Energy-efficient adaptive transmission
power control for wireless body area
networks
ISSN 1751-8628
Received on 10th December 2014
Revised on 8th October 2015
Accepted on 10th November 2015
doi: 10.1049/iet-com.2015.0368
www.ietdl.org
Ali Hassan Sodhro
1,2
,YeLi
1,2
✉
, Madad Ali Shah
2
1
Key Laboratory for Health Informatics of Chinese Academy of Sciences, Shenzhen Institutes of Advanced Technology,
Chinese Academy of Sciences, Shenzhen, People’s Republic of China
2
Electrical Engineering Department, Sukkur Institutes of Business Administration, Sukkur, Pakistan
✉ E-mail: ye.li@siat.ac.cn
Abstract: An important constraint in wireless body area network (WBAN) is to maximise the energy-efficiency of wearable
devices due to their limited size and light weight. Two experimental scenarios; ‘right wrist to right hip’ and ‘chest to right
hip’ with body posture of walking are considered. It is analyzed through extensi ve real-time data sets that due to large
temporal variations in the wireless channel, a co nstant transmission power and a typical conventional transmission
power control (TPC) methods are not suitable choices for WBAN. To overcome these problems a novel energy-efficient
adaptive power control (APC) algorithm is proposed that adaptively adjusts transmission power (TP) level based on the
feedback from base station. The main advantages of the proposed algorithm are saving more energy with acceptable
packet loss ratio (PLR) and lower complexity in impl ementation of desired tradeoff between energy savings and link
reliability. We adapt, optimise and theoretically analyse the required parameters to enhance the system performance.
The proposed algorithm sequentially achieves significant hi gher energy savings of 40.9%, which is demonstrated by
Monte Carlo simulations in MATLAB. However, the only limitation of proposed algorithm is a slightly higher PLR in
comparison to conventional TPC such as Gao’s and Xiao’s methods.
1 Introduction
The combination of wireless communication technologies and
miniature size wearable sensor devices play an important role in
pervasive healthcare system. One of the main contributors in
pervasive healthcare systems is wireless body area network
(WBAN) [1]. The purpose of pervasive healthcare system is to
provide cost-effective health services to anyone, anywhere and
anytime. From user’s perspective low-power, highly reliable and
energy-efficient communication is an important ingredient in
pervasive healthcare systems. In the present era of the energy
crises energy-efficiency is an important factor and cornerstone for
several medical applications in WBAN.
Therefore, an energy-efficient transmission with adaptive power
control (APC) is considered as a key solution that can facilitate
long-term and low-power operations. The WBAN as a simple and
cost-effective health monitoring technology has attracted extensive
attention in inter-disciplinary areas. The variable data rates from
10 kbps to 10 Mbps are used depending on the nature of the
targeted application. The WBAN contains a base station (BS) and
numerous on-body sensor nodes that accumulate vital-sign signals
such as body temperature, blood pressure, electrocardiogram
(ECG) and electroencephalogram (EEG) etc. from the human
body. The accumulated data are sent to electronics health
(eHealth) centres and medical hospitals via the BS. Afterwards,
on-demand facilities are provided to patients, as per their required
needs [2].
The MicaZ nodes with CC2420 radio are extensively used in
WBAN that operates in the 2.4 GHz frequency band, and can
support a 250 kbps data rate. Moreover, these nodes present 32
transmission power (TP) levels (ranging from −25 to 0 dBm
output) selectable at run-time by configuring a register; the output
power in decibel milliwatt (dBm) and respective energy
consumption rate in milliwatt (mW) are outlined in Table 1 for
multiple TP levels [3]. The energy consumption values show large
gaps between the TP levels and these TP levels have tradeoff with
energy-efficiency and link reliability. Therefore, a suitable TP
should be assured according to WBAN requirements. Hence,
much attention needs to be paid at selecting the desired TP level
in on-body sensor nodes for energy saving and link reliability.
Most sensor nodes in WBAN have limited battery lifetime, so
energy-efficiency is considered as a direct need. A variety of
non-medical and medical services may be supported with the help
of WBAN, indeed high reliability is not required for non-medical
services, but medical services must assure high reliability [4, 5].
Therefore, the energy-efficiency and high reliability both will be
supported by WBAN to monitor and manage the patients’ life
efficiently and economically.
TP control (TPC) remained a main area of research in wireless
sensor networks (WSNs). TPC research may be subdivided into
four categories, i.e. a network-level method, a node-level method,
a neighbour-level method and a packet-level method [6]. A single
fixed TP level is used for entire network in network-level method,
whereas the node-level method uses different TP levels for every
sensor node. Different TP levels for different neighbours are used
by each sensor node in neighbour-level method. The discussed
methods are only suitable for invariant channel state, but are
unsuitable to fluctuating channel state. Therefore, the most
appropriate method for unstable channel state is the packet-level
method in which the TP of each transmitter sensor node changes
after receiving feedback from the BS. However, conventional TPC
protocols are not suitable for WBAN because of their static nature.
Therefore, TPC has become an attractive research area in WBAN.
Kim and Eom [4] develop a new TPC algorithm which is based on
short-term and long-term link-state estimations; however, they did
not mention the energy saving. A number of researchers have
proposed TPC protocols for WBAN, but still some problems exist.
Xiao
et al. [7] propose multiplicative increase and additive
decrease TP policy-based TPC algorithm in body area senor
networks, which is complex and gives high reliability, but
consumes more energy. In this algorithm also the TP adjustment
range is confined, and energy consumption could not be
IET Communications
Research Article
IET Commun., 2016, Vol. 10, Iss. 1, pp. 81–90
81
&
The Institution of Engineering and Technology 2016