Scholarly article on topic 'Analyzing and Evaluating Contention Access Period of Slotted CSMA/CA for IEEE802.15.4'

Analyzing and Evaluating Contention Access Period of Slotted CSMA/CA for IEEE802.15.4 Academic research paper on "Computer and information sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Computer Science
OECD Field of science
Keywords
{802.15.4 / "Slotted CSMA/CA" / MAC / Physical / "Contention Access Period."}

Abstract of research paper on Computer and information sciences, author of scientific article — D. Mahmood, Z.A. Khan, U. Qasim, M. Umair Naru, S. Mukhtar, et al.

Abstract IEEE standard 802.15.4 is widely studied due to its huge applicability. This vary study is based upon analyzing techniques and methodology discussed in IEEE 802.15.4 standard in context of contention access period. According to standard specifications, there are two minor differences in CSMA/CA algorithm used in CAP along with different frequency ranges. These frequency ranges are accepted in different geographical regions of this world. Considering medium access algorithm used, one flavor supports ACK frame after successful transmission and other do not. ACK mode of CSMA/CA is widely analyzed analytically and simulated in earlier conducted researches,however, there is no study discussing behaviour of non-ACK mode. This non-ACK mode is prescribed as a different flavor in IEEE 802.15.4 for applications that do not focus on taking ACK packet after every transmission. In this work, we modify a markov chain model for non-ACK mode and compare both modes analytically followed by extensive simulations and discussions. For health care applications ACK mode shows its worth however, in streaming data or playing games, non-ACK mode can be preferred due to its lower delay, higher throughput and lower control load.

Academic research paper on topic "Analyzing and Evaluating Contention Access Period of Slotted CSMA/CA for IEEE802.15.4"

(8)

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedia Computer Science 34 (2014) 204 - 211

The 9th International Conference on Future Networks and Communications (FNC-2014)

Analyzing and Evaluating Contention Access Period of Slotted

CSMA/CA for IEEE802.15.4

D. Mahmooda, Z. A. Khanb, U. Qasimc, M. Umair Narua, S. Mukhtard,

M. I. Akrama, N. Javaida

a COMSATS Institute of Information Technology, Islamabad, Pakistan bInternetworking Program, FE, Dalhousie University, Halifax, Canada c University of Alberta, Alberta, Canada dBahria Univerisity Islamabad, Pakistan

Abstract

IEEE standard 802.15.4 is widely studied due to its huge applicability. This vary study is based upon analyzing techniques and methodology discussed in IEEE 802.15.4 standard in context of contention access period. According to standard specifications, there are two minor differences in CSMA/CA algorithm used in CAP along with different frequency ranges. These frequency ranges are accepted in different geographical regions of this world. Considering medium access algorithm used, one flavor supports ACK frame after successful transmission and other do not. ACK mode of CSMA/CA is widely analyzed analytically and simulated in earlier conducted researches,however, there is no study discussing behaviour of non-ACK mode. This non-ACK mode is prescribed as a different flavor in IEEE 802.15.4 for applications that do not focus on taking ACK packet after every transmission. In this work, we modify a markov chain model for non-ACK mode and compare both modes analytically followed by extensive simulations and discussions. For health care applications ACK mode shows its worth however, in streaming data or playing games, non-ACK mode can be preferred due to its lower delay, higher throughput and lower control load. © 2014 Elsevier B.V. Thisisanopenaccess article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of Conference Program Chairs Keywords: 802.15.4; Slotted CSMA/CA; MAC; Physical; Contention Access Period.

1. Introduction

World is changing, automation in every aspect of life is the current need. For that automation, we humans are now surrounded with processors to process different attributes. Here comes the domain of communication where these processings normally called as information is sent, stored and utilized in a beneficial way. Combining processing, transmitting, storing and utilizing that information gives birth to networks; initially wired and then wireless. Entering

* Nadeem Javaid, www.njavaid.com, Tel.: +92-300-579-2728 E-mail address: nadeemjavaid@comsats.edu.pk, nadeem.javaid@univ-paris12.fr

1877-0509 © 2014 Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of Conference Program Chairs doi:10.1016/j.procs.2014.07.011

wireless domain, there are numerous networks proposed for different applications. Amongst these, wireless sensor networks (WSN's) are emerging field that has enormous potential for applications in it. These networks consists of tiny sensors having limited energy and low computational power.

IEEE 802.15 working group is specified for wireless personal area networks (WPAN)and is further divided into different extensions. Within these extensions, IEEE 802.15.4 [1] is designed for low rate and longer battery life. Due to simplicity, longer life time and wide range of applications, IEEE802.15.4 standard has got focus light in research as well as industrial aspect. This standard contrary to other methodologies (WiFi) stresses on least possible cost for communication between devices with minimal underlying resources. This standard is designed for a communication range of 10 meters along with data rate of 250kbps.

The said standard stipulates MAC and Physical Layers and provides basis to different technologies i.e. ZigBee, ISA100, WirelessHART and MiWi alliances. For MAC sublayer, a device can operate in contention free period or contention access period. In contention free period, a node may reserve guaranteed time slots while in contention access period, carrier sense multiple access with collision avoidance is used with some modifications from earlier standards. In physical layer, IEEE802.15.4 is operable on three different frequency assignments. This work as the title indicates, focuses on workability of contention access period of IEEE802.15.4 under three different frequency assignments.

2. Related Work

In IEEE802.11 [2] if a node that has a packet to transmit, gets channel access, decrements backoff counter by one other wise, it is blocked. Considering IEEE802.15.4, backoff counter decrements whether channel is accessed or found busy. Due to such variations, Researchers were interested in establishing mathematical models regarding IEEE802.15.4. Initial researches [4-7] were conducting using markov chain models without taking superframe structure into account. [8] enhanced the research by considering super frame structure as well. Modeling of said standard is further enhanced by [9] that proposed a model for slotted CSMA/CA. Authors of [10] calculated probabilistic values of reliability, energy consumption and delay which was further modified by [11] and [12]. The study regarding performance analysis of all frequency assignments that are mentioned in standard text [1] was presented by [13]. Considering medium access techniques of contention access period, [14-16] give comparison of different schemes to be applied on body area networks. Access schemes like time and frequency division multiple access, pure and slotted Aloha along with CSMA/CA were observed using key metrics of path loss, delay and good put. CSMA/CA performance in IEEE802.15.4 is widely studied however, for the applications where ACK is not utterly needed after transmission of a packet, there is specified a flavor in standard text that is non-ACK mode. Functioning of CSMA/CA remains same but with absence of this control packet. In this study, we compare both of these modes i.e. ACK and non-ACK using markov transitional model (enhanced version of [10] and [12]) under all three frequency assignments that are mentioned in [1] for the said standard. For comparison analysis, we calculate transmission, delay and channel accessing probabilities using Markov model which are the key features for any protocol and simulate them using MATLAB.

Rest of the paper is organized as: Overview of IEEE 802.15.4 (MAC and Physical Layer aspect) in section 3, Probabilistic Modeling regarding ACK and non-ACK mode of CSMA/CA is presented in section 4 while simulated results and logical reasoning is provided in section 5 followed by conclusion of this paper.

3. An Overview: IEEE 802.15.4

3.1. MAC SUBLAYER

MAC layer provides data as well as management services. Management services tends to interface between upper layers with MAC layer while data services are concerned with seamless functioning of transmission/reception (MPDU's) procedures in accordance with physical layer. Considering IEEE 802.15.4 standard, the major role of this sub layer do not change however, there are some modifications regarding MAC of previous standards [11]. Super frame structure of MAC as illustrated in fig. 1 is classified into two classes; active and inactive. Within Active part of super frame structure, all operations regarding data transmissions and receptions are undertaken. In slotted version,

Fig. 1. Superframe Structure

this active period is divided into equally distributed time slots. Every transmission begins at the start of one of these time slots to ensure minimum wastage of resources due to packet collision if it happens. Besides such functionalities, active period further has two domains. Contention Access Period (CAP) and Contention Free Period (CFP). In contention free period, a point coordinator takes control of channel allocation and every node that opt for CFP is given specified time slots for their transmissions, these nodes do not have to content for channel accessing. Whereas on other hand, CAP is based on best effort procedures. All nodes that opt for CAP mode of active period have to content with each other to get channel access. IEEE 802.15.4 specified CSMA/CA algorithm for efficient accession of channel. Though this protocol is already widely accepted and applied however, in this standard it has some slight changes in accordance with its application scenario and limitations (power and computational limitations). In IEEE 802.15.4 [1] CSMA/CA performs two clear channel assessments convectively to get access of channel which is not in the case of [2] where only one clear channel assessment is under taken.We, in this work focus on contention access period for further analyzing and evaluating. Like any other protocol, the said protocol has different stages. When a network underlying this protocol initiates, a node being part of this network stages it self in idle state. In this state, it actually is waiting for any data packet from upper layers to transmit. Besides three counters are generated as well, i.e. back off counter, contention window counter and retransmission counter. At this stage, MAC layer has to reset the following parameters as:Number of backoffs (NB = 0), Contention window (CW = 2), Backoff exponent (BE = macMinBE) and Retransmission times (RT = 0).

When a packet from network layer reaches MAC layer, the idle state ends and now, the node is in channel sensing state. In between channel sensing and idle state, there is a random back off period within the range of o- to- 2BE-1. Channel access is granted after channel sensing state that comprises of two consecutive successful clear channel assessments (CCA). After completion of 1 st CCA, The contention window counter is decremented by one and after successful CCA2, CW counter reaches to 0. This means that channel is accessed and data packet is to be transmitted ending the state of channel sensing and now it is in transmission state. If CCA fails, back off counter is incremented and packet again has to wait a random back off time till maximum value of macMaxCSMABackoffs and macMaxBE. When value reaches the said threshold, packet is discarded. In transmission state, either packet is successfully transmitted or a

Table 1. MAC parameters for Slotted CSMA/CA (IEEE 802.15.4).

Parameters Value

macMinBE 3

macMaxBE 5

macMaxFrameRetries 3

ACK reception Time aTurnaroundTime + aUnitBackoff Period

LIFS (for Larger packet) 40 Symbols

IFS (for smaller packet) 12 Symbols

collision occurs. If packet is transmitted successfully, node again gets into idle state waiting for next packet to come from network layer. However, if collision occurs retransmission counter is incremented and data packet is set into wait for random back off time. After completion back off time, contention window counter and back off counter are set as default (CW = 2, BE = macMinBE). Afterwards, again channel is to be sensed , accessed and then transmit according to described procedures. There is also a threshold for retransmission retries. If number of retries reaches

Acknowledged Transmissions

Short Frame

Non-Acknowledged Transmissions

Short Fran

"T LIFS () ( sifs ()

Fig. 2. Packet Transmission in IEEE 802.15.4

the value of macMaxFrameRetries, packet is discarded throwing node into idle state. Within all procedure, there is a variation in protocol. ACK mode or non - ACK mode. In ACK mode, after successful reception of a packet, receiver issue this control packet ensuring reception of data packet. If sender fails to receive ACK packet due to any reason, retransmission counter is incremented and packet is to be transmitted again as per protocol. Transmission of ACK packet do not follow any protocol as it is just issued on successful reception of data packet[1]. Considering non - ACK mode, the protocol functioning remain same with absence of this control packet. In this mode, data is transmitted and if no collision occurs, it is understood as successful transmission. Having two clear channel assessment already has eliminated major chances of collision. This statement is discussed briefly in later sections of this study. During communication process, MAC layer specified different vacant spaces between transmission of two frames. These vacant spaces are termed as inter frame space and are placed to avoid any chance of collision due to propagation delay and other factors. There are certain types of inter frame spaces prescribed, however, there are two major types in which we are interested i.e. short inter frame space (SIFS) and large inter frame space (LIFS). (SIFS) are used between very small data/ control frames however, for frames having larger chunks of data (LIFS) are recommended. Fig. 2 expresses this concept in terms of IEEE 802.15.4 specifically.

3.2. PHYSICAL LAYER

According to standard specifications, IEEE 802.15.4 can use any of three assigned frequency bands i.e. 868MHz, 915MHz and 2.4GHz. Each offering 20Kbps, 40Kbps and 250Kbps respectively. Within these bands, there are 27 channels. Channel 0 is specified for 868MHz band, channels ranging 1-10 are reserved for 915MHz band while 10 to 26 are nominated for 2.4GHz band.

In 868MHz and 915MHz bands, BPSK is used as modulation scheme where as in 2.4GHz band, O - QPSK is used along with symbol rates of 20KSymbols/sec, 40KSymbols/sec and 62.5KSymbols/sec respectively. Slotted CSMA/CA is used in Contention Access Period (CAP) which is essence of this study.

IEEE 802.15.4 specifies, certain modes are described for physical layer that should be capable of channel assessment. Any one of these modes can be implemented in network to get desired results. Mode1 states that physical layer reports channel as busy if it finds energy level above a certain threshold. Considering Mode2 if physical layer finds a signal with same modulation / spreading scheme, channel will be reported as busy. In Mode3, physical layer not only sense the energy level but also matches the modulation scheme. If that scheme matches, it declares channel busy.[1]. Considering data transmission and reception, data reception is given priority. A node that is receiving data, will attempt for channel accessing only after it has no more packet to receive. In this work, we focus on IEEE 802.15.4

Table 2. Frequency Assignments for IEEE 802.15.4

Frequency Assignment Number Channels of Channel Bandwidth Symbol Rate Data Rate Modulation Scheme

868-868.6 MHz 902-928 MHz 2.4-2.4835 GHz 1 10 16 600KHz 2MHz 5MHz 20KSymbols/Sec 40KSymbols/Sec 62.5KSymbols/Se 20Kbits/sec 40Kbits/sec c250Kbits/sec BPSK BPSK O-QPSK

by comparing various parameters using three prescribed frequency ranges in standard. Moreover, we implement both flavors of CSMA/CA for IEEE 802.15.4 i.e. acknowledged and non-acknowledged with all prescribed frequency bands. We use Markov model presented by Park et.al [10] for slotted 802.15.4 and modified it for non - ACK mode to simulate (using MATLAB) and validate our discussions. The values are set as in accordance with standard text.

4. IEEE 802.15.4 Operational Modes

Default Counter Default Counter Counter Counter

NB=NB+1 NB=NB+1 BE=Backoff Exponent

RT=Retransmission NB=Number of Backoffs

Fig. 3. Comparison, ACK and Non-ACK mode of CSMA/CA - a Node's Prospective

According to specifications, CSMA/CA is operable in non-ACK and ACK modes (fig. 3). If we look closely on effect of this tiny packet over network performance we will be astonished. As discussed in [4] probability of getting CCA1 busy is dependent on two main features, i.e. data packet and ACK packet. Taking the logical reasoning further ahead, we can conclude that, ACK packet is one reason of not finding free channel during clear channel assessment 1 or absence of this packet may raise chances of getting CCA1 successful. Considering impact of CCA1 over network is enormous. Once CCA1 is successfully done, there is a high chance of getting CCA2 free as well. Now, if we look into details of this algorithm, we can say that, having high chances of CCA1 gives higher probability of gainings access of channel, afterwards lower chances of collision resulting in minimizing probability of going to back off stages, and retrying retransmission attempts. Hence, we achieve better good put, lower delay and increased reliability. Numerous authors have placed their analytical and experimental work considering network performance by taking different parameters. This study focus on both flavors of CSMA/CA (with and without ACK) over prescribed three different frequency ranges. For that we taking further ahead the work of [10] and [3], modified gernalized markov chain for non-ACK mode and simulate it using MAT LAB. Reliability, Delay and Throughput are taken as performance metrics for both modes over all three different frequency bands for a network comprising of 10 nodes.

5. Analysis and Results

5.1. Channel Accessing

Once MAC layer receives a packet to transmit, according to protocol standards, it will go on for a random back off period. On completing of this period, it will start sensing transmission medium so that packet can be transmitted. As discussed earlier and in standard that a node will carry forward two consecutive clear channel assessments to transmit. 1st CCA plays a vital role in channel accessing. If CCA1 is acquired, probability of getting CCA2 will increase. In non-acknowledged mode of CSMA/CA, probability of getting CCA1 increases. Here we simulate the said protocol on three different frequency ranges i.e. 868MHz, 915MHz and 2.4GHz to compare which range suits best for IEEE 802.15.4. The impact of ACK packet at transmission range of 858MHz is highest in comparison with rest of two higher ranges. The reason is simple, as range of communication increases, the less will be time taken by load/ traffic to pass on. Considering Traffic load of 1000 bits per application/node, at 858MHZ, probability of getting CCA1 busy is highest and it gradually increase with respect to increase in traffic load. Same pattern can be observed with the rest of two higher frequency ranges however, probability of getting CCA1 busy is minimized. Comparing

Offered Load (bits/appli

Fig. 4. (a)Busy Channel Probability during CCA1, (b) Busy Channel Probability during CCA2, (c) Discarded Packets Probability due to Channel Access Failure, (d) Discarded Packets Probability due to Retry Limits

results of CCA1 and CCA2, Fig 4a and Fig 4b, we come to know an interesting fact that, probability of getting CCA1 busy is higher than probability of getting CCA2 busy. Logically it should be like that because in CCA2, there is minimal almost negligible chance of getting CCA2 busy due to ACK packet. Length of data packet with respect to ACK packet is longer, hence if CCA1 is done successfully, there is a greater chance of getting CCA2 successfully as well. Considering frequency ranges, results are same, higher the range is, better is the performance. There is a procedure described in protocol as, a node will try to get channel for a finite number of times. When that number is over, packet is discarded. According to simulated results, as probability of getting CCA1 and CCA2 successfully is higher at 2.4GHz. This is also depicted when we simulate probability of discarded packets due to failure in channel accession. Fig 4c illustrates probability of packet drop ratio is maximum in 858MHz, this ratio declines at 915MHz and minimal at 2.4GHz.

5.2. Packet Transmission

As a node gets two CCAs successfully, node immediately transmits packet. Although, the algorithm almost vanishes the chances of collision however, there still are chances of collisions. As in accordance with protocol specifications, if collision occurs, node will try to resend it again for a finite number of iterations. Once that counter reaches its

limit, packet than is discarded for further processing or forwarding. Considering plots as in Fig 4d, there is a different behavior between ACK and non-ACK mode of protocol. Here, ACK mode of CSMA/CA supersedes non-ACK mode in higher frequencies however, both stands same at 858MHz band. At this stage, either a successful transmission happens or packet is to be discarded. Either way, protocol running on each node refresh itself and same procedure is repeated.

6. Performance Metrics

6.1. Reliability

Probability of Successful transmission inversely depends upon'packet drop due to channel access failure and exceeding retry limits. Reliability is considered as number of successful transmissions. In above sections we simulated the probabilities of packet drop. To calculate reliability we have to add these packet drop probabilities and plot them inversely. Considering the simulated result, we come to know that for 915MHz band, initially reliability is very high almost near to 1 but then it decrease dramatically. Reliability for 858MHz band declines in same fashion however, it is less reliable with respect to rest two. As the traffic load increases, reliability decreases gradually for all frequency bands however, for 2.4GHz band, the degree of declination is not that steep showing much better performance for IEEE 802.15.4 (slotted non-beacon enabled CSMA/CA). If we compare the modes non-ACK mode supersedes ACK mode as it bears minimal control overhead along with higher probability of accessing a channel for transmission. Fig 5a expresses the probabilistic curve of reliability as discussed earlier.

(a) (b)

Fig. 5. (a) Reliability Factor (Probability) Vs Traffic Load,(b) Delay(Probability) Vs Traffic Load, (c)Throughput(Probability) Vs Traffic Load 6.2. Delay

Delay is one of the most critical performance criteria of any protocol. At times, for specific applications, delay is not tolerated however, In this study we compare this performance metric for different variations of IEEE 802.15.4 standard. Basically, delay comprises of two parts. One is delay due to protocol algorithm that includes back offs and collisions etc., while the second part is based upon length of packet to be transmitted, turnaround time, and inter frame spaces. Simulations show that, delay is lower at 2.4GHz band as back off slot duration is minimal there with respect to rest of two frequency bands. Considering non-ACK mode, here again it has upper hand due to its simplicity and almost negligible control overhead as shown in fig.5b.

6.3. Throughput

Throughput is another very important performance metric for any network. In under consideration flavors of IEEE 802.15.4,as can be depicted from Fig. 5c, non-ACK mode in 2.4GHz band proves itself best amongst rest of all flavors. Reason is very obvious that at greater spectrum, higher will be throughput. non-ACK mode performs better as there is no control overhead and have high chances of getting clear channel with respect to ACK mode. 915MHz band stands second however, in this band also, non-ACK mode performs better. However, there is left no big difference between non-ACK and ACK mode at 858MHz band that has lowest throughput amongst all.

7. Conclusion

In this study, we focused on contention access period of IEEE 802.15.4. ACK mode and non-ACK mode as prescribed in standard text are analyzed analytically and than simulated to verify impact of ACK packet on algorithm using all three frequency assignments. Our study suggests that for critical health monitoring applications where sensors may or may not be implanted in body, ACK mode show its worth due to low packet drop ration. However, for streaming data or applications that do not need ACK of every transmitted packet, non-ACK mode performs better due to its higher throughput, good put and lower delay, control load leading to longer battery life. Using higher frequency assignments may apparently lead to excessive power usage however, higher the frequency is, lower is the packet drop ratio, delay and easy channel access.

References

1. IEEE TG 15.4, 'Part 15.4:Wireless Medium Access Control (MAC) and Physical Layer(PHY) Specifcations for Low-Rate Wireless Personal Area Networks (WPANs)',IEEE Std., New York, 2011.

2. Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications,IEEE Std. 802.11 Specification, Jun. 26, 1997.

3. S. Pollin, M. Ergen, S. C. Ergen, B. Bougard, F. Catthoor, A. Bahai, andP. Varaiya, Performance analysis of slotted carrier sense IEEE 802.15.4 acknowledged uplink transmissions, in Proc. of IEEE WCNC, 2008, pp.15591564

4. T. R. Park, T. H. Kim, J. Y. Choi, S. Choi, and W. H. Kwon, Throughput and energy consumption analysis of IEEE 802.15.4 slotted CSMA/CA, Electron. Lett., vol. 41, no. 18, pp. 10171019, Sep. 2005.

5. Z. Tao, S. Panwar, D. Gu, and J. Zhang, Performance analysis and a proposed improvement for the IEEE 802.15.4 contention access period, inProc. IEEE WCNC, Las Vegas, NV, Apr. 36, 2006, pp. 18111818.

6. S. Pollin, M. Ergen, S. C. Ergen, B. Bougard, L. V. Perre, F. Catthoor, I. Moerman, A. Bahai, and P. Varaiya, Performance analysis of slotted carrier sense IEEE 802.15.4 medium access layer, in Proc. IEEE GLOBECOM, San Francisco, CA, Nov. 27Dec. 1, 2006, pp. 16.

7. T. Lee, H. R. Lee, and M. Y. Chung, MAC throughput limit analysis of slotted CSMA/CA in IEEE 802.15.4 WPAN,IEEE Commun. Lett., vol. 10, no. 7, pp. 561563, Jul. 2006.

8. I. Ramachandran, A. K. Das, and S. Roy, Analysis of Contention Access Period of IEEE 802.15.4, Univ. Washington, Seattle, WA, UWEE Tech. Rep. UWEETR-2006-0003, Feb. 2006.

9. J. Mi si candV.B.Mi si c, Access delay for nodes with finite buffers in IEEE 802.15.4 beacon enabled PAN with uplink transmissions,Comput. Commun., vol. 28, no. 10, pp. 11521166, Jun. 2005

10. Pangun Park;Di Marco, P.;Soldati, P.;Fischione, C.;Johansson, K.H., "A generalized Markov chain model for effective analysis of slotted IEEE 802.15.4," Mobile Adhoc and Sensor Systems, 2009. MASS '09. IEEE 6th International Conference on , vol., no., pp.130,139, 12-15 Oct. 2009

11. Zayani, M.-H.;Gauthier, V.;Zeghlache, D., "A joint model for IEEE 802.15.4 physical and medium access control layers," Wireless Communications and Mobile Computing Conference (IWCMC), 2011 7th International, vol., no., pp.814,819, 4-8 July 2011

12. Wang, F.;Li, D.;Zhao, Y., "Analysis of CSMA/CA in IEEE 802.15.4," Communications, IET , vol.5, no.15, pp.2187,2195, October 14 2011

13. Alvi, A.N.;Naqvi, S.S.;Bouk, S.H.;Javaid, N.;Qasim, U.;Khan, Z.A., "Evaluation of Slotted CSMA/CA of IEEE 802.15.4," Broadband, Wireless Computing, Communication and Applications (BWCCA), 2012 Seventh International Conference on , vol., no., pp.391,396, 12-14 Nov. 2012

14. Israr, I.; Yaqoob, M.M.; Javaid, N.; Qasim, U.; Khan, Z.A., Simulation Analysis of Medium Access Techniques Broadband, Wireless Computing, Communication and Applications (BWCCA), 2012 Seventh International Conference on , vol., no., pp.602,607, 12-14 Nov. 2012.

15. Rahim, A.;Javaid, N.;Aslam, M.;Rahman, Z.;Qasim, U.;Khan, Z.A., A Comprehensive Survey of MAC Protocols for Wireless Body Area Networks, Broadband, Wireless Computing, Communication and Applications (BWCCA), 2012 Seventh International Conference on , vol., no., pp.434,439, 12-14 Nov. 2012. doi: 10.1109/BWCCA.2012.77.

16. Hayat, S.;Javaid, N.;Khan, Z. A.;Shareef, A.;Mahmood, A.;Bouk, S.H., Energy Efficient MAC Protocols, High Performance Computing and Communication, 2012 IEEE 9th International Conference on Embedded Software and Systems (HPCC-ICESS), 2012 IEEE 14th International Conference on , vol., no., pp.1185,1192, 25-27 June 2012.