=Paper=
{{Paper
|id=Vol-1852/p13
|storemode=property
|title=A security framework for wireless body area
network based smart healthcare system
|pdfUrl=https://ceur-ws.org/Vol-1852/p13.pdf
|volume=Vol-1852
|authors=Mobeen Khan,Muhammad Taha Jilani,Muhammad Khalid Khan,Maaz Bin Ahmed
}}
==A security framework for wireless body area
network based smart healthcare system==
https://ceur-ws.org/Vol-1852/p13.pdf
A security framework for wireless body area
network based smart healthcare system
Mobeen Khan, Muhammad Taha Jilani1, Muhammad Khalid Khan, Maaz Bin Ahmed
Graduate School of Science & Engineering, PAF- Karachi Institute of Economics and Technology,
Karachi, Pakistan
1
e-mail: mtaha.jilani@gmail.com
Abstract—In recent years, there is enormous growth of are small and limited resources, such as power and processing
smart systems that have been developed for range of capabilities [5]. Therefore, an efficient security mechanism for
applications. This development is focused on Internet-of- these resource constrained devices is currently an important
Things (IoT), which will eventually, transform IoT into system
research area.
of systems. Such smart systems are widely utilized in healthcare;
however, the wide scope of such systems is also vulnerable to These days, multiple wireless networks are available around
security and privacy issues. This paper proposes a security us, these wireless devices are interconnected over a traditional
framework for wireless body area network (WBAN) based network like internet which can be used to connect the remote-
smart healthcare system. The proposed framework implies the users in a fast, reliable and cost effective manner. Nevertheless,
security mechanism by considering low power and low
the connected devices over conventional network then there will
resources devices within the WBAN. This is not just reduces
be need for a security mechanism that can restrict attacker
the complexity for resource constrained devices, but
ensures the availability and data integrity in simple, yet effectiveaccess and control. However, conventional network has limited
manner. protection that is not suitable for an IoT network [5]. For
instance, an endeavor can access and manipulate the IoT devices
Keywords—Wireless body area network (WBAN) security; which may have an anonymous result. Therefore, there is a need
Internet-of-Things; smart healthcare; e-Health systems;
for security framework that can be implemented to protect the
ZigBee security framework; IoT architecture
IoT system that cannot be accessed from inside and outside of
network, except authorized entities.
I. INTRODUCTION
The wireless body area networks (WBANs), are one of the
The recent advancements in the information and
low power sensor network [6], which provides efficient and
communication technologies enable the researchers and
reliable infrastructure for healthcare system including
engineers to realize the smart systems for various applications.
implanted, non-implanted and wearable sensor devices for
The realization of such systems will lead to develop the system
human body. These sensor devices are used to capture various
of system, or also known as Internet-of-Things (IoT). The IoT is
symptoms of a patient like heartbeat, body temperature, blood
defined as various tiny physical objects, that are capable to
pressure, respiration and ECG etc. and send these symptoms to a
sense from their surroundings, connected with each other by
Body Network Controller (BNC). BNC is an essential part of a
different networks and share their collected data to convey
WBAN which is capable to capture sensor data and after
information [1]. These IoT based systems can be employed in
processing forwards to the centralized e-Health server. The e-
industrial applications such as remote-sensing, manufacturing,
Health server in turn saves this real time data from various
transportation, smart-homes, smart-communities and healthcare.
patients that can be monitored by his clinician.
With the help of these small interconnected sensors or objects
the automation of certain system can be achieved [2]. An As shown in Figure 1, a BNC collects and processes data
important application is the smart healthcare system, where the from the devices and forward it to a server using a wireless
doctors can monitor the symptoms of a patient remotely by access network. The Wireless access network which is used in
Wireless Body Area Network in IoT environment [3]. However, this scenario is IEEE 802.11ah. This standard is especially
for an IoT environment one of the most important aspect is the designed for IoT devices. There can be use of WLAN
security and privacy, which is still questionable in many 802.11a/b/n standards but IEEE 802.11ah has some additional
proposed IoT architectures. The architectures are not well feature (like long range with low power consumption).
defined, as they have not provided the information, like how
data can be restricted and preserve its integrity. In conventional
networks AES encryption technique can be used for information
security [4]. But it cannot be applied to IoT based devices, that
Copyright © 2017 held by the authors 80
At other side e-Health server also use the public key
algorithm for secure communication with remote users (sucah
as, Doctors). Each user has its own private key and they will
communicate with server by using a session key (session key is
exchanged with the help of public key). By implementing this
method confidentiality and security access of BNC and e-Health
server can be achieved. The availability of system can be
ensured by tracking the heartbeat of a patient, whereas the
unavailability activates the emergency.
Figure 1. A typical WBAN in a smart healthcare system within the IoT The remaining paper is organized as, Section II overviews
network with its: things, network, and applications [7]. the different existing communication protocol and their
vulnerabilities in terms of security, while the Section III
The comparison of various IEEE 802.11 standards is presented
represents our proposed framework and at last, Section IV
in Figure 2. In hospital vicinity, numerous 802.11ah WLAN
concludes this study.
access points are configured and installed. The e-Health server
is connected via IEEE 802.3 Ethernet with these access points. II. RELATED WORK
As explained before, an e-Health server that stores real time data
Wireless Personal Area Network (WPAN) falls in low
is connected to internet that enables doctors to access this
power wireless technology category. WPAN is specified in
information from remote-end. Likewise, conventional network
IEEE 802.15 standard and released in 2005. For wireless
here a vulnerability rises since the connection between BNCs
connectivity IEEE defines PHY and MAC layer specifications
and e-Health server may have insider threats and similarly, it
for fixed, portable, and moving devices within a POS (personal
may face outside attacks. Therefore, it is essential to implement
operating space)[10]. POS is typically extends up to 10 m in all
a mechanism that ensure the security for a smart Hospital [3, 8].
directions of an object, whether it is stationary or in motion. As
shown in the Figure 3 there are different technologies that are
defined within WPAN excluding IEEE 802.11ah:
IEEE 802.11ah
IEEE 802.15.1 is Bluetooth
IEEE 802.15.4 ZigBee
IEEE 802.15.6 WBAN
Figure 2. Comparison of variants of IEEE 802.11 WiFi standard
This research study focuses the three major goals, these are:
(i) Access security of BNC and e-Health server (ii) Data
confidentiality, and (iii) Availability of system. As mentioned
earlier, traditional networks utilizes the state of the art
encryption algorithm, such as 3DES, AES, IDEA etc. However,
these algorithms cannot be implemented in small devices with
limited resources (in term of power consumption and processing
capabilities). This work implements the AES with CTR
(Counter) mode in BNC for encryption [8]. The AES CTR keys Figure 3. Various technologies defined for wireless personal or body
area networks (WPAN/WBAN)
are generated at server side and then sent to BNC via a shared
(private) key. The generated key is shared and then configured
at BNC while patient register themselves. The encryption that is The IEEE 802.11ah standard for the WLAN is developed in
performed by XOR operation with CTR keys, will exchange 2014, that operates in 900 MHz in contrast with other wireless
periodically while BNC is said to be in ideal mode. The logical LAN standards [11]. The main aim behind the development of
operation of XOR, can be implemented easily and efficiently this standard was to extend the range of IEEE 802.11ac
even for a small resource devices [9]. Thus, demonstrates that standard, particularly for low power devices [12]. Utilization of
AES-CTR mode may have same security level as other modes.
81
lower frequencies is not only helpful to extend the range but it another full function device called Router, and the major
has also lower consumption of power. responsibility is to extend the network coverage by providing
routing function. Unlike, Coordinator it is unable to establish a
One of cheap technology that is also useful for a short
network by itself. The end devices, that are reduced function
distance is the Bluetooth. It is widely used, in various devices
devices are neither similar to coordinator nor router, but they
such as mouse, keyboard, headset and most popular for personal
just sent their sensing data to network using Coordinator and
devices communication, for both data and voice[13]. The typical
Routers. Again, they are unable to establish ZigBee network and
range is about 10m however it can be extend up to 100m (using
cannot serve or even assist in routing.
amplifiers) [13]. The cipher algorithm used by Bluetooth uses
stream cipher named E0, which required re-synchronization for
every payload. The main parts of E0 stream cipher are (i) stream
key-generator (ii) payload key-generator and the (iii)
encrypt/decrypt parts [14]. There is variation in key size, that is
8 bits to 128 bits but it depends on both communicating devices.
In spite of these security mechanisms, it is still vulnerable to
different attacks, including Man in the middle, Blue-snarfing
and Viruses [15].
The ZigBee is an extension of IEEE 802.15.4 WPAN
standard developed by ZigBee Alliance. The protocol stack of
ZigBee is built on the top of IEEE 802.15.4[16], as shown in Figure 5. Topology of a ZigBee network
Figure 4, which only define physical and MAC layer for low The Advance Encryption Standard algorithm is used in
power personal area network[17]. In contrast with IEEE 802.11 ZigBee that is based on AES-128 bit key and it provide security
WLAN it is much simpler protocol and supports multiple features on application and network layers [18]. Three different
topologies. The main characteristic are low power, low kinds of keys are used for this purpose, namely, link key,
throughput, and long battery life with secure networking (128 network key and Master key. The link key is utilized for data
bit AES encryption)[18]. Its typical range is 10 to 100 meter confidentiality between nodes, while, to exchange all
LOS (Line Of Sight) normally depends on output power and information securely, each pair of node has a unique link key
environmental conditions. However data can be transmitted to managed at application layer. The network key (which is 128 bit
longer distances with a mesh network. key) is generated by trust center that is shared among all the
devices. The coordinator may become trust center or even any
dedicated device. At last, the master key has prime function to
secure the process of exchange the link keys between two nodes
and it supposed to be preconfigured before deployment
In order to join ZigBee network each node must request
current network key with the help of preconfigured master key
(to avoid stealing of current network key). Network key is
updated time to time by trust center and shared to all nodes by
using current network key [19]. This whole security scenario
works very well but it is inefficient for WBAN in term of
resources (processing and power consumption). AES is a very
complex encryption algorithm which requires high amount of
Figure 4. Layered architecture of ZigBee processing and battery power (implementation of AES in small
battery powered device) to encrypt the data. As mentioned in
The ZigBee network devices are included Coordinator,
Abstract that it is infeasible to implement the conventional
Router and End devices as shown in Figure 5. The core
security mechanisms in very small resource constraint devices.
component of a ZigBee network is based on Coordinator, which
installed initially and work as full-function device which In various implementations of WBAN, AES is not used yet
establish and manages the whole ZigBee network. With a for security to the best of my knowledge. Because end devices
network there will only one Coordinator. Similarly, there will be of WBAN have very limited resources so it is infeasible to
82
implement complete AES algorithm even the CTR mode of The thorough understanding of AES-CTR mode [20], help
block cipher. CTR mode is easiest to implement in contrast with us to implement it. As shown in Figure 7, e-Health server will
other modes, but still not efficient for WBAN devices because registered a patient with a unique patient ID (PID), when he is
to operate the CTR mode, still it have to perform whole AES being admitted in hospital. This information will store against
process with certain values called Counter. Therefore, in this that patient along with 128 bit intermediate cipher texts. Using
work, an implementation of AES counter (CTR) mode is this information, the BNC will be configured with PID and the
proposed, which is feasible for resource constraint devices. previously generated cipher texts will now store at BNC. The
III. PROPOSED FRAMEWORK additional protection through user login and password BNC will
restrict any information modification or retrieval by an
In order to define my purposed idea it is important to get adversary that might reside in hospital vicinity. There will be a
deep understanding of AES-CTR mode [20]. All modern block single BNC in WBAN for all sensors connected to a patient. It
cipher operates in one of the five standard modes which are should be highlighted that noticeable that permanent memory of
ECB, CBC, CFB, OFB and CTR. In CTR mode a variable BNC can store up to 128 bit intermediate cipher texts. These
called counter is initialized to IV (Initial Vector “some specified texts will be used by BNC for data encryption thru XOR
value”) which is incremented linearly or randomly (with the function with plain text data block. At this moment the whole
help of some pseudo random sequence). This value is wrap BAN (comprising BNC and all sensors) will be ready for
around to initial value upon reaching to its maximum allowable deployment.
limit. The size of the counter depends upon variant of
encryption algorithm being used, like for AES-128 the size of
counter is 128 bit.
Figure 6. Basic building-block for counter mode encryption and
decryption Figure 7. Flow-chart for the proposed security mechanism of a WBAN
network
The Figure 6 illustrates the counter mode, where the key
that execute the AES process, must be shared with receiver. The symptoms of a patient will be sent by body sensors to
Initially, by using shared key it will encrypt the counter value the BNC, which in turn, aggregate the data and forwarded to the
with AES-128 encryption and produced the intermediate cipher e-Health server. The BNC that is connected to e-Health server
text (this term will be used in remaining part of this article). For through IEEE 802.11ah access, can received data from various
real time applications, if gateway router encrypts use other sensors that may capture different symptoms. Although, there
modes, such as CBC, so due to longer delays that will be not will be range of sensors, but most common are the body
suitable. Additionally, the intermediate cipher texts can be temperature, heartbeat, respiration, blood pH and pressure etc.
computed prior to operation, or even along this process, since it The Figure 8 depicts various characteristics of a patient’s
depend upon counter values that can be determine earlier. The symptoms.
simplicity of XOR operation, provide fast logical operations that
increase the efficiency of a hardware while comparing with
other complex arithmetic operation [9]. Even, the proposed CTR
mode uses XOR operation for real-time encryption, but it is able
to achieved the security of well-established AES [20].
83
128 bit Encryption
0 10 114 127
PID PAYLOAD TIMESTAMP
PDU format
(a)
128 bit Encryption & Authentication
0 8 18 121 135
Encrypt
PID PAYLOAD TIMESTAMP
ID
PDU format
(b)
Figure 10. PDU Format (a) without ID Encryption (b) with
Figure 8. Various characteristics of a typical patient [21]
Encrypt ID [21]
As mentioned earlier, the implementation of proposed
security framework will provide confidentiality and data It can be seen that for selection of particular intermediate
integrity to the communication between BNC and e-Health cipher text (to decrypt the PDU) the receiver will use this field.
server. For encryption of data BNC will select an intermediate Once it will decrypted, PID at the start of payload will be
cipher text from its memory and will perform XOR operation checked by receiver and validate with stored PID for a
with data as shown in Figure 9. corresponding BNC. This decryption will depend on successful
validation otherwise it will be rejected. The receiver will be
avoided from replay attack by enabling the timestamp within a
PDU. Whereas, based on predefined time limit the received
PDU is also validated on timestamp. It is worth to mention that
before sending a packet to BNC, the server will carry-out the
same procedure. To ensure availability, continuous monitoring
for the arrival rate of packets of certain BNC can performed. If
it does not receive packets of certain BNC within predefined
limit, it will declare the emergency. The emergency indicates
failure of one or more component within the network, which
may include LAN switches, Access Points and BNC etc.
Another important aspect is determining the failure of the
sensors, it can be achieved by monitoring the status in each
Figure 9. Data encryption through XOR operation arriving payload. If the system do not receive any status in the
To provide confidentiality whole 128 bit block (10 bits for payload in predefined limit (as defined in Figure 8) it can
PID and remaining bits are symptoms) is encrypted with one of assumed that the sensor is failed. Because when a sensor is
selected Intermediate cipher text from memory and its reference failed it will not send data to BNC, and in turn, BNC will send
is also attached with payload. Now data block is ready to be all zero’s bits in its associated payload slot in the subsequent
transmitted towards server. packets.
At this point it is briefly defined how data will be encrypted. A IV. CONCLUSION
field named encryption ID will be attached with each PDU
In this paper, different communication technologies for a
(Protocol Data Unit) to indicate particular intermediate cipher smart hospital using IoT system are extensively reviewed, while
text which will be used to encrypt this PDU as shown in Figure the vulnerabilities for security and privacy are highlighted. The
10. paper proposed a security framework for smart healthcare
system based on wireless body area network (WBAN). It is
found to be simple, more effective and low-cost implementation
for the resource constrained devices, particularly in smart
healthcare environment. The proposed work utilizes the well-
established AES encryption algorithm in a decentralized
manner, even for the resource-constraint devices. This
84
implementation not only fulfills the limitations of the resource implementation," in Proceedings of the conference on
constraint devices which includes less processing capability, Design, automation and test in Europe-Volume 1, p.
10246, 2004.
limited battery, limited memory and hardware simplicity but [10] K. S. Sze-Toh, K. C. Yow, "Usage of mobile agent in
also provides a well-defined security mechanism which is the configuring WPANs," in Control, Automation, Robotics and
prime requirement of today’s networks. The proposed Vision, 2002. ICARCV 2002. 7th International Conference
on, pp. 938-943, 2002.
framework not just reduces the complexity for resource
[11] S. Aust, R. V. Prasad, I. G. Niemegeers, "IEEE 802.11 ah:
constrained devices, but ensures the availability and data Advantages in standards and further challenges for sub 1
integrity in simple, yet effective manner. GHz Wi-Fi," in 2012 IEEE International Conference on
Communications (ICC), pp. 6885-6889, 2012.
[12] W. Sun, M. Choi, S. Choi, "IEEE 802.11 ah: A long
range 802.11 WLAN at sub 1 GHz," Journal of ICT
REFERENCES Standardization, vol. 1, pp. 83-108, 2013.
[13] P. McDermott-Wells, "What is bluetooth?," IEEE potentials,
[1] L. Atzori, A. Iera, G. Morabito, "The internet of things: A
vol. 23, pp. 33-35, 2004.
survey," Computer networks, vol. 54, pp. 2787-2805, 2010.
[14] C. T. Hager, S. F. MidKiff, "An analysis of Bluetooth
[2] Q. Zhu, R. Wang, Q. Chen, Y. Liu, W. Qin, "Iot
security vulnerabilities," in Wireless Communications and
gateway: Bridging wireless sensor networks into internet of
Networking, 2003. WCNC 2003. 2003 IEEE, pp. 1825-1831,
things," in Embedded and Ubiquitous Computing (EUC),
2003.
2010 IEEE/IFIP 8th International Conference on, pp.
[15] T. Panse, V. Kapoor, "A review on security mechanism of
347-352, 2010.
Bluetooth communication," International Journal of
[3] H. Fotouhi, A. Cauevic, K. Lundqvist, "Communication and
Computer Science and Information Technologies, vol. 3, pp.
Security in Health Monitoring Systems--A Review," in
3419-3422, 2012.
Computer Software and Applications Conference
[16] J.-S. Lee, Y.-W. Su, C.-C. Shen, "A comparative study of
(COMPSAC), 2016 IEEE 40th Annual, pp. 545-554, 2016.
wireless protocols: Bluetooth, UWB, ZigBee, and Wi-Fi," in
[4] P. Hamalainen, T. Alho, M. Hannikainen, T. D.
Industrial Electronics Society, 2007. IECON 2007. 33rd
Hamalainen, "Design and implementation of low-area and
Annual Conference of the IEEE, pp. 46-51, 2007.
low-power AES encryption hardware core," in 9th
[17] G. Mulligan, "The 6LoWPAN architecture," in Proceedings
EUROMICRO Conference on Digital System Design
of the 4th workshop on Embedded networked sensors, pp.
(DSD'06), pp. 577-583, 2006.
78-82, 2007.
[5] D. Boyle, T. Newe, "Security protocols for use with
[18] P. Baronti, P. Pillai, V. W. Chook, S. Chessa, A. Gotta, Y.
wireless sensor networks: A survey of security architectures,"
F. Hu, "Wireless sensor networks: A survey on the state of the
in Wireless and Mobile Communications, 2007. ICWMC'07.
art and the 802.15. 4 and ZigBee standards," Computer
Third International Conference on, pp. 54-54, 2007.
communications, vol. 30, pp. 1655-1695, 2007.
[6] E. Jovanov, A. Milenkovic, C. Otto, P. De Groen, B.
[19] N. Vidgren, K. Haataja, J. L. Patino-Andres, J. J. Ramirez-
Johnson, S. Warren, G. Taibi, "A WBAN system for
Sanchis, P. Toivanen, "Security threats in ZigBee-
ambulatory monitoring of physical activity and health status:
enabled systems: vulnerability evaluation, practical
applications and challenges," in 2005 IEEE Engineering in
experiments, countermeasures, and lessons learned," in
Medicine and Biology 27th Annual Conference, pp.
System Sciences (HICSS), 2013 46th Hawaii International
3810-3813, 2006.
Conference on, pp. 5132-5138, 2013.
[7] M. Ghamari, B. Janko, R. S. Sherratt, W. Harwin, R.
[20] H. Lipmaa, P. Rogaway, D. Wagner, "CTR-mode
Piechockic, and C. Soltanpur, "A survey on wireless body
encryption," in First NIST Workshop on Modes of Operation,
area networks for ehealthcare systems in residential
2000.
environments," Sensors, vol. 16, p. 831, 2016.
[21] J. Y. Khan, M. R. Yuce, "Wireless body area network
[8] S. Saleem, S. Ullah, H. S. Yoo, "On the Security Issues in
(WBAN) for medical applications," New Developments in
Wireless Body Area Networks," JDCTA, vol. 3, pp. 178-184,
Biomedical Engineering. INTECH, 2010.
2009.
[9] K. Tiri, I. Verbauwhede, "A logic level design
methodology for a secure DPA resistant ASIC or FPGA
85