T1 Introduction
　　he world’s population is aging， and increased health care expenses are affecting the quality of life of elderly people. Thus， inexpensive health care solutions are urgently needed. Recently， wireless body area networking （WBAN） has been proposed as one such solution. Physiological information， such as electrocardiography （ECG） signals， blood pressure readings， and body posture signals， can be acquired by a WBAN and sent to a remote medical center for analysis and treatment [1]. This has the potential to relieve the pressure on public medical resources and increase convenience for patients. A WBAN could reduce the number of times a patient with a chronic disease needs to visit the hospital.
　　A WBAN also has other applications， such as somatic and virtual reality [2]， sports training and fitness analysis， and military applications [3]， [4].
　　A WBAN comprises a coordinator node and several sensor nodes deployed on the body， within the body， or around the body. These nodes usually form a network with a star topology and single?hop communication. Sensor nodes collect information from the human body and transmit it to the coordinator node for processing. A typical WBAN architecture is shown in Fig. 1.
　　The body?centric， health?related， mobile nature of WBAN means there are some special requirements that need to be considered when designing a WBAN system. First， energy efficiency needs to be reflected in the system design because sensor nodes only have limited?capacity batteries that are often inconvenient to recharge or replace （especially for implanted sensors）. Second， traffic in a WBAN primarily comprises vital signs； therefore， latency and throughput need to be guaranteed. Finally， a WBAN is heterogeneous because there are diverse nodes located in different positions， and these nodes have different QoS requirements. There are also the differences between WBAN users. Therefore， is necessary to design protocols that enable the data rate to be scaled and services to be differentiated for different sensor nodes and WBANs.
　　Common wireless technologies may not be suitable for use in a WBAN. For example， Bluetooth [5] only supports a limited number of terminal nodes and single?hop communication. It also consumes a lot of power [6]. Zigbee [7] does not sufficiently guarantee QoS or mobility [6]. The IEEE 802.15.6 Task Group published the WBAN standard in 2012. The purpose of this Task Group was to establish a communication standard optimized for sensors operating on， in or around both human and animal bodies. The standard ensures reliability， QoS， energy efficiency， high data rate， and no interference in a variety of WBAN applications [1]. Many researchers have also investigated WBAN?related topics， including model architecture [8]， PHY and MAC layer design [9]， [10]， adaptive route protocol [11]， and interference mitigation [12]， [13]. 　　Attached to the body， a WBAN moves as the hosts does. In a hospital or nursing home， coexistence of WBANs may cause problems. Adjacent WBANs may use the same channel simultaneously because of limited frequency and dense distribution. Without scheduling or coordination between adjacent WBANs， intra?WBAN packet delivery may suffer. WBANs may also need to compete with other wireless networks for spectrum if both networks use the same spectrum band， e.g.， unlicensed 2.4 GHz ISM. Problems arising from the coexistence of WBANs and other wireless networks can be simplified by regarding the WBAN as a uniformly distributed disturbance point because a WBAN is much smaller than other wireless networks. Some research on the coexistence of WBANs and other wireless networks has been done recently [14]-[16]. Problems arising from coexisting WBANs are more complex due to the fact that the scale of communication of different WBANs is similar and different inter? and intra?WBANs may have heterogeneous characteristics. Such problems hamper the widespread deployment of WBANs.
　　In this paper， we focus on problems arising from the coexistence of multiple WBANs and discuss related interference?mitigation solutions. We also discuss research trends and unresolved issues related to inter?WBAN coexistence.
　　In section 2， we describe inter?WBAN interference problems in detail. In section 3， we introduce and compare inter?WBAN interference mitigation solutions. In section 4， we discuss related open research issues. In section 5， we summarize and conclude the paper.
　　2 Problems Arising from the Coexistence of
　　WBANs
　　Problems with coexisting WBANs occur for three main reasons：
　　1） a large number of WBAN users gathered together. Although WBANs usually only have a short communication range， they may still come close to each other in certain places， such as metro stations， hospitals， nursing homes， or retirement villages. If the WBANs are not scheduled properly， transmission collision may occur.
　　2） limited wireless resources. Most WBANs use the 2.4 GHz ISM band， which is also used by Wi?Fi， Bluetooth， and Zigbee networks （Fig. 2）. The problem is compounded because Wi?Fi APs have been widely deployed both in public places and at home. Therefore， there are insufficient collision?free channel resources to satisfy demand.
　　3） random mobility. In crowded and resource?limited environments， reuse of frequency channels is inevitable. However， the random mobility of WBAN users breaks down the spatial isolation of co?channel WBANs. 　　The distance between WBANs and the distribution intensity of WBANs are both important factors affecting the degree of interference between two WBANs. In [17]， Muhammad et al. conducted experiments on the change in packet loss rate （PLR） in a WBAN in relation to the distance of the WBAN from other interfering WBANs. They found that the distance between WBANs noticeably affects the PLR in each WBAN. In [18]， the relationship between the number of interfering WBANs and packet error rate （PER） was analyzed. The authors found that densely distributed interference sources congest the channel and cause high PER. Thus， closer inter?WBAN distance and denser distribution increases interference between WBANs and degrades communication.
　　Body movements and clothing give rise to a shadowing effect， i.e.， lack of signal reflection， in a WBAN. The antenna angle greatly affects the received signal strength （RSS）， which is given by the received signal?to?interference?plus?noise ratio （SINR） [19]. An unsuitable transmission angle may create serious interference for other WBANs. In addition， the topology of a WBAN affects the degree of interference for other WBANs. For example， WBAN users form a line when queuing in a station or a ring when sitting around a table. In [20]， Wang et al. studied WBANs with a hexagonal lattice structure within an inter?WBAN mesh structure. Geometric probability was then used to model and analyze interference arising from coexistence of WBANs. In [21]， the power?law distribution model of the crowd was introduced to simulate the structure of multiple WBANs， and the distance?based distribution function of interference probability was proposed to estimate the probability of inter?WBAN interference.
　　Apart from the previously mentioned external factors， there are some factors within a WBAN that significantly affect inter?WBAN interference. In [22]， De Silva et al. investigated the relationship between packet delivery rate （PDR）， packet transmission rate， and number of interfering networks. They found that PDR decreases to as low as 65% for a transmission rate of 100 packets/s in a mission?critical WBAN situation. This reduced to 60% when there were eight WBANs in the surrounding environment. In [23]， Sarra et al. investigated transmission power， frequency of data transmission， and packet size in a WBAN when there was both heavy and light interference from the surrounding environment. They found that decreasing the packet size or frequency of data transmission can improve the expected transmission （ETX） and increase PDR. Increasing transmission power also increases the SINR. In [18]， the authors investigated SINR， BER， and probability of collision for TDMA， FDMA and CDMA access schemes when there was inter?piconet coexistence. The authors confirmed that interference from coexisting WBANs significantly degrades the performance of WBANs within the system. The experimental results showed that TDMA and FDMA are better choices than CDMA for mitigating interference when there are coexisting WBANs. 　　3 Solutions for Mitigating Interference
　　Caused by Coexisting WBANs
　　As with intra?WBAN multiple access， inter?WBAN coexistence can be achieved by separating WBANs in the frequency， time， or spatial domains. In the following subsections， we discuss frequency allocation， time scheduling and co?channel interference mitigation. Some other interference?mitigation strategies are also discussed.
　　3.1 Channel Assignment
　　In general， each WBAN should choose a wireless channel for intra?WBAN data transmission. Reasonable channel allocation is the most straightforward way of avoiding inter?WBAN interference. A WBAN is assigned a channel that is different from an adjacent conflicting WBAN. In most scenarios， there is no centralized managing entity； therefore， channels need to be assigned in a distributed way.
　　In [1]， distributed channel?hopping mechanisms were described. Channel hopping is controlled by the hopping sequence， which is generated by the generator polynomial based on the Galois linear feedback shift register （LFSR）. The WBAN coordinator may then change its operating channel periodically according to the Channel Hopping State and Next Channel Hop fields in its beacons without exchanging information. In [18]， [24]， [25]， some other random channel?allocation strategies can be found. These random channel hopping or allocation schemes are easy to implement and are well suited to an environment with a small number of WBANs. However， when WBANs are densely distributed and channel resources are limited， randomized channel?assignment schemes lead to a high probability of channel conflict and significantly degrade performance.
　　Some other channel?assignment solutions use only self?measured interference indicators， such as received signal strength indication （RSSI） and received signal?to?interference ratio （SIR）， which can only be regarded as rough indicators of interfering WBANs. In [26]， the authors propose an interference?aware channel switching （inter?ACS） algorithm that senses whether WBANs are experiencing interference according to the SIR. Depending on the degree of SIR detected by the coordinator， various n?hop channels are assigned to WBANs. This results in better performance than that achieved using uniform or random channel allocation.
　　SIR?based methods are not well suited to crowd scenarios， in which more information about other interfering WBANs needs to be acquired to better allocate channels. In a distributed situation， messages need to be exchanged between WBANs so that WBANs know the interference information of adjacent WBANs. In [27]， Deylami et al. proposed a distributed dynamic coexistence management （DCM） mechanism to improve the performance of coexisting WBANs. In DCM， each WBAN listens the channel and extracts beacon information from other WBANs when it suffers beacon loss. If there is an insufficient gap for coexistence， the WBAN switches to another channel. In [28] and [29]， Movassaghi et al. designed a scheme based on the interference region （IR） for eliminating inter?WBAN interference. In this scheme， each WBAN records the information of interfering nodes in the IR and shares this information with all the other WBANs in its vicinity. A channel?reuse strategy similar to that in cellular networks is then used. With this strategy， distant WBANs can use the same channel， and orthogonal sub?channels are assigned to WBANs that are close to each other and experiencing interference. 　　If more information is exchanged between WBANs， channel assignment can be modeled as a graph coloring problem. Network topology is modeled as a graph G = （V，E）， where the vertex set V represents an individual WBAN， and the edge of set E of G shows that the two connected vertices （WBANs） conflict with each other if allocated the same channel. The colors of the graph show the channel resources. Hence， channel allocation can then be transformed into a graph coloring problem with fastest speed and least colors. In [30]， a Random Incomplete Coloring （RIC） algorithm with low time complexity and high spatial reuse was proposed. In this algorithm， quick inter?WBAN scheduling （IWS） is achieved through random?value coloring， and the reuse efficiency of channel resources is guaranteed by incomplete coloring. In [31]， a combination of cooperative scheduling and graph coloring was proposed to eliminate inter?WBAN interference. The main idea of this scheme is that adjacent WBAN pairs form a cluster that can allocate channels using the same method as that in [28] and further distribute resources between clusters by using the graph coloring method. With graph coloring methods， perfect superframe synchronization is assumed. However， such synchronization cannot be guaranteed because of the randomness of WBANs. Furthermore， interference should also be detected while building the graph model.
　　By passing messages between WBANs， the channel?assignment scheme ensures a higher channel reuse rate. The main problem with this scheme is that effective channel allocation cannot be guaranteed in some scenarios， such as when WBANs are densely distributed or when frequency spectrum is scarce. Furthermore， frequent interference detection and information exchange consumes wireless resources and energy. Therefore， the channel?assignment scheme is only appropriate when there is a relatively large frequency spectrum to be allocated or when there is a relatively small number of coexisting WBANs.
　　3.2 Time Rescheduling
　　The TDMA?based scheme is usually used for intra?WBAN communication to avoid sensor collision. In such a scheme， non?overlapping time slots are allocated to different sensors for transmission. Similarly， inter?WBAN interference can also be mitigated by time rescheduling. When an individual WBAN occupies the channel only part of the time， e.g.， during the channel sensing period or inactive period of superframe in [1]， [5] and [7]， the other WBANs can temporarily access the channel to transmit data. This is a more efficient way of utilizing the channel. 　　In [32]-[37]， the authors use time rescheduling to mitigate inter?WBAN interference. In [32]， the authors propose a BAN?BAN interference mitigation （B2IRS） beacon?enabled strategy in which the coordinator of the WBAN collects information about adjacent WBANs at the end of superframe active time. In this way， multiple WBANs do not access the channel at the same time. In [33]， Kim et al. proposed an asynchronous internetwork interference avoidance scheme called AIIA. This scheme was implemented in the coordinator of the WBAN and maintained a table containing the time scheduling information about adjacent WBANs. Thus， every WBAN could transmit in a conflict?free time slot according to the information table. In [34]， information was exchanged between proximal coordinators of WBANs in order to arrange the transmission sequence of WBANs according to application priority. In [35] and [36]， Mahapatro et al. focused on the fairness between WBANs when there was an insufficient number of time slots. If all WBANs accessed the channel in a round?robin way according to their time slot requirements， WBANs with fewer nodes would have a long wait. The authors then proposed a scheme in which WBAN containing fewer nodes could occupy more time slots and found that average wait times between WBANs remained the same. In time?rescheduling solutions， complete and up to date information about adjacent WBANs is prerequisite. In [37]， Wen et al. used a CSMA?like mechanism to transmit the beacon without collision caused by an incomplete neighbor list.
　　Some other contention?based time rescheduling approaches are found in [38]-[42]. In [38]， [39]， the listening strategy is similar to that in [32]； however the channel is sensed when the beacon needs to be sent. A Poisson distribution model is established to generate the probability of channel sensing. Then， the beacon is sent to reserve the channel once the channel is idle. However， if the channel is busy， there needs to be a wait period before resensing. In [40]-[41]， Dong et al. modified the inter?WBAN time?scheduling scheme to uniformly access the channel when there is no cooperation between WBANs. At the same time， an opportunistic relaying （OR） strategy was introduced to select the minimum interference link inside a WBAN. This enhances coexistence between WBANs. However， this scheme was not practical because links were selected according to channel measurements rather than a prediction scheme， and an increase in the number of relays had the potential to sacrifice system performance. 　　Time scheduling requires complex arrangement strategies to avoid inter?WBAN interference. With scheduling?based methods of inter?WBAN interference mitigation， frequent coordination consumes a lot of energy. With contention?based methods of inter?WBAN interference mitigation， broadcasting information before accessing the channel consumes a lot of energy. In addition， it is also a challenge for a WBAN to obtain transmission information about other WBANs in the vicinity in a timely way. The key issue for effective time scheduling is timely updating of information. Therefore， time?scheduling methods may be more appropriate when the inter?WBAN topology is changing slowly.
　　3.3 CoChannel Interference Mitigation
　　Generally speaking， resource management strategies， such as channel assignment and time rescheduling， are the most straightforward ways of reducing interference between coexisting WBANs. Each WBAN can work in an orderly manner based on the established resource allocation strategy. However， sometimes there may be problems with resource management because WBANs are densely deployed and the inter?WBAN topology changes quickly. In this case， we can only adjust the transmission power to improve the performance of coexisting densely distributed WBANs. Interference between WBANs may still exist， and a proper method should be used to mitigate it.
　　Adjusting the transmission power is a natural way of improving the performance of coexisting densely distributed WBANs. However， although decreasing the transmission power of one WBAN mitigates the interference experienced by that WBAN， it also degrades the performance of that WBAN. Game theory is usually introduced to trade off performance for energy consumption in a cooperative or non?cooperative manner. The importance of performance or energy consumption can be adjusted by setting the fixed or adaptive weighting factor in the application?based utility function. According to Nash Equilibrium （NE）， anyone who changes their strategy alone without consideration of the strategies of other participants will benefit less [43]. That is， each rational participant is not motivated to change their strategy unilaterally if they seek the maximum possible benefit.
　　In [21] and [43]-[46]， the authors describe power?control game for inter?WBAN interference mitigation. In [21]， [43]， [45]， and [46]， a power control game involving a trade?off between throughput and power consumption was proposed to mitigate interference arising from coexisting WBANs. The existence of and uniqueness of the NE solution was discussed. Fast?convergence algorithms such as Harmonic Mean （HM） [43]； Proactive Power Update （PAPU） [46]； and Best Response Iteration [21]， [45] have been proposed to converge to the unique NE. In [44]， Wu et al. created a decentralized inter?user interference suppression （DISG） algorithm for a WBAN. They proposed a power?control game based on the SINR and power use to choose a suitable channel and transmission power. After proving the existence of the NE， the authors derived the condition that would guarantee the uniqueness of the NE and introduced the No?Regret Learning algorithm to optimize the NE. 　　Most of works referring to the power?control game for WABNs only consider the single link inside the WBAN， which is impractical because a WBAN comprises many sensors with various QoS requirements， link gains， etc. The convergence speed of the power?updating algorithm must keep up with any change in the inter?WBAN topology； otherwise， the solution obtained in the current iteration is only optimal for the earlier period， not the current period.
　　In [47]， Kazemi et al. do not use game theory to mitigate interference arising from coexistence of WBANs. Instead， they propose a power controller based on reinforcement learning （RL） that exploits environmental information such as received interference， previous transmission power， and SINR. In [48]， the authors propose a fuzzy power controller （FPC） that exploits the SINR， transmission power， and interference power level. An FPC uses genetic algorithms （GAs） to optimize parameters and can maximize throughput and minimize power consumption by updating the optimal transmission power according to the current environment.
　　As well as power?adjustment strategies， adaptive strategies can be used to mitigate inter?WBAN interference. In [13]， Yang et al. proposed a scheme that included adaptive modulation， adaptive data rate， and adaptive duty cycle. They also introduced the interference mitigation factor （IMF）， which is used to quantitatively analyze the effectiveness of their proposed scheme.
　　Co?channel interference?suppression methods such as power control based on game theory， power controller， and adaptive schemes are passive ways of mitigating inter?WBAN interference. Passively mitigating inter?WBAN interference involves constantly adjusting the transmission strategy according to the communication environment rather than actively altering the communication environment. This means compromising on performance when there is severe interference. In such situations， resource allocation strategies are ineffective.
　　3.4 Other Strategies
　　There are also other strategies for mitigating inter?WBAN interference [1]， [21]， [22]. In [1]， beacon?shifting strategies are introduced to protect the most important part of intra?WBAN communication， i.e.， transmission of the beacon， and make it easier for WBANs to coexist. The coordinator of a WBAN can transmit its beacon at different offsets relative to the start of the beacon periods in order to avoid repeated beacon collisions. The coordinator can also schedule allocation conflicts that occur as a result of coexistence. The offsets are decided by the unique beacon shifting sequence chosen by the coordinator. 　　Some auxiliary information can also be used to mitigate inter?WBAN interference. In [21]， the authors use social interaction detection to estimate the distance of interference. In [22]， De Silva et al. designed a fixed sensor network to predict potential interference. An interference prediction module was used to obtain the location of WBANs and RSSI， and the likelihood of interference was predicted according to the distance and RSSI level. Finally， a resource arbitrator was assigned channels that were different to those of the relevant WBANs in order to avoid potential interference.
　　3.5 Summary and Discussion
　　Tables 1 and 2 summarize different inter?WBAN interference mitigation solutions. These proposed solutions are all designed for different particular scenarios. When there are sufficient resources and a slowly changing network topology， resource?allocation methods such as channel assignment and time rescheduling are the most straightforward methods for mitigating inter?WBAN interference. However， in some scenarios， such as densely deployed WBANs， limited frequency resource and frequently changing network topology， resource allocation may not be appropriate because of the lack of channels and insufficient assigning time. Co?channel interference mitigation is suitable in such scenarios. With co?channel interference mitigation， the transmission strategy is adjusted according to the environment so that performance degradation is minimized. The density of WBAN deployment， rate of topology change， and number of time and frequency resources all affect the choice of inter?WBAN interference mitigation solution. Although we have classified these solutions as channel assignment， time rescheduling， and co?channel interference mitigation， they are not mutually exclusive. Co?channel interference mitigation solutions can complement resources?allocation solutions. To further improve existing solutions， there are some unresolved issues that need to be addressed.
　　4 Open Research Issues
　　4.1 Link Diversity
　　In [21]， [44]-[46]， intra?WBAN communication is modeled as a single link for simplified analysis. However， the links from intra?WBAN sensors to the coordinator are different because sensors are positioned differently and the body is moving. Hence， link diversity inside the WBAN needs to be considered when dealing with inter?WBAN interference.
　　4.2 Unequal QoS Requirements
　　Another aspect that is often overlooked is varied QoS requirements for sensors inside the WBAN and for different WBAN users. Intra?WBAN sensors may have different requirements in terms of transmission latency， data rate， priority and PLR， and the amount of interference they experience. Thus， the WBAN should not be regarded as a homogeneous network when designing an interference?mitigation solution. There are also personalized WBANs to serve individual needs， and these tend to have different performance demands. These different demands should not be ignored as well. 　　4.3 Mobility
　　A characteristic of a WBAN is random mobility. This is the combination of sensor mobility， due to body movements， and WBAN mobility， due to daily activities. Sensor mobility results in a change of intra?WBAN topology and internal link gain， which may lead to a failure of convergence of the power control game in a co?channel interference mitigation solution. Unconscious movements change the inter?WBAN topology and cause problems with the resource management strategy for inter?WBAN coexistence. For example， two WBAN users who are initially far apart may be assigned the same channel according to a certain resource?allocation strategy. When the users move close to each other， a collision occurs. Mobility creates some serious challenges for current interference mitigation solutions. However， in order to make the deployment of WBANs more widespread， more attention has to be paid to mobility. In [1]， [40]， [49]， [50]， the authors have made some preliminary attempts to analyze mobility in relation to mitigation of inter?WBAN interference.
　　4.4 Auxiliary Information
　　A WBAN is a body?centric network and the human behavior and neighboring environment would affect inter?WBAN coexistence. For example， the density of WBANs is very different in a coffee shop or subway station. The social relations between WBAN users may also affect the distance between WBANs. Though some efforts made in [21]， [22] have considered the environmental support， the social attributes of WBAN are still not be utilized sufficiently and need further investigated.
　　5 Conclusion
　　This paper presented a deep analysis of the inter?WBAN coexistence issue and provided a broad overview of the inter?WBAN coexistence and interference mitigation strategies. The solutions to solve the coexistence problem， including channel assignment strategies， time rescheduling strategies， co?channel interference mitigation strategies， etc.， were summarized in this work. Some constructive suggestions were also proposed for the study of inter?WBAN coexistence problem in the future.
　　References
　　[1] IEEE. （2012）. 802.15.6?2012—IEEE standard for local and metropolitan area networks—part 15.6： wireless body area networks [Online]. Available： standards.ieee.org/findstds/standard/802.15.6?2012.html
　　[2] D. M. Barakah and M. Ammad?uddin， “A survey of challenges and applications of wireless body area network （WBAN） and role of a virtual doctor server in existing architecture，” in 2012 Third International Conference on Intelligent Systems， Modelling and Simulation （ISMS）， Kota Kinabalu， Malaysia， 2012， pp. 214-219. doi： 10.1109/ISMS.2012.108. 　　[3] S. Ullah， H. Higgins， B. Braem， et al.， “A comprehensive survey of wireless body area networks，” Journal of Medical Systems， vol. 36， no. 3， pp. 1065-1094， Jun. 2012. doi： 10.1007/s10916?010?9571?3.
　　[4] M. Chen， S. Gonzalez， A. Vasilakos， et al.， “Body area networks： a survey，” Mobile Networks and Applications， vol. 16， no. 2， pp. 171-193， Apr. 2011. doi： 10.1007/s11036?010?0260?8.
　　[5] IEEE. （2010）. 802.15.1?2005—IEEE standard for information technology—local and metropolitan area networks—specific requirements—part 15.1a： wireless medium access control （MAC） and physical layer （PHY） specifications for wireless personal area networks （WPAN） [Online]. Available： standards.ieee.org/findstds/standard/802.15.1?2005.html
　　[6] E. Georgakakis， S. A. Nikolidakis， D. D. Vergados， and C. Douligeris， “An analysis of bluetooth， zigbee and bluetooth low energy and their use in WBANs，” Wireless Mobile Communication and Healthcare， vol. 55， pp. 168-175， 2011. doi： 10.1007/978?3?642?20865?2_22.
　　[7] IEEE. （2011）. 802.15.4?2011—IEEE standard for local and metropolitan area networks??part 15.4： low?rate wireless personal area networks （LR?WPANs） [Online]. Available： standards.ieee.org/findstds/standard/802.15.4?2011.html
　　[8] L. Januszkiewicz， “Simplified human body models for interference analysis in the cognitive radio for medical body area networks，” in 8th International Symposium on Medical Information and Communication Technology （ISMICT）， Firenze， Italy， Apr. 2014， pp. 1-5. doi： 10.1109/ISMICT.2014.6825246.
　　[9] R. Chavez?Santiago， K. E. Nolan， O. Holland， et al.， “Cognitive radio for medical body area networks using ultra wideband，” IEEE Wireless Communications， vol. 19， no. 4， pp. 74-81， Aug. 2012. doi： 10.1109/MWC.2012.6272426.
　　[10] I. Kirbas， A. Karahan， A. Sevin， and C. Bayilmis， “isMAC： an adaptive and energy?efficient MAC protocol based on multi?channel communication for wireless body area networks，” KSII Transactions on Internet and Information Systems， vol. 7， no. 8， pp. 1805-1824， Aug. 2013. doi： 10.3837/tiis.2013.08.004.
　　[11] B. R. Gonchikar， V. Geetha， and R. Kunabeva， “An adaptive routing mechanism for wireless body area network （WBAN），” International Journal of Innovative Research and Development， vol. 3， no. 6， Jun. 2014.
　　[12] H. Moungla， A. Jarray， A. Karmouch， and A. Mehaoua， “Cost?effective reliability?and energy?based intra?WBAN interference mitigation，” in IEEE Global Communications Conference （GLOBECOM）， Austin， USA， Dec. 2014， pp. 2399-2404. doi： 10.1109/GLOCOM.2014.7037167. 　　[13] W. Yang and K. Sayrafian?Pour， “Interference mitigation using adaptive schemes in body area networks，” International Journal of Wireless Information Networks， vol. 19， no. 3， pp. 193-200， Sept. 2012. doi： 10.1007/s10776?012?0192?2.
　　[14] T. Hayajneh， G. Almashaqbeh， S. Ullah， and A. V. Vasilakos， “A survey of wireless technologies coexistence in WBAN： analysis and open research issues，” Wireless Networks， vol. 20， no. 8， pp. 2165-2199， Nov. 2014. doi： 10.1007/s11276?014?0736?8.
　　[15] N. Torabi and V. C. M. Leung， “Realization of public m?health service in license?free spectrum，” IEEE Journal of Biomedical and Health Informatics， vol. 17， no. 1， pp. 19-29， Jan. 2013. doi： 10.1109/TITB.2012.2227117.
　　[16] Y. Kim， S. Lee， and S. Lee， “Coexistence of ZigBee?based WBAN and WiFi for health telemonitoring systems，” IEEE Journal of Biomedical and Health Informatics， vol. PP， no. 99， Jan. 2015. doi： 10.1109/JBHI.2014.2387867.
　　[17] M. U. Memon， L. X. Zhang， and B. Shaikh， “Packet loss ratio evaluation of the impact of interference on zigbee network caused by Wi?Fi （IEEE 802.11b/g） in e?health environment，” in IEEE 14th International Conference on e?Health Networking， Applications and Services （Healthcom）， Beijing， China， Oct. 2012， pp. 462-465. doi： 10.1109/HealthCom.2012.6379462.
　　[18] A. Zhang， D. B. Smith， D. Miniutti， et al.， “Performance of piconet co?existence schemes in wireless body area networks，” in IEEE Wireless Communications and Networking Conference （WCNC）， Sydney， Australia， Apr. 2010， pp. 1-6. doi： 10.1109/WCNC.2010.5506746.
　　[19] L. W. Hanlen， D. Miniutti， D. Smith， et al.， “Co?channel interference in body area networks with indoor measurements at 2.4 GHz： distance?to?interferer is a poor estimate of received interference power，” International Journal of Wireless Information Networks， vol. 17， no. 3-4， pp. 113-125， Dec. 2010. doi： 10.1007/s10776?010?0123?z.
　　[20] W. Xuan and C. Lin， “Interference analysis of co?existing wireless body area networks，” in IEEE Global Telecommunications Conference （GLOBECOM 2011）， Houston， USA， Dec. 2011， pp. 1 -5. doi： 10.1109/GLOCOM.2011.6133624.
　　[21] Z. Zhang， H. Wang， C. Wang， and H. Fang， “Interference mitigation for cyber?physical wireless body area network system using social networks，” IEEE Transactions on Emerging Topics in Computing， vol. 1， no. 1， pp. 121-132， Jun. 2013. doi： 10.1109/TETC.2013.2274430.
　　[22] B. de Silva， A. Natarajan， and M. Motani， “Inter?user interference in body sensor networks： preliminary investigation and an infrastructure?based solution，” in Sixth International Workshop on Wearable and Implantable Body Sensor Networks， Berkeley， USA， Jun. 2009， pp. 35-40. doi： 10.1109/BSN.2009.36. 　　[23] E. Sarra， H. Moungla， S. Benayoune， and A. Mehaoua， “Coexistence improvement of wearable body area network （WBAN） in medical environment，” in IEEE International Conference on Communications （ICC）， Sydney， Australia， 2014， pp. 5694-5699. doi： 10.1109/ICC.2014.6884229.
　　[24] W. Lee， S. H. Rhee， Y. Kim， and H. Lee， “An efficient multi?channel management protocol for wireless body area networks，” in International Conference on Information Networking， Chiang Mai， Thailand， Jan. 2009， pp. 1-5.
　　[25] C. Ren， H. Yu， C. Ma， et al.， “An energy?efficient spectrum handoff and access mechanism for the WBAN system，” in IEEE/CIC International Conference on Communications in China—Workshops （CIC/ICCC）， Xi’an， China， Aug. 2013， pp. 125-130. doi： 10.1109/ICCChinaW.2013.6670580.
　　[26] J. Mahapatro， S. Misra， M. Manjunatha， and N. Islam， “Interference?aware channel switching for use in WBAN with human sensor interface，” in 4th International Conference on Intelligent Human Computer Interaction （IHCI）， Kharagpur， India， Dec. 2012， pp. 1-5. doi： 10.1109/IHCI.2012.6481814.
　　[27] M. N. Deylami and E. Jovanov， “A distributed scheme to manage the dynamic coexistence of ieee 802.15.4?based health?monitoring WBANs，” IEEE Journal of Biomedical and Health Informatics， vol. 18， no. 1， pp. 327-334， Jan. 2014. doi： 10.1109/JBHI.2013.2278217.
　　[28] S. Movassaghi， M. Abolhasan and D. Smith， “Smart spectrum allocation for interference mitigation in wireless body area networks，” in IEEE International Conference on Communications （ICC）， Sydney， Australia， Jun. 2014， pp. 5688-5693. doi： 10.1109/ICC.2014.6884228.
　　[29] S. Movassaghi， M. Abolhasan， D. Smith， and A. Jamalipour， “AIM： adaptive internetwork interference mitigation amongst co?existing wireless body area networks，” in IEEE Global Communications Conference （GLOBECOM）， Austin， USA， Dec. 2014， pp. 2460-2465. doi： 10.1109/GLOCOM.2014.7037177.
　　[30] S. H. Cheng and C. Y. Huang， “Coloring?based inter?wban scheduling for mobile wireless body area networks，” IEEE Transactions on Parallel and Distributed Systems， vol. 24， no. 2， pp. 250-259， Feb. 2013. doi： 10.1109/TPDS.2012.133.
　　[31] S. Movassaghi， M. Abolhasan， and D. Smith， “Cooperative scheduling with graph coloring for interference mitigation in wireless body area networks，” in IEEE Wireless Communications and Networking Conference （WCNC）， Istanbul， Turkey， Apr. 2014， pp. 1691-1696. doi： 10.1109/WCNC.2014.6952484.
　　[32] P. R. Grassi， V. Rana， I. Beretta， and D. Sciuto， “B2IRS： a technique to reduce BAN?BAN interferences in wireless sensor networks，” in Ninth International Conference on Wearable and Implantable Body Sensor Networks （BSN）， London， UK， May 2012， pp. 46-51. doi： 10.1109/BSN.2012.30. 　　[33] E. Kim， S. Youm， T. Shon， and C. Kang， “Asynchronous inter?network interference avoidance for wireless body area networks，” The Journal of Supercomputing， vol. 65， no. 2， pp. 562-579， 2013. doi： 10.1007/s11227?012?0840?4.
　　[34] A. Jamthe， A. Mishra， and D. P. Agrawal， “Scheduling schemes for interference suppression in healthcare sensor networks，” in IEEE International Conference on Communications （ICC）， Sydney， Australia， Jun. 2014， pp. 391-396. doi： 10.1109/ICC.2014.6883350.
　　[35] J. Mahapatro， S. Misra， M. Manjunatha， and N. Islam， “Interference mitigation between WBAN equipped patients，” in Ninth International Conference on Wireless and Optical Communications Networks （WOCN）， Indore， India， Sept. 2012， pp. 1-5. doi： 10.1109/WOCN.2012.6331909.
　　[36] J. Mahapatro， S. Misra， M. Mahadevappa， and N. Islam， “Interference?aware MAC scheduling and admission control for multiple mobile WBANs used in healthcare monitoring，” International Journal of Communication Systems， vol. 28， no. 7， pp. 1352-1366， May 2015. doi： 10.1002/dac.2768.
　　[37] S. Wen， G. Yu， and W. Wai?Choong， “A lightweight inter?user interference mitigation method in body sensor networks，” in IEEE 8th International Conference on Wireless and Mobile Computing， Networking and Communications （WiMob）， Barcelona， Spain， Oct. 2012， pp. 34-40. doi： 10.1109/WiMOB.2012.6379098.
　　[38] K. Seungku， K. Seokhwan， K. Jin?Woo， and E. Doo?Seop， “Flexible beacon scheduling scheme for interference mitigation in body sensor networks，” In 9th Annual IEEE Communications Society Conference on Sensor， Mesh and Ad Hoc Communications and Networks （SECON）， Seoul， South Korea， Jun. 2012， pp. 157-164. doi： 10.1109/SECON.2012.6275772.
　　[39] S. Kim， S. Kim， J. Kim， and D. Eom， “A Beacon Interval Shifting Scheme for Interference Mitigation in Body Area Networks，” Sensors， vol. 12， no. 8， pp. 10930-10946， Aug. 2012. doi： 10.3390/s120810930.
　　[40] D. Jie and D. Smith， “Cooperative body?area?communications： Enhancing coexistence without coordination between networks，” in IEEE 23rd International Symposium on Personal Indoor and Mobile Radio Communications （PIMRC）， Sydney， Australia， Sept. 2012， pp. 2269 -2274. doi： 10.1109/PIMRC.2012.6362733.
　　[41] D. Jie and D. Smith， “Opportunistic relaying in wireless body area networks： coexistence performance，” in IEEE International Conference on Communications （ICC）， Budapest， Hungary， 2013， pp. 5613-5618.
　　[42] Z. Xie， G. Huang， J. He， and Y. Zhang， “A clique?based WBAN scheduling for mobile wireless body area networks，” Procedia Computer Science， vol. 31， pp. 1092-1101， 2014. doi： 10.1016/procs.2014.05.364. 　　[43] Z. Lei， L. Bin， C. Chang， and W. C. Chang， “Bayesian game based power control scheme for inter?WBAN interference mitigation，” in IEEE Global Communications Conference （GLOBECOM）， Austin， USA， Dec. 2014， pp. 240-245. doi： 10.1109/GLOCOM.2014.7036814.
　　[44] G. Wu， J. Ren， F. Xia， et al.， “A game theoretic approach for interuser interference reduction in body sensor networks，” International Journal of Distributed Sensor Networks， vol. 2011， article ID 329524， 2011. doi： 10.1155/2011/329524.
　　[45] M. Nazir and A. Sabah， “Cooperative cognitive WBAN： from game theory to population dynamism，” in 3rd International Congress on Ultra Modern Telecommunications and Control Systems and Workshops （ICUMT）， Budapest， Hungary， Oct. 2011， pp. 1-6.
　　[46] R. Kazemi， R. Vesilo， E. Dutkiewicz， and F. Gengfa， “Inter?network interference mitigation in wireless body area networks using power control games，” in International Symposium on Communications and Information Technologies （ISCIT）， Tokyo， Japan， Oct. 2010， pp. 81-86. doi： 10.1109/ISCIT.2010.5664908.
　　[47] R. Kazemi， R. Vesilo， E. Dutkiewicz， and P. L. Ren， “Design considerations of reinforcement learning power controllers in wireless body area networks，” in IEEE 23rd International Symposium on Personal Indoor and Mobile Radio Communications （PIMRC）， Sydney， Australia， Sept. 2012， pp. 2030-2036. doi： 10.1109/PIMRC.2012.6362688.
　　[48] R. Kazemi， R. Vesilo， and E. Dutkiewicz， “A novel genetic?fuzzy power controller with feedback for interference mitigation in wireless body area networks，” in IEEE 73rd Vehicular Technology Conference （VTC Spring）， Yokohama， Japan， May 2011， pp. 1-5. doi： 10.1109/VETECS.2011.5956462.
　　[49] Z. Chen and M. L. Sichitiu， “N?body： social based mobility model for wireless ad hoc network research，” in 7th Annual IEEE Communications Society Conference on Sensor Mesh and Ad Hoc Communications and Networks （SECON）， Boston， USA， Jun. 2010， pp. 1-9. doi： 10.1109/SECON.2010.5508280.
　　[50] M. Nabi， M. Geilen， and T. Basten， “MoBAN： a configurable mobility model for wireless body area networks，” in 4th International ICST Conference on Simulation Tools and Techniques， Barcelona， Spain， Mar. 2011， pp. 168-177. doi： 10.4108/icst.simutools.2011.245511.
　　Manuscript received： 2015?03?26