appli-
cle. The protocol was implemented nor simulated. An cations [80], the PAR also outlines a large number of
adaptation of this protocol was used in [78]. This pro- applications that can be served by the proposed stan-
tocol divides time into frames in which only one node is dard, going from classical medical usage, e.g. EEG and
allowed to transmit. The scheduling order is derived by ECG monitoring, to personal entertainment systems.
applying the Earliest Deadline First algorithm. Omeni In 2008, a Call for Proposals on physical layer and
et al. [74] propose a MAC protocol for a star-networked MAC layer protocols was issued [81]. The large num-
WBAN that supports TDMA to reduce the probabil- ber of responses, 64 in total, confirmed the industry
ity of collision and idle listening. Each slave node is as- interest. Currently, the responses are being evaluated
signed a slot by the central node. When an alarm occurs at monthly meetings, while some proposals are merged.
11
L H L
7 Network layer
H H
Developing efficient routing protocols in WBANs is a
nontrivial task because of the specific characteristics D
L
of the wireless environment. First of all, the available
bandwidth is limited, shared and can vary due to fad- Low-temperature Destination
ing, noise and interference, so the protocol’s amount node
Sender
of network control information should be limited. Sec-
ondly, the nodes that form the network can be very Fig. 6 An example of LTR and ALTR. The white arrows indi-
heterogeneous in terms of available energy or comput- cate the LTR-path. The shaded arrows show the adapted path
ing power. of ALTR. When the path has three hops, the routing algorithm
switches to shortest path routing.
Although a lot of research is being done toward en-
ergy efficient routing in ad hoc networks and WSNs [82],
the proposed solutions are inadequate for WBANs. For (hot spots) [50]. Packets are withdrawn from heated
example, in WSNs maximal throughput and minimal zones and rerouted through alternate paths. TARA suf-
routing overhead are considered to be more important fers from low network lifetime, a high ratio of dropped
than minimal energy consumption. Energy efficient ad- packets and does not take reliability into account. An
hoc network protocols only attempt to find routes in improvement of TARA is Least Temperature Routing
the network that minimize energy consumption in ter- (LTR) and Adaptive Least Temperature Routing (ALTR)
minals with small energy resources, thereby neglecting [84] that reduces unnecessary hops and loops by main-
parameters such as the amount of operations (measure- taining a list in the packet with the recently visited
ments, data processing, access to memory) and energy nodes. ALTR switches to shortest hop routing when
required to transmit and receive a useful bit over the a predetermined number of hops is reached in order
wireless link. Most protocols for WSNs only consider to lower the energy consumption. An example of LTR
networks with homogeneous sensors and a many-to-one and ALTR is given in Fig. 6. A smarter combination
communication paradigm. In many cases the network of LTR and shortest path routing is Least Total Route
is considered as a static one. In contrast, a WBAN has Temperature (LTRT) [36]. The node temperatures are
heterogeneous mobile devices with stringent real-time converted into graph weights and minimum tempera-
requirements due to the sensor-actuator communica- ture routes are obtained. A better energy efficiency and
tion. Specialized protocols for WBANs are therefore a lower temperature rise is obtained, but the protocol
needed. has as main disadvantage that a node needs to know the
In the following, an overview of existing routing temperature of all nodes in the network. The overhead
strategies for WBANs is given. They can be subdivided of obtaining this data was not investigated.
in two categories: routing based on the temperature of
the body and cluster based protocols.
7.2 Cluster Based Routing
7.1 Temperature Routing “Anybody” [35] is a data gathering protocol that uses
clustering to reduce the number of direct transmissions
When considering wireless transmission around and on to the remote base station. It is based on LEACH [85]
the body, important issues are radiation absorption and that randomly selects a cluster head at regular time in-
heating effects on the human body. To reduce tissue tervals in order to spread the energy dissipation. The
heating the radio’s transmission power can be limited cluster head aggregates all data and sends it to the
or traffic control algorithms can be used. In [83] rate base station. LEACH assumes that all nodes are within
control is used to reduce the bioeffects in a single-hop sending range of the base station. Anybody solves this
network. Another possibility is a protocol that balances problem by changing the cluster head selection and
the communication over the sensor nodes. An exam- constructing a backbone network of the cluster heads.
ple is the Thermal Aware Routing Algorithm (TARA) The energy efficiency is not thoroughly investigated and
that routes data away from high temperature areas reliability is not considered. Another improvement of
12
LEACH is Hybrid Indirect Transmissions (HIT) [86], In [92] the reliability of CICADA was evaluated and
which combines clustering with forming chains. Doing additional mechanisms were proposed in order to im-
so, the energy efficiency is improved. Reliability, how- prove the reliability even further, such as the random-
ever, is not considered. ization of schemes and overhearing the control messages
This overview clearly shows that routing protocols sent by the siblings.
for WBANs is an emerging area of research, the pro- BodyQos [93] addresses three unique challenges brought
tocols described above were only developed in the last by BSN applications. It uses an asymmetric architec-
two years. ture where most of the processing is done at the cen-
tral device. Second, they have developed a virtual MAC
(V-MAC) that can support a wide variety of different
8 Cross-layer Protocols MACs. Third, an adaptive resource scheduling strategy
is used in order to make it possible to provide statistical
Cross-layer design is a way to improve the efficiency bandwidth guarantees as well as reliable data communi-
of and interaction between the protocols in a wireless cation in WBANs. The protocol has been implemented
network by combining two or more layers from the pro- in nesC on top of TinyOS.
tocol stack. This research has gained a lot of interest The desired quality of service will affect the energy
in sensor networks [87, 88]. However, little research has consumption. For example, to obtain a lower packet
been done for WBANs. loss, the transmit power can be increased, which raises
Ruzelli et al. propose a cross-layer energy efficient the energy consumption. It is therefore important to
multi-hop protocol built on IEEE 802.15.4 [46]. The achieve the right balance between power consumption
network is divided into time zones where each one takes and the desired reliability of the system.
turn in the transmission. The nodes in the farthest
timezone start the transmission. In the next slot, the
farthest but one sends its data and so on until the sink 10 Security
is reached. The protocol almost doubles the lifetime
compared to regular IEEE 802.15.4. The protocol was The communication of health related information be-
developed for regular sensor networks, but the authors tween sensors in a WBAN is subject to the follow-
claim its usefulness for WBANs. ing security requirements: data confidentiality, data au-
CICADA [34] uses a data gathering tree and con- thenticity, data integrity and data freshness [94]. Data
trols the communication using distributed slot assign- confidentiality means that the transmitted information
ment. It has low packet loss and high sleep ratios while is strictly private and can only be accessed by autho-
the network flexibility is preserved. It also enables two- rized persons, e.g. the doctor attending the patient. It is
way communication. Data-aggregation and the use of usually achieved by encrypting the information before
a duty cycle even further improved the lifetime of the sending it using a secret key and can be both symmet-
network. rically and asymmetrically. Data authenticity provides
Another approach for cross layering is completely a means for making sure that the information is sent by
discarding the layered structure and implementing the the claimed sender. For this, a Message Authentication
required functionality in different modules which inter- Code (MAC 3 ) is calculated using a shared secret key.
act and can be changed easily [89]. A first attempt for Data integrity makes sure that the received information
WBANs using this method is described in [90]. has not been tampered with. This can be inspected by
verifying the MAC. Data freshness guarantees that the
received data is recent and not a replayed old message
9 Quality of Service
to cause disruption. A much used technique is to add a
counter which is increased every time a message is sent.
The research on QoS solutions is extensive for gen-
The security mechanisms employed in WSNs do gen-
eral ad hoc networks. However, these QoS solutions
erally not offer the best solutions to be used in WBANs
are designed for more powerful devices which are of-
for the latter have specific features that should be taken
ten line-powered. Most of these solutions do not apply
into account when designing the security architecture.
to WSN or WBAN applications. Several QoS solutions
The number of sensors on the human body, and the
specific for WSNs have been proposed, but these solu-
range between the different nodes, is typically quite
tions mainly focus on one or a few QoS features such
limited. Furthermore, the sensors deployed in a WBAN
as reliability, delay, bandwidth specification or reser-
vation [91]. For WBANs, researchers have shown little 3 MAC is written in italic in order to avoid confusion with the
effort to provide QoS solutions. abbreviation of Medium Access Control
13
are under surveillance of the person carrying these de- 11 Existing Projects
vices. This means that it is difficult for an attacker to
physically access the nodes without this being detected. Several research groups and commercial vendors are al-
When designing security protocols for WBANs, these ready developing the first prototypes of WBANs. How-
characteristics should be taken into account in order ever, this research mainly focuses on building a sys-
to define optimized solutions with respect to the avail- tem architecture and service platform and in lesser ex-
able resources in this specific environment. Although tent on developing networking protocols. In this sec-
providing adequate security is a crucial factor in the tion, we provide a non-exhaustive overview of projects
acceptance of WBANs, little research has been done in for WBANs.
this specific field. One of the most crucial components Otto et al. [6] and Jovanov et al. [32] present a sys-
to support the security architecture is its key manage- tem architecture which both handles the communica-
ment. Further, security and privacy protection mech- tion within the WBAN and between the WBANs and
anisms use a significant part of the available resources a medical server in a multi-tier telemedicine system.
and should therefore be energy efficient and lightweight. The communication between the sensors and the sink
A solution for data integrity and freshness was pro- is single-hop, slotted and uses ZigBee or Bluetooth. The
posed in [95]. Their integrity algorithm is based on the slots are synchronized using beacons periodically sent
measurement of a permissible round trip time threshold by the sink. They use off-the-shelf wireless sensors to
and is computational feasible. Authentication is done design a prototype WBAN such as the Tmote sky plat-
by calculating a MAC with a random sequence of num- form from formerly Moteiv [100], now sentilla [101].
bers. This sequence is determined at the initialization The Tmote sky platform is also used in the CodeBlue-
phase. project [102,103] where WBANs are used in rapid disas-
ter response scenarios. A wearable computer attached
In [96] a security mechanism was added to CICADA.
to the patient’s wrist, i.e. a Tmote Sky mote, forms
Doing so, CICADA-S became one of the first proto-
an ad hoc wireless network with a portable tablet PC.
cols where appropriate security mechanisms are incor-
They developed a wireless two-lead ECG, a wireless
porated into the communication protocol while address-
pulse oximeter sensor and a wireless electromyogram
ing the life-cycle of the sensors. It was shown that the
(EMG).
integration of key management and secure, privacy pre-
serving communication techniques has low impact on Ayushman [104] is a sensor network based medical
the power consumption and throughput. monitoring infrastructure that can collect, query and
analyze patient health information in real-time. A wire-
Another promising solution for key management is less ECG, gait monitoring and environment monitoring
the use of biometrics. Biometrics is a technique com- was developed using off-the-shelf components with a
monly known as the automatic identification and ver- Mica2 wireless transceiver. Further, the necessary soft-
ification of an individual by his or her physiological ware for consulting the data at a remote client was de-
and/or behavioral characteristics [97]. In [12] an algo- veloped.
rithm based on biometric data is described that can
The Human++ project by IMEC-NL [10] aims “to
be employed to ensure the authenticity, confidentiality
achieve highly miniaturized and autonomous sensor sys-
and integrity of the data transmission between the per-
tems that enable people to carry their personal body area
sonal device and all other nodes. Algorithms that use
network.”. An ambulatory EEG/ECG system with a
the heartbeat to generate a key are proposed in [98,99].
transmitter working on 2.4 GHz was developed. This
In [65] body-coupled communication (BCC) is used system can run for approximately 3 months using 2 AA
to associate new sensors in a WBAN. As BCC is limited batteries. In order to obtain a longer autonomy, the
to the body, this techniques can be used to authenticate project also investigates energy scavenging with ther-
new sensors on the body. moelectric generators (TEG). In 2006, a wireless pulse
The developers of WBANs will have to take into oximeter was presented, fully powered by the patient’s
account the privacy issues. After all, a WBAN can be body heat. Further, the project investigates new wire-
considered as a potential threat to freedom, if the appli- less technologies such as UWB to make an ultra-low
cations go beyond “secure” medical usage, leading to a power transmitter.
Big Brother society. Social acceptance would be the key The European MobiHealth project [105] provides a
to this technology finding a wider application. There- complete end-to-end mHealth platform for ambulant
fore, considerable effort should be put in securing the patient monitoring, deployed over UMTS and GPRS
communication and making sure that only authorized networks. The MobiHealth patient/user is equipped with
persons can access the data. different sensors that constantly monitor vital signals,
14
e.g. blood pressure, heart rate and electrocardiogram a lot of open research issues. On the data link layer,
(ECG). Communication between the sensors and the more WBAN specific MAC-protocols need to be devel-
personal device is Bluetooth or ZigBee based and is oped that take into account the movement of the body,
single-hop. The major issues considered are security, i.e. the mobility of the nodes, additional low-power fea-
reliability of communication resources and QoS guar- tures such as an adaptive duty cycle for lowering the idle
antees. listening and overhearing, the use of the human physi-
The French project BANET [106] aims to provide ology such as heart beat to ensure time synchronization
a framework, models and technologies to design opti- and so on. Concerning the network layer, a promising
mized wireless communication systems targeting the research track is the combination of thermal routing
widest range of WBAN-based applications, in the con- with more energy efficient mechanisms. More efficient
sumer electronics, medical and sport domains. They fo- QoS-mechanisms are needed, for example based on the
cus on the study of the WBAN propagation channel, BodyQos framework. Other interesting open research
MAC protocols and coexistence of WBANs and other issues are mobility support embedded in the protocol,
wireless networks. security, inter operability and so on. In order to define a
The German BASUMA-project (Body Area System globally optimal system, it might be necessary to unite
for Ubiquitous Multimedia Applications) [107] aims at several of these mechanisms in a cross-layer protocol.
developing a full platform for WBANs. As communica- The use energy scavenging was not addressed in de-
tion technique, a UWB-frontend is used and a MAC- tail in this paper but is nevertheless important. With
protocol based on IEEE 802.15.3. This protocol also a smart combination of lower energy protocols and en-
uses time frames divided into contention free periods ergy scavenging, the optimal solution for achieving au-
(with time slots) and contention access periods (CSMA/CA).
tonomous Body Area Networks can be reached. For a
A flexible and efficient WBASN solution suitable for WBAN, energy scavenging from on-body sources such
a wide range of applications is developed in [108]. The as body heat and body vibration seems very well suited.
focus lies on posture and activity recognition applica- The ultimate goal is to create a small and smart band-
tions by means of practical implementation and on-the- aid containing all necessary technology for sensing and
field testing. The sensors are WiMoCA-nodes, where communication with a base station. Very preliminary
sensors are represented by tri-axial integrated MEMS examples can be found in the Sensium-platform [74]
accelerometers. and the Human++-project [10].
The Flemish IBBT IM3-project (Interactive Mobile
Medical Monitoring) focuses on the research and im-
plementation of a wearable system for health monitor-
ing [109]. Patient data is collected using a WBAN and
analyzed at the medical hub worn by the patient. If an 13 Conclusions
event (e.g. heart rhythm problems) is detected, a signal
is sent to a health care practitioner who can view and In this survey, we have reviewed the current research
analyze the patient data remotely. on Wireless Body Area Networks. In particular, this
work presents an overview of the research on the prop-
12 Open Research Issues agation in and on the human body, MAC-protocols,
routing protocols, Quality of Service and security. To
The discussions above clearly show that, although a lot conclude, a list of research projects is given and open
of research is going on, still a lot of open issues exist. research issues are discussed.
Several researchers have already started studying A WBAN is expected to be a very useful technol-
the propagation of electromagnetic waves in and on the ogy with potential to offer a wide range of benefits to
body and a few models for the physical layer are pro- patients, medical personnel and society through contin-
posed. It should be noticed that none of them take the uous monitoring and early detection of possible prob-
movements of the body into account, although move- lems. With the current technological evolution, sensors
ments can have severe impact on the received signal and radios will soon be applied as skin patches. Do-
strength, as described in Section 5.2. Further, new emerg- ing so, the sensors will seamlessly be integrated in a
ing technologies such as galvanic coupling and trans- WBAN. Step by step, these evolutions will bring us
formation of information via the bones offer promising closer to a fully operational WBAN that acts as an en-
results and need to be investigated more thoroughly. abler for improving the Quality of Life. We feel that
Although some protocols already exist that take care this review can be considered as a source of inspiration
of the data link layer and networking, this area still has for future research directions.
15
for computer assisted physical rehabilitation,” Journal of for healthcare and assisted living environments. New York,
NeuroEngineering and Rehabilitation, vol. 2, no. 1, pp. 16– NY, USA: ACM, 2007, pp. 37–42.
23, March 2005. 47. R. C. Shah and M. Yarvis, “Characteristics of on-body
33. A. Bhargava and M. Zoltowski, “Sensors and wireless com- 802.15.4 networks,” in Wireless Mesh Networks, 2006.
munication for medical care,” in Database and Expert Sys- WiMesh 2006. 2nd IEEE Workshop on, Reston, VA, USA,,
tems Applications, 2003. Proceedings. 14th International 2006, pp. 138–139.
Workshop on, Sep. 2003, pp. 956–960. 48. T. S. Rappaport, Wireless Communication: Principles and
34. B. Latré, B.Braem, I.Moerman, C. Blondia, E. Reusens, Practice 2nd edition. Prentice Hall, 2002.
W. Joseph, and P. Demeester, “A low-delay protocol for 49. S. K. S. Gupta, S. Lalwani, Y. Prakash, E. Elsharawy, and
multihop wireless body area networks,” in 4th Annual In- L. Schwiebert, “Towards a propagation model for wireless
ternational Conference on Mobile and Ubiquitous Systems: biomedical applications,” in Communications, 2003. ICC
Networking & Services, 2007, Workshop PerNets, Philadel- ’03. IEEE International Conference on, vol. 3, May 2003,
phia, PA, USA, 6-10 August 2007, pp. 479–486. pp. 1993–1997.
35. T. Watteyne, S. Augé-Blum, M. Dohler, and D. Barthel, 50. Q. Tang, N. Tummala, S. K. S. Gupta, and L. Schwiebert,
“Anybody: a self-organization protocol for body area net- “Communication scheduling to minimize thermal effects
works,” in Second International Conference on Body Area of implanted biosensor networks in homogeneous tissue,”
Networks (BodyNets), Florence, Italy, 11-13 June 2007. IEEE Transactions on Biomedical Engineering, vol. 52,
36. D. Takahashi, Y. Xiao, F. Hu, J. Chen, and Y. Sun, no. 7, pp. 1285–1294, Jul. 2005.
“Temperature-aware routing for telemedicine applications 51. A. J. Johansson, “Wave-propagation from medical
in embedded biomedical sensor networks,” EURASIP Jour- implants-influence of body shape on radiation pattern,”
nal on Wireless Communications and Networking, vol. in 24th Annual Conference and the Annual Fall Meeting
2008, no. Article ID 572636, 2008, 11 pages. of the Biomedical Engineering Society, Proceedings of the
37. A. Ylisaukko-oja, E. Vildjiounaite, and J. Mantyjarvi, Second Joint EMBS/BMES Conference, vol. 2, 2002, pp.
“Five-point acceleration sensing wireless body area network 1409–1410.
- design and practical experiences,” iswc, vol. 00, pp. 184– 52. E. Reusens, W. Joseph, G. Vermeeren, L. Martens, B. Latré,
185, 2004. B. Braem, C. Blondia, and I. Moerman, “Path-loss models
38. N. T. Dokovski, A. T. van Halteren, and I. A. Widya, “Ba-
for wireless communication channel along arm and torso:
nip: Enabling remote healthcare monitoring with body area
Measurements and simulations,” in IEEE Antennas and
networks,” in FIDJI 2003 International Workshop on Sci-
Propagation Society International Symposium 2007, Hon-
entific Engineering of Distributed Java Applications, Lux-
olulu, HI, USA, 9-15 June 2007, pp. 336–339.
embourg, ser. Lecture notes in Computer Science, N. Guelfi,
53. L. Roelens, S. Van den Bulcke, W. Joseph, G. Vermeeren,
E. Astesiano, and G. Reggio, Eds., vol. 2952/2004. Heidel-
and L. Martens, “Path loss model for wireless narrowband
berg: Springer Verlag, 2004, pp. 62–72.
communication above flat phantom,” Electronics Letters,
39. K. E. Wac, R. Bults, A. van Halteren, D. Konstantas, and
vol. 42, no. 1, pp. 10–11, Jan. 2006.
V. F. Nicola, “Measurements-based performance evalua-
54. T. Zasowski, G. Meyer, F. Althaus, and A. Wittneben,
tion of 3g wireless networks supporting m-health services,”
“Propagation effects in UWB body area networks,” in Ultra-
in Multimedia Computing and Networking 2005. Edited
Wideband, 2005. ICU 2005. 2005 IEEE International Con-
by Chandra, Surendar; Venkatasubramanian, Nalini. Pro-
ference on, Sep. 2005, pp. 16–21.
ceedings of the SPIE, Volume 5680, pp. 176-187 (2004).,
55. A. Fort, J. Ryckaert, C. Desset, P. De Doncker,
S. Chandra and N. Venkatasubramanian, Eds., Dec. 2004,
P. Wambacq, and L. Van Biesen, “Ultra-wideband channel
pp. 176–187.
40. A. Milenkovic, C. Otto, and E. Jovanov, “Wireless sensor model for communication around the human body,” IEEE
networks for personal health monitoring: Issues and an im- Journal on Selected Areas in Communications, vol. 24, pp.
plementation,” Computer Communications, Wireless Sen- 927–933, Apr. 2006.
sor Networks and Wired/Wireless Internet Communica- 56. T. Zasowski, G. Meyer, F. Althaus, and A. Wittneben,
tions, vol. 29, no. 13-14, pp. 2521–2533, August 2006. “UWB signal propagation at the human head,” IEEE
41. O. O. Olugbara, M. O. Adigun, S. O. Ojo, and P. Mudali, Transactions on Microwave Theory and Techniques, Apr.
“Utility grid computing and body area network as enabler 2006.
for ubiquitous rural e-healthcare service provisioning,” in 57. B. Braem, B. Latré, I. Moerman, C. Blondia, E. Reusens,
e-Health Networking, Application and Services, 2007 9th W. Joseph, L. Martens, and P. Demeester, “The need for
International Conference on, Taipei, Taiwan,, Jun. 2007, cooperation and relaying in short-range high path loss sen-
pp. 202–207. sor networks,” in First International Conference on Sen-
42. I. Chlamtac, M. Conti, and J. Liu, “Mobile ad hoc network- sor Technologies and Applications (SENSORCOMM 2007),
ing: imperatives and challenges.” Ad Hoc Networks, vol. 1, Valencia, Spain, 14-20 October 2007, pp. 566–571.
no. 1, pp. 13–64, 2003. 58. A. Fort, C. Desset, J. Ryckaert, P. De Doncker,
43. I. F. Akyildiz and I. H. Kasimoglu, “Wireless sensor and ac- L. Van Biesen, and P. Wambacq, “Characterization of the
tor networks: research challenges,” Ad Hoc Networks, vol. 2, ultra wideband body area propagation channel,” in Ultra-
no. 2, pp. 351–367, October 2004. Wideband, 2005. ICU 2005. 2005 IEEE International Con-
44. T. Zasowski, “A system concept for ultra wideband (UWB) ference on, Sep. 2005.
body area networks,” Ph.D. dissertation, PhD Thesis, ETH 59. M. Di Renzo, R. M. Buehrer, and J. Torres, “Pulse shape
Zürich, No. 17259, 2007. distortion and ranging accuracy in uwbbased body area net-
45. G.-Z. Yang, Ed., Body Sensor Networks. Springer-Verlag works for fullbody motion capture and gait analysis,” in
London Limited, 2006. IEEE Globecom 2007, November 2007, pp. 3775 – 3780.
46. A. G. Ruzzelli, R. Jurdak, G. M. O’Hare, and P. V. D. Stok, 60. D. Neirynck, “Channel characterisation and physical layer
“Energy-efficient multi-hop medical sensor networking,” in analysis for body and personal area network development,”
HealthNet ’07: Proceedings of the 1st ACM SIGMOBILE Ph.D. dissertation, University of Bristol, UK, November
international workshop on Systems and networking support 2006.
17
61. T. Zimmerman, “Personal area networks: Nearfield intra- 77. D. Cavalcanti, R. Schmitt, and A. Soomro, “Performance
body communication,” IBM Systems Journal, vol. 35, no. 3, analysis of 802.15.4 and 802.11e for body sensor network
pp. 609–617, 1996. applications,” in 4th International Workshop on Wearable
62. M. S. Wegmueller, A. Kuhn, J. Froehlich, M. Oberle, N. Fel- and Implantable Body Sensor Networks (BSN 2007), vol.
ber, N. Kuster, and W. Fichtner, “An attempt to model the Volume 13. Springer Berlin Heidelberg, 2007, pp. 9–14.
human body as a communication channel,” IEEE Transac- 78. E. Farella, A. Pieracci, L. Benini, and A. Acquaviva, “A
tions on Biomedical Engineering, vol. 54, no. 10, pp. 1851– wireless body area sensor network for posture detection,”
1857, Oct. 2007. in ISCC ’06: Proceedings of the 11th IEEE Symposium on
63. K. Hachisuka, Y. Terauchi, Y. Kishi, T. Hirota, K. Sasaki, Computers and Communications. Washington, DC, USA:
H. Hosaka, and K. Ito, “Simplified circuit modeling and IEEE Computer Society, 2006, pp. 454–459.
fabrication of intrabody communication devices,” in Solid- 79. B. Heile, “IEEE 802.15 TG 6 PAR,” IEEE15-07-0575/r9,
State Sensors, Actuators and Microsystems, 2005. Digest IEEE-SA, December 2007.
of Technical Papers. TRANSDUCERS ’05. The 13th In- 80. D. Lewis, “802.15 TG 6 Call for Applications - Response
ternational Conference on, vol. 1, Jun. 2005, pp. 461–464. Summary,” IEEE15-08-0407r6, IEEE-SA, July 2008.
64. L. Zhong, D. El-Daye, B. Kaufman, N. Tobaoda, T. Mo- 81. A. Astrin, “802.15 TG 6 Call for Proposals (CFP),”
hamed, and M. Liebschner, “Osteoconduct: Wireless body- IEEE15-08-0829r1, IEEE-SA, November 2008.
area communication based on bone conduction,” in Proc. 82. K. Akkaya and M. Younis, “A survey on routing proto-
Int. Conf. Body Area Networks (BodyNets), June 2007. cols for wireless sensor networks,” Ad Hoc Networks, vol. 3,
65. T. Falck, H. Baldus, J. Espina, and K. Klabunde, “Plug no. 3, pp. 325–349, 2005.
83. H. Ren and M. Q. H. Meng, “Rate control to reduce bioef-
’n play simplicity for wireless medical body sensors,” Mob.
fects in wireless biomedical sensor networks,” in Mobile and
Netw. Appl., vol. 12, no. 2-3, pp. 143–153, 2007.
66. I. Demirkol, C. Ersoy, and F. Alagoz, “MAC protocols for Ubiquitous Systems - Workshops, 2006. 3rd Annual Inter-
wireless sensor networks: a survey,” IEEE Communications national Conference on, San Jose, CA,, Jul. 2006, pp. 1–7.
84. A. Bag and M. A. Bassiouni, “Energy efficient thermal
Magazine, vol. 44, no. 4, pp. 115–121, Apr. 2006.
aware routing algorithms for embedded biomedical sensor
67. P. Baronti, P. Pillai, V. Chook, S. Chessa, A.Gotta, and
networks,” in Mobile Adhoc and Sensor Systems (MASS),
Y. F. Hu, “Wireless sensor networks: A survey on the state
2006 IEEE International Conference on, Vancouver, BC,,
of the art and the 802.15.4 and zigbee standards,” Computer
Oct. 2006, pp. 604–609.
Communications, vol. 30, no. 7, pp. 1665–1695, May 2007.
85. W. R. Heinzelman, A. Chandrakasan, and H. Balakrish-
68. P. Johansson, M. Kazantzidis, R. Kapoor, and M. Gerla,
nan, “Energy-efficient communication protocol for wireless
“Bluetooth: an enabler for personal area networking,” IEEE
microsensor networks,” in System Sciences, 2000. Proceed-
Network, vol. 15, no. 5, pp. 28–37, Sep./Oct. 2001.
ings of the 33rd Annual Hawaii International Conference
69. IEEE 802.15.4-2003: IEEE Standard for Information Tech-
on, Jan 2000, pp. 8020–8024.
nology - Part 15.4: Wireless Medium Access Control and
86. M. Moh, B. J. Culpepper, L. Dung, T.-S. Moh, T. Hamada,
Physical Layer specifications for Low Rate Wireless Per-
and C.-F. Su, “On data gathering protocols for in-body
sonal Area Networks.
biomedical sensor networks,” in Global Telecommunica-
70. ZigBee Alliance, official webpage: http://www.zigbee.org.
tions Conference, 2005. GLOBECOM ’05. IEEE, vol. 5,
71. N. F. Timmons and W. G. Scanlon, “Analysis of the per-
Nov./Dec. 2005.
formance of IEEE 802.15.4 for medical sensor body area 87. R. Madan, S. Cui, S. Lall, and N. A. Goldsmith, “Cross-
networking,” in First Annual IEEE Communications So- layer design for lifetime maximization in interference-
ciety Conference on Sensor and Ad Hoc Communications limited wireless sensor networks,” IEEE Transactions on
and Networks, 2004. IEEE SECON, Oct. 2004, pp. 16–24. Wireless Communications, vol. 5, no. 11, pp. 3142–3152,
72. H. Li and J. Tan, “An ultra-low-power medium access con-
Nov. 2006.
trol protocol for body sensor network,” in 27th Annual In- 88. T. Melodia, M. Vuran, and D. Pompil, “The state of the
ternational Conference of the Engineering in Medicine and art in cross-layer design for wireless sensor networks,” in
Biology Society, 2005. IEEE-EMBS, Shanghai,, 2005, pp. EuroNGI Workshop on Wireless and Mobility, ser. LNCS
2451–2454. 3883, July 2005, pp. 78–92.
73. I. E. Lamprinos, A. Prentza, E. Sakka, and D. Koutsouris, 89. E. De Poorter, B. Latré, I. Moerman, and P. Demeester,
“Energy-efficient MAC protocol for patient personal area Sensor and Ad-Hoc Networks: Theoretical and Algorith-
networks,” in 27th Annual International Conference of the mic Aspects, ser. Lecture Notes Electrical Engineering.
Engineering in Medicine and Biology Society, 2005. IEEE- Springer, June 2008, vol. 7, ch. Universal Framework for
EMBS, Shanghai,, 2005, pp. 3799–3802. Sensor Networks.
74. O. C. Omeni, O. Eljamaly, and A. J. Burdett, “Energy effi- 90. B. Latré, E. De Poorter, I.Moerman, and P. Demeester,
cient medium access protocol for wireless medical body area “Mofban: a lightweight framework for body area networks,”
sensor networks,” in Medical Devices and Biosensors, 2007. Lecture Notes in Computer Science, proceedings of Embed-
ISSS-MDBS 2007. 4th IEEE/EMBS International Summer ded and Ubiquitous Computing (EUC 2007), vol. 4808, pp.
School and Symposium on, Cambridge, UK,, Aug. 2007, pp. 610–622, December 2007.
29–32. 91. D. Chen and P. K. Varshney, “Qos support in wireless sen-
75. H. Li and J. Tan, “Heartbeat driven medium access control sor networks: A survey,” in Int. Conference on Wireless
for body sensor networks,” in HealthNet ’07: Proceedings Networks (ICWN 2004). CSREA Press, June 2004.
of the 1st ACM SIGMOBILE international workshop on 92. B. Braem, B. Latré, C. Blondia, I. Moerman, and P. De-
Systems and networking support for healthcare and assisted meester, “Improving reliability in multi-hop body sensor
living environments. Puerto Rico, USA: ACM, 11 June networks,” in Second International Conference on Sensor
2007, pp. 25–30. Technologies and Applications ( SENSORCOMM 2008),
76. N. Golmie, D. Cypher, and O. Rebala, “Performance anal- Cap Esterel, France, 25-31 August 2008, pp. 342–347.
ysis of low rate wireless technologies for medical appli- 93. G. Zhou, J. Lu, C.-Y. Wan, M. Yarvis, and J. Stankovic,
cations,” Computer Communications, vol. 28, no. 10, pp. “Bodyqos: Adaptive and radio-agnostic qos for body sensor
1266–1275, June 2005. networks,” April 2008, pp. 565–573.
18