, pp. 205-216 http://dx.doi.org/10.14257/ijseia.2015.9.9.18 Adaptive MAC Protocol for Emergency Data Transmission in Wireless Body Sensor Networks Rae Hyun Kim and Jeong Gon Kim 1 Department of Electronic Engineering, Korea Polytechnic University (KPU) Si Heung City, Kyunggi Do, 429-793, KOREA hjkl525@naver.com, jgkim@kpu.ac.kr Abstract Recently, due to the increase in social intertest for advanced wireless telecommunication techniques, activated ubiquitous environments and healthcare, the application range of WBSN has been consistently expanding. WBSN environment has objective of collecting various bio-signals created directly or indirectly from in and out of the body, connecting with external networks and monitoring the display tool. Various biosignals created in WBSN environment have unpredictable rate of emergency data occurrence by its nature. Such emergency data should be transmitted with priority compared to other periodic signals for prompt handling according to the condition of the patient. MAC protocol proposed in this thesis is in form of TDMA based on competitive reserved allocation by time-slot in CSMA/CA environment. This MAC protocol classifies the types of bio-signals with each different characteristic and settles the priority by type and the minimum transmission delay time of data accordingly. Especially for transmission of emergency data, MAC protocol reduces the transmission delay of emergency data and packet loss by setting priority. Keywords: WBSN, WBAN, DTD-MAC, MED-MAC(1), MED-MAC(2), CSMA/CA, TDMA, GTS, PET 1. Introduction WBSN (Wireless Body Sensor Network) refers to a network that allows data transmission in and out of the body based on WSN, (Wireless Sensor Network) [1] and it could be considered as an applied network composed based on WBAN (Wireless Body Area Network) [2] that is the standard of IEEE 802.15.6 regarding telecommunication method and transmission quality. Figure 1 shows the arrangement diagram of WSBN. With the purpose of monitoring the condition of patients at real-time and transmitting to external network, WBSN collects various bio-signal data from the devices or nodes located on the surface or inside of human body. It is forecasted that the WBSN will replace the current wired method of medical observation and monitoring environment [3]. When numbers of nodes transmit data at the same time in the environment of sensor network, CSMA/CA (Carrier Sensed Multiple Access/Collision Avoidance) [4], known widely as MAC protocol, performs operational degradation at serious level and triggers extremely high consumption of energy from repeated attempts of transmission, due to high frequency of idle listening and packet collision. Therefore, there had already been many research results and generalizations that TDMA (Time Division Multiple Access) method is advantageous in many ways regarding electricity consumption and reduction in delay of devices or nodes [5, 6]. Meanwhile, there may be general and urgent types of data among the data that the nodes transmit to coordinators. Particularly, the emergency 1 : Corresponding Author ISSN: 1738-9984 IJSEIA Copyright c 2015 SERSC
data should have the condition to be transmitted immediately after the occurrence by its nature. Figure 1. WBSN Network Environment IEEE 802.15.4 MAC Protocol [7] that has been applied for transmission of bio-signals until recently a hybrid method that applies both competition-oriented, which is universally used for data processing in WBSN environment, and schedule-oriented method. However, the allocation method of GTS (Guaranteed Time Slot) of IEEE 802.15.4 MAC Protocol applies FIFS (First In First Service) queuing method where the channels are allocated by the order or arrival, resulting inevitable delay in transmission, and is inadequate for transmitting and handling of emergency data that should be transmitted without any delay at the time of occurrence [8]. The purpose of this thesis is to reduce and minimize the rate of inevitable transmission delay and packet loss in the process of handling emergency data occurring at random rate with the same method of handling the general data. In addition, two forms of MAC protocol that guarantees fast processing of emergency data as well as enhanced processing of general data through process of packet data by varying structures of Super-Frame according to the occurrence rate of urgent and general data. The structure of this thesis is as follows In Chapter 2, the existing methods of DTD (Decrease of Transmission Delay)-MAC Protocol [9] that reduces the delay in transmission and packet loss in IEEE 802.15.4 MAC Protocol and WBSN environment will be discussed. In Chapter 3, the two types of MAC protocols, MED-MAC (1) Protocol and MED-MAC(2) Protocol [10], will be proposed and discussed in order to reduce the delay in transmission of emergency data and the packet loss. In Chapter 4, the functions of MAC protocol proposed will be analyzed through simulation and comparison with the existing methods. Finally in Chapter 5, the conclusion and the future research topics and directions will be suggested. 2. Related Research 2.1. IEEE 802.15.4 MAC Protocol Figure 2. IEEE 802.15.4 MAC Protocol 206 Copyright c 2015 SERSC
IEEE 802.15.4 MAC Protocol forms Super-Frame structure where synchronization is executed through Beacon Frame as shown in Figure 2. It is composed of the active period where communication of data is carried out between the nodes and the coordinators and the inactive period where there is no communication of data. The active period is composed of 16 even slots and is formed based on sections of CAP (Contention Access Period), a competition-oriented method, and GTS (Guaranteed Time Slot) of CFP (Contention Free Period), a non-competition-oriented method that IEEE802.15.4 MAC Protocol may be referred as hybrid MAC Protocol. The device may be allocated with 1 coordinator or more than 1 GTS. In addition, since there are 16 even slots in active period, as the more devices are allocated with GTS, the length of CAP section is relatively reduced as well. IEEE 802.15.4 MAC Protocol may allocate GTS to maximum 7 devices within the active period, and when all 7 are allocated, the rest of the devices requesting GTS may not receive the GTS. WBSN environment is the assembly composed of each different node with limited numbers. In addition, since the main data is bio-signals, there are restrictions regarding energy. With different bio-signals of each patient, the data should guarantee reliability and fairness while being made easy for observation of biosignals of patients at real time through display tools. 2.2. DTD (Decrease of Transmission Delay) - MAC Protocol Figure 3. DTD-MAC Protocol Super-Frame Figure 3 shows the Super-Frame structure of DTD-MAC protocol in the existing research method. In the Super-Frame of DTD-MAC protocol, the action performed in the frame of PSAP section is allocating and scheduling the information related to GTS Time- Slot allocation of next period in current period. GTS Time-Slot is assembled in order based on the Time-Slot reservation information received from PSAP frame of previous period. Beacon frame synchronizes the GTS Time-Slot information received from PSAP frame of previous period in current period and settles the order of GTS Time-Slot. The priority of each bio-signal node proposed in this thesis referred the functional requirements of WBAN as shown in Table 1 below. Table 1. Requirements of WBAN Application [11] Copyright c 2015 SERSC 207
When each bio-signal nodes, corresponding to ECG, EEG and EMG, attempts to transmit the data from the coordinator for channel allocation and reaches the maximum transmission delay of 250ms, the original priority or each node is neglected and is given with the most prior for channel allocation compared to other nodes. In addition, in the section where not all of the nodes transmit the data for channel allocation, the original priority is referred for each node in the buffer. However, when the maximum transmission delay reaches 250ms after the delay of the buffer data of each node, there is change in priority temporarily and the corresponding node is guaranteed with the most prior transmission. On the other hand, the packets of nodes that failed to transmit the data at the same time are counted as transmission delay or loss. DTD-MAC protocol has principle of transmitting data within 250ms, the maximum transmission delay from its occurrence. However, DTD-MAC protocol only considered the environment where there is no emergency data that there may be possible disadvantages due to lack of flexible response in processing of emergency data. For emergency data that may occur at irregular rates, data should be transmitted in real time to be effective. However, since DTD-MAC protocol uses the same method for handling and transmitting the emergency data that requires real-time transmission, and the general data, continued delay in transmission of emergency data and accordingly, high energy consumption and packet loss occur. 3. Proposed MAC Protocol Figure 4. MED-MAC Protocol Super-Frame Figure 4 shows the Super-Frame of MED-MAC protocol that secured the flexibility in processing of emergency data by adding UP (Urgent Period) frame section to the original DTD-MAC Super-Frame. The proposed MAC protocol adopts the variable Super-Frame structure where the structure changes according to the premeasured relative occurrence rate of urgent and general data. Therefore, the length of UP frame will be varied according to the data amount of Emergency Data Node measured from PET frame for each period. If there is no emergency data of each node measured from PET frame in above Super-Frame, UP frame will disappear and there will be conversion to GTS Time-Slot section for processing and transmission of general data. The Inactive Period frame refers to the section where the schedule information of GTS Time-Slot reservation allocation and UP frame of next period and the data are not transmitted within PET frame. Figure 5 below shows the competition process of channel allocation for general or emergency data of each different node during equal transmission period. 208 Copyright c 2015 SERSC
Figure 5. An Example of PET Frame for Channel Competition As shown in Figure 5, the emergency data may be classified with special tag added, unlike general data. It indicates when bio-signals with different features are transmitted at the same time through 1 slot. Especially in this thesis, transmission of emergency data will be prioritized than the general data by principle while guaranteeing the equal functionality in transmission and processing of general data. Through the two proposed methods, handling of emergency data according to its occurrence rate will be controlled for an optimized solution of processing general and emergency data will be sought as the most important consideration of this thesis. 3.1. Prioritized Allocation and Transmission Method for Emergency Data (MED-MAC(1)) In an environment with high occurrence rate or emergency data, all types of emergency data are guaranteed with priority for channel allocation and transmission before the nodes of all other general data under situations where emergency and general data of each node or emergency or general data themselves compete for channel allocation. Therefore, smooth transmission of data is guaranteed for emergency data flowing in real-time through fast processing of vast amount of emergency data, improving the credibility of emergency data transmission and the QOS of system. On the other hand, the standard of transmission within 250ms is applied for the general data, but with priority to emergency data in all cases. In addition, when the emergency data compete with each other for channel allocation, the priority is given as the priority arranged in advance. However, it is expected to have disadvantage of relative delay in transmission of general data and increase in packet loss will occur compared to the environment where the occurrence rate or emergency data is between 10~20%, compared to MED-MAC(2) Protocol. 3.2. Transmission Method to Reduce Delay in Emergency Data (MED-MAC(2)) In a situation where urgent and general data of each node or the emergency and general data compete for channel allocation, each emergency data is classified according to the priority standards (ECG > EEG > EMG) and the emergency data should be transmitted within the maximum transmission delay of 100ms by principle. If the emergency data is not transmitted within the suggested minimum transmission delay time, it results transmission delay and packet loss. In contrast, all general data that occurred in the same period or competed for channel allocation should be transmitted within the maximum transmission delay of 250ms by principle. If it fails, data flows into the buffer and are kept in sustained standby condition and the transmission delay here is added up to the total transmission delay time as well. The general data that gave way to the emergency data in competition for channel allocation should be transmitted within 250ms by principle. If the general data had reached the maximum transmission delay of 250ms while the emergency data had not, and when the two attempts for channel allocation, the general data that reached the maximum transmission delay is allocated with the channel. When emergency data competes with another emergency data for channel allocation, the existing priority rank is applied for classification and allocation. If one of he emergency data of node that reached Copyright c 2015 SERSC 209
the maximum transmission delay period while the other emergency data with higher priority had not, the priority is temporarily exchanged for allocation to the urgent data that arrived first. If 2 or more different nodes attempt to compete for channel allocation, the existing priority ranking should be reflected for allocation under the condition where all nodes reach the maximum transmission delay in current period. However, the emergency data of the node that exceeded the maximum transmission delay, while not being allocated with channel, are added up to the packet loss and the transmission delay. Therefore, in the environment where emergency data intermittently occurs, the fast processing of emergency data and adequate level of processing of general data will likely be guaranteed. 4. Performance Comparison In this thesis, WBSN network environment is composed of Star-Topology and simulated with Visual Studio C++, and the length of 1 Time-Slot was defined as 50ms. Each period is composed of 100 Time-Slots and the average collected from the 100 periods were calculated. The priority of general data was ECG > EEG > EMG, and the occurrence rate of general data of EEG was defined as twice of ECG, EEG as 18 times of EMG, based on WBAN requirements. The maximum transmission delay of each node was arranged to be 250ms from the moment of occurrence by principle. If the data is not transmitted within the maximum transmission delay of 250ms, it was added to the total transmission delay and packet loss. The occurrence rate of emergency data was defined as he 10%, 20% and 30%, and an environment for competition with general data was arranged. The priority was defined as ECG > EEG > EMG in order same as the general data. The functional assessment was executed under assumption that the general and emergency data will occupy 60% and 120% of the whole Time-Slot. Also, the maximum transmission delay was defined as 100ms for the emergency data for the method of transmitting the emergency data always prior to the general data (MED-MAC (1)) and the competitive condition with general data (MED-MAC (2)) was applied for the functional assessment, together with the comparison evaluation of the original DTD-MAC protocol with the average packet loss. In the simulation of this research, it was assumed that 60% of transmittable Time-Slot will be the targeted bio-signals for transmission and that 120% of data traffic will occur. In addition, the condition under 10% of emergency data and 90% of general data and under 30% of emergency data and 70% of general data was carried out respectively. Figure 6 and 7 presents the simulation of the average transmission delay of general data and the simulation of average transmission delay of emergency data under environment of 60% occurrence of whole Time-Slot and under 10% of emergency data and 90% of general data. Here, MED-MAC (1) is the transmission method of prioritized allocation, suggested in 3.1, and MED-MAC (2) shows the result of transmission method of reducing delay, suggested in 3.2. From the two images, the transmission delay of MED-MAC (1) is smaller in both, and for the general data, the average transmission delay of DTD-MAC Protocol or MED-MAC (1), MED-MAC (2) and the three MAC-protocol show the value that are almost equal. 210 Copyright c 2015 SERSC
Figure 6. Average Time Delay of General Data in case of 60 % of Data Traffic in Available Time Slot (E.D : 10%, G.D : 90%) Figure 7. Average Time Delay of Emergency Data in case of 60 % of Data Traffic in Available Time Slot (E.D : 10%, G.D : 90%) Figure 8 and 9 below presents the simulation results of average transmission delay of each general and emergency data under condition of 60% of occurrence rate of data traffic out of the total Time-slot, with 30% of emergency data and 70% of general data. As the occurrence rate of emergency data increases, it is found that the suggested method of general data faces slightly lower time delay. In addition, as the two suggested method, MED-MAC (1) and MED-MAC (2), face smaller gap of functional differences as the traffic of emergency data increases. Figure 8. Average Time Delay of General Data in case of 60 % of Data Traffic in Available Time Slot (E.D : 30%, G.D : 70%) Copyright c 2015 SERSC 211
Figure 9. Average Time Delay of Emergency Data in case of 60 % of Data Traffic in Available Time Slot (E.D : 30%, G.D : 70%) Figure 10 and 11 below presents the results of average transmission delay simulation of each general and emergency data under condition of 120% of occurrence rate of data traffic out of the total possible Time-Slot with 10% of emergency data and 90% of general data. As the data traffic increases, there is less and less time delay for general data, and the lowest transmission delay of suggested method occurs only for emergency data. Also, the rate of emergency data occurs intermittently that the delay of MED-MAC (1) is maintaining a very low value compared to that of MED-MAC (2). Figure 10. Average Time Delay of General Data in case of 120 % of Data Traffic in Available Time Slot (E.D : 10%, G.D : 90%) Figure 11. Average Time Delay of Emergency Data in case of 120 % of Data Traffic in Available Time Slot (E.D : 10%, G.D : 90%) 212 Copyright c 2015 SERSC
Figure 12 and 13 presents the results of average transmission delay simulation for each general and emergency data under condition of 120% occurrence rate of data traffic out of the total Time-Slot with 30% of emergency data and 70% of general data. As the data traffic and the occurrence rate of emergency data increases, all methods show the delay time of almost the same level for the general data. Also, the difference of delay in emergency data of the two suggested methods, MED-MAC (1) and MED-MAC (2), is shown to be very similar as the traffic of emergency data increases. Figure 12. Average Time Delay of General Data in case of 120 % of Data Traffic in Available Time Slot (E.D : 30%, G.D : 70%) This shows the possibilities for more effectiveness of MED-MAC (2) than MED-MAC (1) as the data traffic increases and the rate of emergency data increases. Figure 13. Average Time Delay in Case of 120 % of Data Traffic in Available Time Slot (E.D : 30%, G.D : 70%) 5. Conclusion In this thesis, MAC protocol with the objective of quick processing of emergency data and reduction in the transmission delay of emergency data under WBSN environment was suggested and assessed. DTD-MAC protocol that was taken for precedent research under WBSN environment does not have the standards related to processing of emergency data that the emergency data were processed with the same standards of transmission delay for general data, resulting inevitable increase in transmission delay and inadequate environment for emergency data, requiring real-time processing. The two MAC protocol proposed in this thesis provide methods where the emergency data are given with the absolute priority depending on the amount of emergency and general data and another method that guarantees the consistent transmissible function of general data by restricting the maximum transmission delay of emergency data to 100ms. As a result of functional assessment the prioritized transmission method was very effective when the rate of Copyright c 2015 SERSC 213
emergency data was low, through the experiment. In addition, in an environment where data traffic rapidly increases, the method of applying maximum delay of emergency data, had guaranteed an adequate level of function in transmission and processing of general data, showing the possibility of realizing optimized processing of emergency data, and this led to the need of research for solution of appropriate combination. References [1] J. Su Kim, J. Hyun Lee and K. Wook Rim, Energy Efficient Key Management Protocol in Wireless Sensor Networks, IJSIA, vol. 4, no. 2, (2010) April 1-12. [2] J. Sung Choi and J. Gon Kim, An Improved MAC Protocol for WBAN through modified frame structure, International Journal of Smart Home (IJSH), vol.8, no. 2, (2014). [3] O. G. Morchon, H. Baldus and D. S. Sanchez, Resource Efficient Security for Medical Body Sensor Networks, Wearable and Implantable Body Sensor Networks, BSN International Workshop, 4pp, (2006) April. [4] LAN-MAN Standards Committee of the IEEE Computer Society. Wireless LAN Medium Access Control(MAC) and Physical Layer(PHY) Specification. IEEE, New York, NY, (1997). [5] S. A. Gopalan and J. Park, Energy Efficient MAC Protocols for Wireless Body Area Network: Survey, International Congress on Ultra Modern-Telecommunications and Control Systems and Workshops, (2010) October, pp. 739-744. [6] H. Kwon and S. Lee, Energy-efficient multi-hop transmission in Body Area Networks, IEEE PIMRC, Tokyo, (2009), pp. 2141-2146. [7] IEEE 802.15.4 Standard-2003, Part 15.4: Wireless Medium Access Control(MAC) and Physical Layer(PHY) Specifications for Low-Rate Wireless Personal Area Network(LR-WPANs), (2003). [8] H. Geon Lee, K. Lee and Y. Shin, A Priority Based MAC Protocol for Emergency Data Transmission in Wireless Body Area Networks, Telecommunication Journal of the institute of electronics engineers of korea, vol. 49, no. 4, (2012). [9] R. Hyeon Kim and J. Gon Kim, Delay Reduced MAC Protocol for Bio Signal Monitoring in the WBSN Environment, 2015 6 th International Workshop Series, Advanced Science and Technology Letters, (Networking and Communication 2015), vol. 94, pp. 42-46. [10] R. Hyeon Kim, P. Soo Kim and J. Gon Kim, A Study on MAC Protocol for emergency data transmission in Wireless Bio Signal Monitoring Environment, 2015 7 th International Interdisciplinary Workshop Series, Advanced Science and Technology Letters, (Mechanical Engineering 2015), vol. 108, pp. 93-98. [11] R. Cavallarari, F. Martelli, R. Rosini, C. Buratti and R. Verdone, A Survey on Wireless Body Area Networks: Technologies and Design Challenges, IEEE Communications Surveys & Tutorials, Third Quarter, vol. 16, no. 3, (2014). Authors Rae Hyeon Kim, was born in Daegu, Korea on May 24, 1990. He Graduated Medical Engineering Department at ANDONG Science College in February 2012. He Graduated Electronics Engineering Department at Korea Polytechnic University in February 2014. He Enrolled in Masters in Electronics Engineering at Korea Polytechnic University since February 2014 until present. Interests: WBAN, WBSN realization and applied techniques, Cooperative transmission, MIMO Jeong Gon Kim, was born in Seoul, Korea on May 24, 1969. He received the B.S., M.S. and Ph.D, degrees all in electrical engineering from Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea in 1991, 1993 and 1998, respectively. From 1998 to 1999, he was the 214 Copyright c 2015 SERSC
Post Doctoral Research Fellow at the University of Hawaii at Manoa, USA, From 1999-2001, he joined R&D center of LG Telecom, Korea and is involved in IMT-2000 radio access technology development. From 2001-2003, he was also involved in 3GPP physical layer standardization, concentrating on the TDD mode in the Telecommunication Research center of Samsung Electronics. Since 2003, he is now a Professor at the Department of Electronics Engineering of Korea Polytechnic University. His research Interests now include the design and performance analysis of wireless communication system, specially 5G mobile communication, MIMO, cooperative communication, and WBAN application. Copyright c 2015 SERSC 215
216 Copyright c 2015 SERSC