Multi-Layer Protocol Tracing in a GPRS Network. Gurtov, A., Passoja, M., Aalto, O., & Raitola, M. Proceedings of the IEEE Vehicular Technology Conference (Fall VTC2002), Vancouver, Canada, September, 2002.
Paper bibtex @article{ Gurtov02a,
author = {A. Gurtov and M. Passoja and O. Aalto and M. Raitola},
title = {Multi-Layer Protocol Tracing in a {GPRS} Network},
journal = {Proceedings of the IEEE Vehicular Technology Conference (Fall VTC2002), Vancouver, Canada},
year = {2002},
month = {September},
annote = {Abstract: This paper presents a performance evaluation of GPRS accomplished by combination of measurement at the end hosts and tracing inside the network. The multi-layer tracing approach allows not only observing, but also understanding the network performance. With end-to-end measurements we assess data rates, latency, and buffering experienced by users in a live GPRS network. Comparing the results to our previous measurements shows a notable improvement in the network and terminals over past two years. Mobility tests while driving in the urban environment quantify the interval, duration and data loss caused by cell reselections. In the test lab, multi-layer tracing of radio, link and transport protocols gives a closer picture of GPRS performance. For instance, TCP interacts inefficiently with resource allocation at the RLC layer and fragmentation at the LLC layer. Finally, we illustrate delay spikes and data losses during a cell reselection by tracing of signaling messages during a cell update and routing area update procedures. Annika: The performance of data transmission in Sonera's live and test GPRS networks is investigated in this paper. The GPRS implementation is based on release 97. The coding scheme used was CS-2. The network was configured for LLC in unacknowledged mode and RLC in acknowledged mode. VJ header compression was disabled. Throughput, latency, buffering, and mobility (cell reselections) were investigated. Both TCP and RLC data was captured (with Tcpdump and Nethawk) and analyzed. The maximum throughput was 43kbps in DL and 21kbps in UL. The round trip time in an unloaded network was 0.5-1.1s (with a typical value of 0.7s). Many interesting results are presented, and only some of them are included here. The buffers in the GPRS network are too large (50kbps per user in DL), cell reselections lead to delay spikes at the TCP level that can cause spurious timeouts. Increases in round trip times occurred due to frequent TBF allocations and releases at the RLC layer. The RLC acknowledgment scheme wasted resources in some cases. The authors also found some performance degrading bugs in the GPRS networks.},
url = {papers/Gurtov02_multilayer_TCP_GPRS.pdf},
bibdate = {Sunday, July 07, 2002 at 06:54:44 (CEST)},
submitter = {Johan Garcia}
}
Downloads: 0
{"_id":"wWnYLonW2WPRJuPGw","bibbaseid":"gurtov-passoja-aalto-raitola-multilayerprotocoltracinginagprsnetwork-2002","downloads":0,"creationDate":"2015-12-04T23:32:58.190Z","title":"Multi-Layer Protocol Tracing in a GPRS Network","author_short":["Gurtov, A.","Passoja, M.","Aalto, O.","Raitola, M."],"year":2002,"bibtype":"article","biburl":"http://www.cs.kau.se/cs/prtp/prtp.bib","bibdata":{"annote":"Abstract: This paper presents a performance evaluation of GPRS accomplished by combination of measurement at the end hosts and tracing inside the network. The multi-layer tracing approach allows not only observing, but also understanding the network performance. With end-to-end measurements we assess data rates, latency, and buffering experienced by users in a live GPRS network. Comparing the results to our previous measurements shows a notable improvement in the network and terminals over past two years. Mobility tests while driving in the urban environment quantify the interval, duration and data loss caused by cell reselections. In the test lab, multi-layer tracing of radio, link and transport protocols gives a closer picture of GPRS performance. For instance, TCP interacts inefficiently with resource allocation at the RLC layer and fragmentation at the LLC layer. Finally, we illustrate delay spikes and data losses during a cell reselection by tracing of signaling messages during a cell update and routing area update procedures. Annika: The performance of data transmission in Sonera's live and test GPRS networks is investigated in this paper. The GPRS implementation is based on release 97. The coding scheme used was CS-2. The network was configured for LLC in unacknowledged mode and RLC in acknowledged mode. VJ header compression was disabled. Throughput, latency, buffering, and mobility (cell reselections) were investigated. Both TCP and RLC data was captured (with Tcpdump and Nethawk) and analyzed. The maximum throughput was 43kbps in DL and 21kbps in UL. The round trip time in an unloaded network was 0.5-1.1s (with a typical value of 0.7s). Many interesting results are presented, and only some of them are included here. The buffers in the GPRS network are too large (50kbps per user in DL), cell reselections lead to delay spikes at the TCP level that can cause spurious timeouts. Increases in round trip times occurred due to frequent TBF allocations and releases at the RLC layer. The RLC acknowledgment scheme wasted resources in some cases. The authors also found some performance degrading bugs in the GPRS networks.","author":["Gurtov, A.","Passoja, M.","Aalto, O.","Raitola, M."],"author_short":["Gurtov, A.","Passoja, M.","Aalto, O.","Raitola, M."],"bibdate":"Sunday, July 07, 2002 at 06:54:44 (CEST)","bibtex":"@article{ Gurtov02a,\n author = {A. Gurtov and M. Passoja and O. Aalto and M. Raitola},\n title = {Multi-Layer Protocol Tracing in a {GPRS} Network},\n journal = {Proceedings of the IEEE Vehicular Technology Conference (Fall VTC2002), Vancouver, Canada},\n year = {2002},\n month = {September},\n annote = {Abstract: This paper presents a performance evaluation of GPRS accomplished by combination of measurement at the end hosts and tracing inside the network. The multi-layer tracing approach allows not only observing, but also understanding the network performance. With end-to-end measurements we assess data rates, latency, and buffering experienced by users in a live GPRS network. Comparing the results to our previous measurements shows a notable improvement in the network and terminals over past two years. Mobility tests while driving in the urban environment quantify the interval, duration and data loss caused by cell reselections. In the test lab, multi-layer tracing of radio, link and transport protocols gives a closer picture of GPRS performance. For instance, TCP interacts inefficiently with resource allocation at the RLC layer and fragmentation at the LLC layer. Finally, we illustrate delay spikes and data losses during a cell reselection by tracing of signaling messages during a cell update and routing area update procedures. Annika: The performance of data transmission in Sonera's live and test GPRS networks is investigated in this paper. The GPRS implementation is based on release 97. The coding scheme used was CS-2. The network was configured for LLC in unacknowledged mode and RLC in acknowledged mode. VJ header compression was disabled. Throughput, latency, buffering, and mobility (cell reselections) were investigated. Both TCP and RLC data was captured (with Tcpdump and Nethawk) and analyzed. The maximum throughput was 43kbps in DL and 21kbps in UL. The round trip time in an unloaded network was 0.5-1.1s (with a typical value of 0.7s). Many interesting results are presented, and only some of them are included here. The buffers in the GPRS network are too large (50kbps per user in DL), cell reselections lead to delay spikes at the TCP level that can cause spurious timeouts. Increases in round trip times occurred due to frequent TBF allocations and releases at the RLC layer. The RLC acknowledgment scheme wasted resources in some cases. The authors also found some performance degrading bugs in the GPRS networks.},\n url = {papers/Gurtov02_multilayer_TCP_GPRS.pdf},\n bibdate = {Sunday, July 07, 2002 at 06:54:44 (CEST)},\n submitter = {Johan Garcia}\n}","bibtype":"article","id":"Gurtov02a","journal":"Proceedings of the IEEE Vehicular Technology Conference (Fall VTC2002), Vancouver, Canada","key":"Gurtov02a","month":"September","submitter":"Johan Garcia","title":"Multi-Layer Protocol Tracing in a GPRS Network","type":"article","url":"papers/Gurtov02_multilayer_TCP_GPRS.pdf","year":"2002","bibbaseid":"gurtov-passoja-aalto-raitola-multilayerprotocoltracinginagprsnetwork-2002","role":"author","urls":{"Paper":"http://www.cs.kau.se/cs/prtp/papers/Gurtov02_multilayer_TCP_GPRS.pdf"},"downloads":0},"search_terms":["multi","layer","protocol","tracing","gprs","network","gurtov","passoja","aalto","raitola"],"keywords":[],"authorIDs":[],"dataSources":["6WGcSu2Ku7pZzqCcg"]}