VIEWS: 40 PAGES: 5 CATEGORY: Emerging Technologies POSTED ON: 9/29/2012
International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) is an online Journal in English published bimonthly for scientists, Engineers and Research Scholars involved in computer science, Information Technology and its applications to publish high quality and refereed papers. Papers reporting original research and innovative applications from all parts of the world are welcome. Papers for publication in the IJETTCS are selected through rigid peer review to ensure originality, timeliness, relevance and readability. The aim of IJETTCS is to publish peer reviewed research and review articles in rapidly developing field of computer science engineering and technology. This journal is an online journal having full access to the research and review paper. The journal also seeks clearly written survey and review articles from experts in the field, to promote intuitive understanding of the state-of-the-art and application trends. The journal aims to cover the latest outstanding developments in the field of Computer Science and engineering Technology.
International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) Web Site: www.ijettcs.org Email: email@example.com, firstname.lastname@example.org Volume 1, Issue 2, July – August 2012 ISSN 2278-6856 Analyzing the effect of timer timeout and packet length on the performance of error control protocols Shashi Kumar D1, Nagamani A2 1 Christ University, Faculty of Engineering, Kaniminike, Bangalore-56, India 2 Dept. of Electronics and Communication Engineering, Bhaskar Engineering College, Hyderabad, India Abstract: In the present day computer networks quality of consider the Stop-And-Wait protocol because it is known service is a major issue, which invariably depends on the to perform poorly . network performance under various network constraints such In both cases, an error is detected at the sender by either a as flow control, error control, bandwidth, data traffic, buffer timer interrupt or a NACK from the receiver. At this handling etc. In this paper we are Analyzing the Effect of point, if the sender backs up to the first packet in error, Timer timeout and Packet length on the performance of error i.e., the first packet that is not yet ACKed, and restarts control protocols and conclusions are drawn the transmission, the strategy is referred to as Go-Back-N Keywords: Quality of Service, Go-Back-N ARQ, Selective . Repeat ARQ, Timer Timeout, Packet Length, Throughput 1. INTRODUCTION In computer networks, mostly we come across three types of errors. Lost frames Lost Acknowledgements Delayed Acknowledgements Most Error Control techniques are based on either Error Detection Scheme (e.g., Parity checks, CRC), or Retransmission Scheme. Error control schemes that Figure 1: Go-Back-N ARQ with lost frame  involve error detection and retransmission of lost or corrupted frames are referred to as Automatic Repeat If, on the other hand, the sender retransmits only those Request (ARQ) error control. packets which are in error, the strategy is called selective- Error control protocols, as the name implies, these repeat. Fig. 1, Fig. 2.and Fig. 3 shows the flow diagrams protocols are used to recover from errors. When some of Go-Back-N and Selective Repeat user submits a message to the network to be transmitted ARQ’s. to another user, the message is usually split into packets which are transmitted over the network and reassembled at the other end. A packet in transit encounters one or more channels, (e.g., satellite, copper wire or optical fiber), and nodes (or routers), which route the packet to the destination. These intermediate elements can induce errors in the packet in that either the packet could get garbled, or dropped altogether. The former is usually due to random electrical noise in the channels while the latter is due to buffer overruns at the nodes and is caused by contention for resources, a phenomenon often referred to as congestion. Two error control protocols that we are primarily interested in analysis are the Go-Back-N and the Selective Repeat protocols. We shall however not Figure 2 Go-Back-N ARQ with lost Acknowledgement  Volume 1, Issue 2 July-August 2012 Page 190 International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) Web Site: www.ijettcs.org Email: email@example.com, firstname.lastname@example.org Volume 1, Issue 2, July – August 2012 ISSN 2278-6856 (Timer Timeout vs. Throughput) Tout SL/No Ps X (ms) 0 0 1 0 1250 0.0367 2 11 1500 0.3267 3 98 1750 0.5367 4 161 Figure 3: Selective Repeat ARQ with lost frame  2000 0.6733 5 202 2. SIMULATIONS AND RESULTS 2250 0.6767 6 203 The analysis on the performance of error control 2500 0.68 protocols, such as Go-Back-N ARQ and Selective Repeat 7 204 ARQ is done by conducting experiments using 2750 0.68 BENCHMARK LAN TRAINER KIT and the simulation 8 204 software. 3000 0.6867 9 206 2.1 Setting the configuration menu In both, the sender and receiver machines the following values needed to be entered in the configuration windows. After calculating the timer timeout values for various data rates experiments are conducted for constant experimental duration. Then the values of successful transmission are noted for different timer timeout values and packet lengths. TABLE 1: Configuration windows at sender and receiver PC 1 PC 2 Node Id 0 Node Id 0 Figure 4: Timer Timeout Vs. Throughput Protocol CSMA/C Protocol CSMA/C D D Baud rate 8Kbps (at Baud Rate 8Kbps (at TABLE 3: Results for Go-Back-N ARQ both the both the (Packet length Vs. Throughput) config config menu and menu and NEU) NEU) Pl in SL/No Ps X Duration 100s Duration 100s Bytes Packet 1000 bytes Packet 1000 bytes length Length 1 0 0 0 Inter 400 ms Inter 400 ms Packet Packet 2 500 314 0.5233 Delay Delay Direction Sender Direction Receiver 3 750 250 0.6250 A. Go-Back-N ARQ 4 1000 199 0.6633 TABLE 2: Results for Go-Back-N ARQ 5 1250 0 0 Volume 1, Issue 2 July-August 2012 Page 191 International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) Web Site: www.ijettcs.org Email: email@example.com, firstname.lastname@example.org Volume 1, Issue 2, July – August 2012 ISSN 2278-6856 Figure 5: Packet length Vs. Throughput B. Selective Repeat ARQ Figure 6: Timer Timeout Vs. Throughput TABLE 4: Results for Selective Repeat ARQ (Timer Timeout Vs. Throughput) TABLE 5: Results for Selective Repeat ARQ (Packet length Vs. Throughput) SL/No Tout (ms) Ps X SL/No Pl in Ps X Bytes 1 0 0 0 1 0 0 0 2 1250 88 0.2933 2 500 298 0.4967 3 1500 128 0.4267 3 750 229 0.5725 4 1750 187 0.6233 5 2000 187 0.6233 4 1000 191 0.6367 6 2250 204 0.68 5 1250 0 0 7 2500 199 0.6732 8 2750 195 0.65 9 3000 193 0.6433 Figure 7: Packet length Vs. Throughput C. Comparison of Go-Back-N ARQ & Selective Repeat ARQ Volume 1, Issue 2 July-August 2012 Page 192 International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) Web Site: www.ijettcs.org Email: email@example.com, firstname.lastname@example.org Volume 1, Issue 2, July – August 2012 ISSN 2278-6856 with Go-Back-N protocol for a given packet length. In the other hand Go-Back-N ARQ will give better throughput for lower values of Packet length than compared to the Selective Repeat ARQ. 4. FORMULA USED Formula used to calculate throughput is. Figure 8: Timer Timeout Vs. Throughput Where X = Throughput Ps =Successfully Transmitted packets Pl= Packet Length in Bytes De=Duration of Experiment in seconds Dr=Data rate in Kbps Tout= Timer Timeout value in seconds 5. CONCLUSION AND FUTURE WORK In this paper we made an analysis of the effect of timer timeout values and packet length on throughput of data link layer error control protocols, conducting experiments using BENCH MARK LAN TRAINER KIT and simulation software without considering the bit errors. Further enhancement can be made to analyze the effect of timer timeout and packet length on the throughput of the Figure 9: Packet length Vs. Throughput system considering bit errors. Also we can analyze the effect of data rate on throughput. 3. RESULT ANALYSIS The simulation results are obtained for Go-Back-N ARQ References and Selective Repeat ARQ by varying packet length and Timer timeout values, using BENCHMARK LAN  Zwaenepoel, W., “Protocols for Large Data Transfers TRAINER KIT and the simulation software. on Local Networks”, Proc. of the ACM Sigcomm, pp Fig. 4 and Fig. 6 show the plots of Timer timeout Vs. 22-32, 1985 Throughput for Go-Back-N ARQ and Selective Repeat  Tanenbaum, A.S., Computer Networks, Prentice- ARQ. In this experiment, the value of packet length, Hall Inc.,Englewood Cliffs, NJ 07632, 1981 1000Bytes and duration of the experiment, 300 seconds is  S. Lin, D. J. Costello, Jr., and M. J. Miller, taken for the simulation. From plots it is clear that as we “Automatic-repeat-requesterror-control scheme,” increase the timer timeout, the throughput of the system IEEE Commun., vol. 22, pp. 5-17, Dec. 1984 is increasing up to certain level and then it is almost same  M. Moeneclaey, H. Bruneel, I. Bruyland, and D. Y. for all other values. For the current scenario, the Chung, “Throughput optimization for a generalized throughput is increasing up to the timeout value of 2250; stop-and-wait ARQ scheme, ”IEEE Trans.Commun., thereafter it is almost equal for all other values. vol. COM-34, pp. 205-207, Feb. 1986 Fig. 5 and Fig. 7 show the plots of Packet length Vs.  Anagnostou, M.E. and E.N. Protonotarios, Throughput for Go-Back-N ARQ and Selective Repeat "Performance analysis of the selective repeat ARQ ARQ. From plots it is very clear that the throughput of protocol," IEEE Trans.Commun., vol. COM-34, pp. the system will increase up to certain range of packet 127-135, Feb. 1986. lengths then it returns back to zero and continuous to be AUTHOR in zero for further increase in the packet length. Shashi Kumar D received the B.E degree Fig. 8 and Fig.9 gives us the comparison results of Go- from SJC Institute of Technology in 2007 and Back-N ARQ and Selective Repeat ARQ protocols for MTech degree in Electrical Engineering from Timer timeout Vs. Throughput and Packet length Vs. Indian Institute of Technology Madras in Throughput respectively. From these plots we can infer 2010. During June, 2010- July, 2011, he that for lesser timer timeout values we will get more stayed in ICFAI University Dehradun as a throughput in case of Selective Repeat when compared Volume 1, Issue 2 July-August 2012 Page 193 International Journal of Emerging Trends & Technology in Computer Science (IJETTCS) Web Site: www.ijettcs.org Email: email@example.com, firstname.lastname@example.org Volume 1, Issue 2, July – August 2012 ISSN 2278-6856 Faculty member, from July-2011 to till date working as Assistant Professor in Christ University, Faculty of Engineering. Nagamani A received the B.Tech from Bhojreddy College of Engineering & Technology for women in 2004, MTech degree in Electrical Engineering from Indian Institute of Technology Madras in 2010, From August 2010 to till date working as Assistant Professor in Bhaskar Engineering College (JB group of institutions). Volume 1, Issue 2 July-August 2012 Page 194
Pages to are hidden for
"Analyzing the effect of timer timeout and packet length on the performance of error control protocols"Please download to view full document