Internetworking with TCP/IP (2nd ed.), vol. I
Internetworking with TCP/IP (2nd ed.), vol. I
Radio link protocols for cellular data
Wireless information networks
Performance evaluation of TCP/RLP protocol stack over CDMA wireless link
Wireless Networks - Special issue on wireless multimedia networking
Modeling TCP throughput: a simple model and its empirical validation
Proceedings of the ACM SIGCOMM '98 conference on Applications, technologies, architectures, and protocols for computer communication
TCP source activity and its impact on call admission control in CDMA voice/data network
MobiCom '98 Proceedings of the 4th annual ACM/IEEE international conference on Mobile computing and networking
TCP/IP performance over 3G wireless links with rate and delay variation
Proceedings of the 8th annual international conference on Mobile computing and networking
I-TCP: indirect TCP for mobile hosts
ICDCS '95 Proceedings of the 15th International Conference on Distributed Computing Systems
Performance Study of Link Layer and MAC Layer Protocols to Support TCP in 3G CDMA Systems
IEEE Transactions on Mobile Computing
Link adaptation and power control for streaming services in EGPRS wireless networks
IEEE Journal on Selected Areas in Communications
Improving the performance of reliable transport protocols in mobile computing environments
IEEE Journal on Selected Areas in Communications
Analytical Model of TCP NewReno through a CTMC
EPEW '09 Proceedings of the 6th European Performance Engineering Workshop on Computer Performance Engineering
Hi-index | 14.99 |
To bring TCP-based services to the mobile devices in a cellular network, it is necessary that TCP be extended over the wireless link. However, the performance of TCP severely degrades in a wireless medium. Hence, radio link protocols (RLPs) are used as an interface between TCP and the physical medium. RLPs fragment TCP segments into frames and use robust error correcting codes and fast retransmission schemes to shield the channel related losses from TCP, thus preventing TCP throughput degradation. In this paper, we show the limitations of the existing RLPs, which do not differentiate the frames generated from the same TCP segment. We claim that if selective frames are made more robust to transmission failures, then the performance of RLP and, hence, TCP can be improved. We identify such decisive frames and categorize them as crucial and noncrucial. Our claim is based on the fact that initial frames can afford a few trials of retransmissions, whereas the later ones cannot. We treat the frames differentially with respect to forward error correcting (FEC) coding and automatic repeat request (ARQ) schemes. We consider specific cases of FEC and ARQ strategies and show the qualitative difference in the performance of the RLP through analysis and simulations. The gain in the performance is more prominent when both FEC and ARQ (hybrid-ARQ) are used. The increase in TCP throughput with the proposed RLP is also demonstrated.