J Med Syst (2013) 37:9961DOI 10.1007/s10916-013-9961-4

ORIGINAL PAPER

A Secure RFID-based WBAN for Healthcare Applications

Sana Ullah · Atif Alamri

Received: 3 March 2013 / Accepted: 14 July 2013 / Published online: 6 September 2013© Springer Science+Business Media New York 2013

Abstract A Wireless Body Area Network (WBAN) allowsthe seamless integration of small and intelligent invasive ornon-invasive sensor nodes in, on or around a human bodyfor continuous health monitoring. These nodes are expectedto use different power-efficient protocols in order to extendthe WBAN lifetime. This paper highlights the power con-sumption and security issues of WBAN for healthcareapplications. Numerous power saving mechanisms are dis-cussed and a secure RFID-based protocol for WBAN isproposed. The performance of the proposed protocol isanalyzed and compared with that of IEEE 802.15.6-basedCSMA/CA and preamble-based TDMA protocols usingextensive simulations. It is shown that the proposed protocolis power-efficient and protects patients’ data from adver-saries. It is less vulnerable to different attacks comparedto that of IEEE 802.15.6-based CSMA/CA and preamble-based TDMA protocols. For a low traffic load and a singlealkaline battery of capacity 2.6 Ah, the proposed proto-col could extend the WBAN lifetime, when deployed onpatients in hospitals or at homes, to approximately fiveyears.

Keywords IEEE 802.15.6 · MAC · Security · RFID

Introduction

Wireless Body Area Networks (WBANs) are becomingincreasingly important for future health care systems. They

S. Ullah (�) · A. AlamriDepartment of Information System, College of Computerand Information Sciences, King Saud University,Riyadh, Saudi Arabiae-mail: sanajcs@hotmail.com

have enough capability to collect biological informationfrom the users in order to maintain their optimal healthstatus. WBANs are able to detect and possibly predict thedeteriorating conditions of patients and are also able to mon-itor chronic diseases including cardiovascular and asthmadiseases [1]. This kind of unobtrusive health monitoring notonly improves the quality of life but also provides computer-assisted rehabilitation to the patients. WBANs are generallycomprised of in-body and on-body area networks. Thein-body networks allow communication between invasivenodes and the coordinator using Medical Implant Commu-nications Service (MICS) band, while the on-body networksuse unlicensed Industrial, Scientific, and Medical (ISM) andUltra-wideband (UWB) bands for communication betweennon-invasive nodes and the coordinator. Because both inva-sive and non-invasive sensor WBAN nodes are miniaturizedand have limited power capacity, they require novel power-efficient solutions at network, Medium Access Control(MAC), and physical layers. This paper highlights the majorWBAN issues that are not yet addressed by the researchcommunity and discusses the power saving mechanismsfor WBAN. This paper also analyzes the contention- andschedule-based power saving mechanisms for WBAN andproposes a secure RFID-based protocol for low-power com-munication. Extensive simulations are conducted to analyzeperformance of the proposed protocol with that of IEEE802.15.6-based CSMA/CA [2] (based on the contention-based mechanism) and preamble-based TDMA protocols[3] (based on the schedule-based mechanism) in terms ofpower consumption, bandwidth utilization and security. Theproposed protocol extends the WBAN network lifetimefrom months to years. It also supports a secure wakeup pro-cess that prevents adversaries from attacking the network.

The rest of the paper is organized as follows.Section “Related work” presents the related work on con

9961, Page 2 of 9 J Med Syst (2013) 37:9961

tention- and schedule-based MAC protocols for WBAN.Section “Major WBAN issues”, “Power saving mecha-nisms for WBAN” present the major WBAN issues andpower saving mechanisms. Section “RFID-based powersaving mechanism” presents the secure RFID-based proto-col for WBAN and the simulation results. The final sectionconcludes our work.

Related work

MAC protocols play a significant role in maximizing thelifetime of WBAN by controlling the dominant sources ofenergy waste, i.e., collision, idle listening, overhearing, andcontrol packet overhead. A collision occurs when more thanone node transmits data at the same time. Idle listeningoccurs when a node listens to an idle or a free channel.Overhearing occurs when a node receives a packet that isnot destined to it. The control packet overhead is the addi-tion of control information to the payload. Multiple researchwork has focused on developing novel MAC protocolsfor WBANs. This work is mainly divided into contention-and schedule-based protocols. In the contention-based pro-tocols, the nodes contend for the channel to send data;Carrier Sensor Multiple Access with Collision Avoidance(CSMA/CA) protocol is the best example of contention-based protocols where the nodes contend for the channelusing a random backoff period. In the schedule-based orTime Division Multiple Access (TDMA) protocols, thechannel is divided into multiple and fixed or dynamic timeslots which are used for data transmission. The followingsections review the existing contention- and TDMA-basedprotocols for WBANs and identify the need for a secureRFID-based protocol.

Contention-based protocols for WBANs

Many researchers have directly adapted the contention-based or CSMA/CA protocol defined in IEEE 802.15.4 [4]due to the fact that this standard supports low data rateapplications with low-power consumption. The authors of[5] considered the contention-based IEEE 802.15.4 MACfor periodic and asymmetric WBAN traffic, however theyhave not considered the real-WBAN scenario where manynodes may generate aperiodic traffic. Similarly, Li et al.has proved that the performance of unslotted CSMA/CAIEEE 802.15.4 mode is better than that of the beaconmode in terms of throughput and latency [6]. Anotherstudy presented in [7] has discouraged the use of IEEE802.15.4-based CSMA/CA because of unreliable ClearChannel Assessment (CCA) and heavy collision problems.The authors of [8] investigated the contention-based slottedALOHA protocol for WBAN in terms of throughput and

energy consumption. A random contention-based protocolis presented in [9], which enhances quality of service formultiple WBAN users by considering inter-WBAN interfer-ence. Otal et al. proposed a contention-based method thatreduces packet collisions and thereby saves extra energycaused by collisions and idle listening [10]. We investi-gated the performance of contention-based WBANs fordifferent frequency bands and suggested useful directionsfor network provisioning and optimized packet size [11].Authors of [12] and [13] developed a Markov chain modelof contention-based WBANs under saturated and non-saturated conditions and studied the effectiveness of dif-ferent random access periods. The contention access delayfor WBAN is studied in [14] where the contention windowis dynamically adjusted using listening window and delaytolerance.

Schedule-based protocols for WBANs

The schedule-based protocols have also attracted the atten-tion of many WBAN researchers. In [15], the authors pro-posed a novel TDMA protocol that solves overhearing andprotocol overhead problems by exploiting the fixed WBANtopology. In order to prolong the WBAN network lifetime,the authors of [16] proposed a TDMA protocol that uses anout-of-band wakeup channel for low-power consumption.Another TDMA-based protocol is presented in [17] wherethe authors considered directional MAC with multi-beamantennas for enabling simultaneous communication in alldirections. We proposed a dual-channel TDMA-based pro-tocol in [18] and [19] where communication between nodesand the coordinator is based on traffic patterns. Anotherstudy presented in [20] proposed a battery-aware TDMAprotocol that utilizes battery discharge dynamics, wirelesschannel models, and packet queuing characteristics. Anultra low power TDMA MAC protocol is introduced in [21],which reduces idle listening by using a separate wakeupreceiver for wakeup purposes. The authors of [22] proposeda priority-based TDMA protocol by using two separate anddedicated periods for life-critical and consumer electron-ics applications. The TDMA protocols presented in [23]and [24] target periodic WBAN traffic without consideringaperiodic and life-critical traffic. In [25], the authors pro-posed an adaptive TDMA protocol that reduces overhear-ing and idle listening problems by using periodic wakeupschedules.

The above work on contention- and schedule-based pro-tocols for WBANs have several limitations. Most of theseprotocols do not address the security issues in WBANs. Inaddition, they consume significant power due to collisionoverhead and frequent synchronization. The secure RFID-based protocol could solve these problems by providing asecure wakeup process that must be able to prevent adver-

J Med Syst (2013) 37:9961 Page 3 of 9, 9961

saries from attacking the network and be able to extend thenetwork lifetime from months to years.

Major WBAN issues

Although the current literature has explained some of theWBAN issues [26–29], the security and power consumptionissues have received limited attention.

WBAN security issues

Because WBANs are expected to collect important andlife-critical information from the patients, attacking themwould lead the patients to serious consequences [30]. MajorWBAN security issues are: 1) data confidentially andintegrity, which protects the data from disclosure and alter-ation, 2) data authentication, which confirms the originalidentify of the source, 3) data freshness, which confirmsthe freshness of the data and prevents it from replayingusing old keys, 4) availability, which confirms the avail-ability of patients data to the physician, and 5) securemanagement, which distributes key among the nodes andthe coordinator for encryption and decryption operation.Several types of attacks are possible on WBANs at phys-ical, MAC, network, and transport layer [30]. Attacks onthe MAC layer may stop the entire operation of the net-work. For example, adversaries may maximize his access tothe medium by manipulating its backoff, thus keeping thechannel busy all the time [31]. In other words, the backoffmanipulation attacks prevent the legitimate users to uti-lize the bandwidth efficiently. Another possible attack is acollision attack, where the adversaries send dummy pack-ets to cause collisions on the channel. Some smart MACprotocols may prevent the backoff manipulation and the col-lision attacks, however they are still open to replay attacks,where the adversaries penetrates the network by gainingenough knowledge of the protocol operation. For example,the supeframe structure of IEEE 802.15.4 could be attackedby creating interference in the slot allocation process usingthe beacons. Most of these attacks could be prevented byintroducing a secure wakeup method.

WBAN power consumption issues

The dominant sources of power consumption are collision,idle listening, overhearing, and control packet overhead.Existing works in Section “Related work” do not providereliable MAC protocols to control all these sources. Thepower consumption sources must be controlled in orderto extend the battery life of a node from months to yearswithout intervention. The power consumption requirementsmay vary for different applications, depending on the traffic

heterogeneity, Quality of Service (QoS), and network adapt-ability. Because WBANs are expected to move frequently,this may cause additional power consumption due to fastmobility. Another important factor affecting the power con-sumption of WBAN is changes in the network topology andthe body position. For example, the power consumption andthe average throughput are considerably affected when aperson is standing or sitting on a chair [32]. One of the solu-tions to improve WBAN lifetime is to scavenge energy fromthe environment including body heat and body variation.However this method should consider the localized SpecificAbsorption Rate (SAR), which must be minimized sincethe nodes are deployed in close proximity to or inside thehuman body [26]. Another solution is to use a dual channelconcept; a separate channel for sending control and synchro-nization packets and a data channel for sending data packets.This concept is already adapted in [18] and [28], howevermost of these protocols use wakeup radio, which has hard-ware implementation cost that is not affordable by WBANs.The best solution is to attach RFID tags to the WBAN nodesfor triggering the data channel to send or receive data.

Power saving mechanisms for WBAN

This section discusses the contention-based, scheduled-based, hybrid, duty cycling and wakeup radio mechanismsfor WBAN.

In contention-based mechanisms, the nodes compete forthe channel to transmit data. If more the one node try toaccess the channel at the same time, a collision may occur.The slotted-ALOHA and CSMA/CA protocols are based oncontention-based mechanisms. In the CSMA/CA protocol,the nodes contend for the channel using a random back-off counter uniformly distributed over the interval [1, CW]where CW is the contention window. The backoff counter isdecremented until it reaches zero where nodes transmit thedata. If the channel is busy, the backoff counter is suspendeduntil the end of the current transmission. The CW is also

Fig. 1 RFID-based WBAN node with wakeup and data circuitry

9961, Page 4 of 9 J Med Syst (2013) 37:9961

Fig. 2 Resource allocation process of the proposed protocol

doubled for each number of failures, this however dependson the standard. For example, the IEEE 802.15.6-basedCSMA/CA protocol doubles the CW for even number offailures only. Further details about the IEEE 802.15.6-basedCSMA/CA are present in [2] and [11].

In schedule-based or TDMA mechanisms, the channelis divided into a number of fixed or dynamic slots thatare used for data transmission. The nodes turn on theirradios only in their own data slots, thus saving the powerconsumed due to overhearing and idle listening. The prob-lems with TDMA protocols is the additional overheadcaused by synchronization of the superframe and data slotboundaries. Preamble-based TDMA protocol is based onschedule-based mechanisms, where a preamble is used forbroadcasting the destination node information. The nodesreceive the preamble and record the boundaries of TDMAslots for data transmission. Other common TDMA protocolsare HEED [33], FLAMA [34], and LEACH [35].

Fig. 3 Pseudo code of the proposed protocol

Fig. 4 Pseudo code of the preamble-based TDMA protocol

The hybrid mechanisms combine the benefits ofcontention- and schedule-based mechanisms. They adaptcontention-based mechanisms for a low traffic load andswitch to schedule-based mechanisms for a high trafficload. The IEEE 802.15.4 MAC protocol adapts the hybridmechanisms by combing the CSMA/CA and TDMA pro-tocols in one superframe structure [4]. It uses the slottedCSMA/CA for normal traffic and the TDMA for real-timeand bursty traffic. Other standards such as IEEE 802.15.3[36] and IEEE 802.15.6 standards have also adapted thehybrid approach in their MAC protocols.

Initialization: CWmin, CWmax, Count 1: for each n do 2: Count ++ 3: Select x random (0, CWmin)4: if channel is idle 5: x - - 6: if (n succeeded) 7: Transmit data 8: else 9: if (n failed) 10 if (count is even) 11: Double CW 12: else 13: CW is unchanged 14: end if 15: end if 16: end if 17: else18: freeze x until channel is idle 19: end if 20: end for

Fig. 5 Pseudo code of the IEEE 802.15.6-based CSMA/CA protocol

J Med Syst (2013) 37:9961 Page 5 of 9, 9961

Table 1 Simulationparameters CCA 63/Symbol rate CW 16

pSIFS 50µs Wakeup/Beacon packet 80 bits

CSMA slot length CCA + 20µs MAC header 56 bits

MACFooter 16 bits Propagation delay 1µs

Idle power 5µW Transmitting power 27mW

Receiving power 1.8mW Switching time 1.2ms

The duty cycling mechanisms allow the nodes to alterbetween sleep and wakeup periods. The nodes switch offtheir radios when there is no activity on the network, thiscan save significant amount of energy consumed due to idlelistening. Generally, the nodes adapt a periodic or an adap-tive wakeup period. In both cases, the nodes wake up for ashorter duration to check the channels status and to receivesynchronization information. The SMAC [37] and TMAC[38] are examples of duty cycling mechanisms.

The wakeup radio mechanisms are categorized into one-channel and dual-channel wakeup radio mechanisms. In theone-channel wakeup radio mechanism, the wakeup processand the data transmission are done on the same channel.The source node first sends a wakeup packet and waits for areply. As soon as this wakeup packet hits the receivers activeperiod, the request is acknowledged. Otherwise the sourcenode continuously transmits the wakeup packet until it isacknowledged. The one-channel wakeup radio mechanismis good for application with flexible delay requirements.In the dual-channel wakeup radio mechanism, the wakeupprocess and the data transmission are done on a separatecontrol channel and data channel, respectively. Because thewakeup radio packets are sent on a control channel, thetarget node immediately wakes up within tolerable delay.The wakeup radio channel has considerably low power con-sumption compared to that of a data channel, however itcomes with an additional cost of hardware implementation.

RFID-based power saving mechanism

RFID systems have already played a significant role in awide range of applications including manufacturing, objecttracking, inventory control, smart environments, and health-care systems. One of the main reasons of RFID success isits cheap and quick implementation without an additionalcost. In a typical RFID systems, the RFID tags are attachedto the objects for different applications. These tags storeinformation that is further read by the RFID reader. TheRFID tags are categorized into active, passive, and semi-passive tags. The active tags have enough power sourcesand have the ability to initiate communication. The pas-sive tags can only receive data and can be powered up bythe reader. The semi-passive tags have limited power sourceonly for internal processing. Because these tags are inex-pensive and power-efficient, they can be easily integratedinto WBAN. They can be attached to the WBAN nodesand can be used for the wakeup purpose only (and notfor storing information as done in traditional RFID sys-tems). Our RFID-based protocol considers a secure RFIDwakeup method that triggers the data channel for com-munication. The proposed protocol uses a separate controlchannel for wakeup and synchronization packets and a datachannel for original data transmission as given in Fig. 1.During the wakeup process, the nodes send an RFID secu-rity code (the security codes are randomly distributed by

Fig. 6 Power consumption fordifferent packet generationperiod

9961, Page 6 of 9 J Med Syst (2013) 37:9961

Fig. 7 WBAN network lifetimefor an average packet generationperiod of 50 seconds

the coordinator) in the RFID wakeup packet, which is com-pared by the coordinator with the code already stored inits circuitry. When the two RFID security codes are same,the data channel is triggered for transmission. However,when the codes are different, the RFID wakeup circuitryaborts the wakeup process. The secure RFID wakeup pro-cess prevents adversaries from penetrating into network andfrom using different attacking methods including backoffmanipulation, collision, and reply attacks. One of the rea-sons is that the adversaries are unable to grab the beaconand the resource allocation information until they followthe secure wakeup process. Once the wakeup process issuccessfully done, the coordinator sends beacon on thedata channel to the nodes for synchronization as given inFig. 2. The beacon contains information about the super-frame structure and data slot boundaries (the beacon anddata frame formats are not included here to space limitationproblems).

We conduct a preliminary study on analyzing the perfor-mance of the proposed protocol with that of IEEE 802.15.6-based CSMA/CA and preamble-based TDMA protocol interms of power consumption, bandwidth utilization andsecurity. We develop a discrete event custom simulatorin C++ that implements the basic operation of all theprotocols including the secure RFID-based protocol. Thepseudo codes of the secure RFID-based, preamble-basedTDMA and IEEE 802.15.6-based CSMA/CA protocols aregiven in Figs. 3, 4, and 5, where x represents the dataslots in the superframe S(t). Because we are interestedin MAC layer performance, the physical layer parame-ters are not considered in the simulations. We considera single WBAN star topology where the communica-tion flow is in upward direction towards the coordinator.The nodes are triggered to generate Poisson traffic. Forthe CSMA/CA, the minimum and maximum CWs aretaken the same for all nodes. The CW is doubled for

Fig. 8 Bandwidth utilization ofthe protocols

J Med Syst (2013) 37:9961 Page 7 of 9, 9961

Fig. 9 Decrease in bandwidthutilization using random andweak attackers

even number of failures and is unchanged for odd num-ber of failures. The simulation parameters are listed inTable 1.

The comparison of the average power consumption ofIEEE 802.15.6-based CSMA/CA, preamble-based TDMA,and secure RFID-based protocols is presented in Fig. 6.It can be seen that the proposed protocol is effectivein reducing power consumption of the nodes by adjust-ing their wakeup and sleep schedules. The overall powerconsumption decreases as a function of packet generationor packet inter-arrival period, this result is obvious sincelarger packet generation period allows nodes to remainin sleep mood for enough duration. Figure 7 shows theWBAN lifetime for a packet generation period of 50 sec-onds. The figure considers an alkaline battery of capac-ity 2.6 Ah. The average lifetime of a node is aroundthree months for IEEE 802.15.6-based CSMA/CA, oneyear for preamble-based TDMA, and five years for thesecure RFID -based protocol. These results are valid fora low traffic load with an error free channel. This fig-ure also shows the average lifetime of an Electroncar-diogram (ECG) node when deployed into a network of20 nodes including Oxygen Saturation (SpO2), respira-tion, and temperature nodes. The average packet generationperiod of ECG, SpO2, respiration, and temperature nodesare around 44 seconds. It can be seen that the proposedprotocol extends the average lifetime of the ECG nodeto approximately four years. Figure 8 shows the band-width efficiency for all protocols, where the data andcontrol represent the time spent in transmitting data andcontrol packets, respectively. The bandwidth utilization isalmost the same for all protocols, however the overheadof control packets is different, for example, control pack-ets transmitted by the preamble-based TDMA are over-heard by all other nodes. In the proposed protocol, the

overhead of control packets is almost negligible because itsends RFID wakeup signal whenever required, thus reduc-ing overhearing and idle listening problems. In order toanalyze different attacks on all the protocols, we considertwo types of attackers: a random attacker, which has smartattacking capabilities and can use backoff manipulation,collision, and relay attacks, simultaneously, and a weakattacker, which has limited capabilities to attack the net-work. As illustrated in Fig. 9, random attackers decreasethe bandwidth utilization of preamble-based TDMA, IEEE802.15.6-based CSMA/CA, and secure RFID-based proto-cols by 60 %, 70 %, and 20 %, respectively. The IEEE802.15.6-based CSMA/CA is mostly affected by backoffmanipulation and collisions attacks where the attackers triedto select a small CW in order to keep the channel busyall the time. The proposed protocol, however, is less vul-nerable to attacks due to its strong and secured wakeupprocess.

Conclusion

This paper first reviewed the major issues and power sav-ing mechanisms in the context of WBAN. It then pro-posed a secure RFID-based protocol that considered RFIDfor wakeup purpose only. The preliminary study analyzedand compared performance of the proposed protocol withthat of IEEE 802.15.6-based CSMA/CA and preamble-based TDMA protocols under different parameters includ-ing power consumption and security. The simulation resultsshowed that the proposed protocol is power-efficient and isless vulnerable to attacks compared to the other protocols.In future, we will extend the secure RFID-based proto-cols for heterogeneous WBAN applications including heartactivity monitoring.

9961, Page 8 of 9 J Med Syst (2013) 37:9961

Acknowledgments The authors extend their appreciation to theDeanship of Scientific Research at King Saud University for fundingthis work through the research group project no. RGP-VPP-220.

Conflict of interest The authors declare that they have no conflictof interest.

References

1. Ullah, S., Higgins, H., et al., A comprehensive survey of wire-less body area networks - On PHY, MAC, and network layerssolutions. J. Med. Syst. 36(3):1065–1094, 2012.

2. IEEE 802.15.6, Wireless Medium Access Control (MAC)andPhysical Layer (PHY) Specifications for Wireless Personal AreaNetworks (WPANs) used in or around a body, 2012.

3. PB-TDMA website, http://www.isi.edu/nsnam/ns/doc/node176.html Data visited: 21 Feb. 2013

4. IEEE Std.802.15.4, Wireless medium access control (MAC) andphysical layer (PHY) specifications for low data rate wirelesspersonal area networks (WPAN). IEEE: Piscataway, NJ, USA2006

5. Timmons, N. F., and Scanlon, W. G., Analysis of the performanceof IEEE 802.15.4 for medical sensor body area networking. In:The Proceedings of IEEE SECON. pp. 47. Santa Clara, CA, USA,2004.

6. Li, C., Li, H. B., Kohno, R., Performance evaluation of IEEE802.15.4 for wireless body area network. In: Proceedings of IEEEInternational Conference on ICC Workshops, pp. 15. Dresden,Germany, 2009.

7. Zhen, B., Li, H. B., R. Kohno., IEEE body area networks and med-ical implant communications. In: The Proceeding of the ICST 3rdInternational Conference on Body Area Networks. Tempe, Ariz,USA, 2008.

8. Kynsijarvi, L., Goratti, L., Tesi, R., Iinatti, J., Hamalainen, M.,Design and performance of contention based MAC protocols inWBAN for medical ICT using IR-UWB, 21 international confer-ence on personal. In: Indoor and Mobile Radio CommunicationsWorkshops (PIMRC Workshops), pp. 107111, 26-30 Sept. 2010.

9. Cheng, S., Huang, C. Y., Tu, C. C., RACOON: A multiuser QoSdesign for mobile wireless body area networks. J. Med. Syst.35:1277–1287, 2011.

10. Otal, B., Alonso, L., Verikoukis, C., Highly reliable energy-savingMAC for wireless body sensor networks in healthcare systems.IEEE. J. Sel. Areas Commun. 27(4):553–565, 2009.

11. Ullah, S., Chen, M., Kwak, K. S., Throughput and delay anal-ysis of IEEE 802.15.6-based CSMA/CA protocol. J. Med. Syst36(6):3875–3891, 2012.

12. Rashwand, S., Misic, J. V., Khazaei, H., IEEE 802.15.6 undersaturation: Some problems to be expected. J. Commun. Netw.13(2):142–148, 2011.

13. Rashwand, S., and Jelena, V., Misic: Performance evaluation ofIEEE 802.15.6 under non-saturation condition. IEEE GLOBE-COM 2011:1–6, 2011.

14. Huq, M. A., Dutkiewicz, E., Fang, G., Liu, R. P., Vesilo, R., MEBMAC: Improved channel access scheme for medical emergencytraffic in WBAN. In: International Symposium on Communi-cations and Information Technologies 2012 (ISCIT 2012), pp.371-376, 2012.

15. Marinkovi, E. M. P., Spagnol, C., Faul, S., Marnane, W. P.,Energy-efficient low duty cycle MAC protocol for wireless bodyarea networks. IEEE Trans. Inf. Technol. Biomed. 13(6):915–925,2009.

16. Moshaddique, A. A., Ullah, N., Chowdhury, M. S., Islam, S. M.R., Kyungsup, K., A power efficient MAC protocol for wire-less body area networks. EURASIP J. Wirel. Commun. Netw.2012.

17. Hussain, M. A., Alam, M. N., Kwak, K. S., Directional MACapproach for wireless body area networks. Sensors 11(1):771–784, 2012.

18. Ullah, S., and Kwak, K. S., An ultra low-power and traffic-adaptive medium access control protocol for wireless body areanetwork. J. Med. Syst. 36(3):1021–1030, 2012.

19. Kwak, K. S., and Ullah, S., A traffic-adaptive MAC protocol forWBAN. In: IEEE Globecom, pp 1286–1289, 2010.

20. Su, H., and Zhang, X., Battery-dynamics driven TDMA MACprotocols for wireless body-area monitoring networks in health-care applications. IEEE. J. Sel. Areas Commun. 27(4):424–434,2009.

21. Marinkovic, S. J., and Popovici, E. M., Power Efficient Net-working Using a Novel Wake-up Radio. IEEE PervasiveHealth,2011.

22. Zhang, Y., and Dolmans, G., Priority-guaranteed MAC protocolfor emerging wireless body area networks. Ann. Telecommun.66(3-4):229–241, 2011.

23. Li, C., and Kohno, H. B., Reservation-based dynamic TDMA pro-tocol for medical body area networks. IEICE. Trans. Commun.92(2):387–395, 2009.

24. Li, H. M., and Tan, J. D., Heartbeat driven MAC for body sen-sor networks. In: The Proceedings of the International Workshopon Systems and Networking Support for Healthcare and AssistedLiving Environments, pp. 2530. San Juan, Puerto Rico, 2008.

25. Rahim, A., Javaid, N., Aslam, M., Qasim, U., Khan, Z. A.,Adaptive-reliable medium access control protocol for wirelessbody area networks. In: 9th IEEE Communications Society Con-ference on Sensor, Mesh and Ad Hoc Communications andNetworks (SECON 2012), 2012.

26. Latre, B., Braem, B., Moerman, I., Blondia, C., Demeester, P.,survey on wireless body area networks. Wirel. Netw. 17:(1), 2011.

27. Cheng, S., and Huang, C., RACOON: A multiuser QoS design formobile wireless body area networks. J. Med. Syst., 1277–1287,2011.

28. Marinkovic, S. J., and Popovici, E. M., Power efficient networkingusing a novel wake-up radio. In: The Proceedings of 5th Inter-national Conference on Pervasive Computing Technologies forHealthcare, pp. 139–143, 2011.

29. Chen, M., et. al., Body area networks: A survey. ACM/SpringerMob. Netw. Appl. 16(2): 171–193, 2011.

30. Saleem, S., et al., A study of IEEE 802.15.4 security frameworkfor wireless body area networks. Sensors 11(2):1383–1395, 2011.

31. Radosavac, S., Cardenas, A. A., Baras, J. S., Moustakides, G. V.,Detecting IEEE 802.11 MAC layer misbehavior in ad hoc net-works: Robust strategies against individual and colluding attack-ers. J. Comput. Secur 15:103–128, 2007.

32. Shah, R. C., and Yarvis, M., Characteristics of on-body 802.15.4networks. In: Proceedings of 2nd IEEE Workshop on WirelessMesh Networks. Reston, VA, USA, pp. 138–139, 2006.

33. Younis, O., and Fahmy, S., Heed: A hybrid, energy-efficient, dis-tributed clustering approach for adhoc sensor networks. IEEETrans. Mob. Comp. 3:366–379, 2004.

34. Rajendran, V., Aveces, G. L., Obraczka, K., Energy-efficient,application-aware medium access for sensor networks. In: Pro-ceedings of IEEE Mobile Adhoc and Sensor Systems Conference.Washington, DC, USA, p. 8630, 2005.

35. Heinzelman, W. R., et al., Energy-efficient communication proto-col for wireless microsensor networks. In: Proceedings of the 33rd

J Med Syst (2013) 37:9961 Page 9 of 9, 9961

Annual Hawaii International Conference on System Sciences,Maui Kula, HI, USA, 2000.

36. IEEE 802.15.3 Website, http://www.ieee802.org/15/pub/TG3.html Data visited: 23 Feb. 2013

37. Ye, W., and Estrin, H. J., An energy-efficient MAC protocol forwireless sensor networks. In: Proceedings of Twenty-First Annual

Joint Conference of the IEEE Computer and CommunicationsSocieties. Miami, FL, USA, pp. 1567-1576, 2002.

38. Dam, V. T., and Langendoen, K., An adaptive energy-efficientMAC protocol for WSNs. In: Proceedings of the 1st ACM Con-ference on Embedded Networked Sensor Systems (Sen-Sys). LosAngeles, CA, pp. 171-180, 2003.