Brief paper: A neuro-adaptive congestion control scheme for round trip regulation
Automatica (Journal of IFAC)
Fairness guarantees in a neural network adaptive congestion control framework
IEEE Transactions on Neural Networks
Optimal flow control for utility-lifetime tradeoff in wireless sensor networks
Computer Networks: The International Journal of Computer and Telecommunications Networking
A multi-channel token ring protocol for QoS provisioning in inter-vehicle communications
IEEE Transactions on Wireless Communications
Hi-index | 0.01 |
We propose a TCP-friendly Additive Increase and Multiplicative Decrease (AIMD) based Datagram Congestion Control Protocol (DCCP) protocol for supporting multimedia traffic in hybrid wireless/wired networks. We further demonstrate how to select the protocol parameters to fairly and efficiently utilize network resources with the consideration of quality of service (QoS) requirements. Since the parameter selection procedure requires only the exchange of parameters among the application, the transport layer protocol, and the link layer protocol, our approach preserves the end-to-end semantics of the transport layer protocol and the layered structure of the Internet. Extensive simulations are performed to evaluate the proposed protocol. It is shown that the AIMD protocol can appropriately regulate multimedia traffic to efficiently utilize the wireless link and fairly share the network resources with coexisting TCP flows, and it can provide satisfactory QoS for delay-sensitive multimedia applications. In addition, AIMD protocol can outperform the non-responsive User Datagram Protocol (UDP) when transporting multimedia traffic over hybrid wireless/wired networks. With satisfactory QoS provisioning, end-systems have more incentives to voluntarily regulate multimedia traffic with an AIMD-based congestion controller, which is vital for network stability, integrity, and future proliferation.