دانلود رایگان مقاله TCPJGNC: پروتکل کنترل حمل و نقل برای شبکه رادیویی شناختی مولتی هاپ

Abstract

Cognitive radio (CR) has emerged as a promising solution to enhance spectrum utilization. In cognitive radio networks (CRNs), the secondary users (SUs) can opportunistically exploit frequency bands when the primary users (PUs) do not occupy the bands. However, TCP performance in CRNs may suffer from significant degradation due to this feature. In this paper, we investigate the limitations of TCP in multi-channel multi-radio multi-hop CRNs, and propose a novel transmission control protocol called TCPJGNC (TCP Joint Generation Network Coding, JGNC) based on network coding. In TCPJGNC, we dynamically adjust the number of packets involved in network coding according to the wireless communication environment to achieve better decoding probability. In the meantime, a coding scheme based on JGNC is provided which can reduce the number of retransmissions in TCPJGNC. In addition, we modify the TCP mechanism to fit into CRNs by considering the features of CRNs. An analysis of approximate expected throughput in TCPJGNC is provided and the simulation results indicate that TCPJGNC can significantly improve the network performance in terms of throughput, bandwidth efficiency and average end-to-end delay. To the best of our knowledge, TCPJGNC is the first transmission control protocol for multi-hop CRNs from a network coding perspective.

7. Conclusion

In this paper, we propose a novel transport protocol TCPJGNC for multi-hop CRNs from a network coding perspective. We jointly consider network coding and modifications in TCP mechanism for CRNs to improve the TCP performance. We first analyze the decoding probability of JGNC over multi-hop CRNs, which considers packet loss rate, PU activity, and the probability of linearly independent. Then, we incorporate JGNC into TCP and modify the TCP from the following four aspects: PU behavior, spectrum sensing, spectrum changing, and TCP itself, for multi-hop CRNs. Simulation results were presented to illustrate the effectiveness of TCPJGNC. It is shown that TCP throughput, bandwidth utilization efficiency and average end-to end delay can be improved significantly by modifying traditional TCP mechanism and using network coding technology. Network layer and MAC layer will be considered for TCP performance optimization of CRNs in our future work.