Improving TCP performance in wired-wireless networks by using a novel adaptive bandwidth estimation mechanism

Abstract

This paper presents a novel dynamic bandwidth estimation mechanism for improving TCP (Transmission Control Protocol) performance in wired-cum-wireless networks. The key idea here is to continuously measure the bandwidth used by a TCP flow via monitoring the rate of returning acknowledgements (ACKs) and the round-trip time (RTT) values. The distinguishing feature of this mechanism (compared to other mechnaisms such as that in TCP Westwood) is that it exploits the burstiness pattern of ACK arrivals and estimates the available bandwidth more accurately. In the proposed bandwidth estimation mechanism, the bandwidth sample is calculated by distributing a burst of ACKs over an off period based on degree of congestion and burstiness in the network. The estimation technique is robust against burstiness of ACK arrival and type of loss (e.g., wireless loss, congestion loss). Simulation results obtained using ns-2 reveal that, a new variant of TCP New-Reno based on this adaptive bandwidth estimation technique, which is referred to as TCP Prairie, provides significant throughput performance improvement over TCP New-Reno and TCP Westwood under congestion and/or wireless loss scenarios. Also, compared to TCP Westwood, TCP Prairie is observed to be more friendly towards TCP New-Reno.

WLC47-6: TCP Performance Improvement through Inter-layer Enhancement with Mobile IPv6

“…Therefore, we need a way to optimize CWND/SSTHRESH values by quickly probing the route's available capacity. The optimal congestion window size can be determined based on the bandwidth estimation [16]. In addition, updating the congestion window is based solely on ACKs for data sent through the new network.…”

Receiver-Based ACK Splitting Mechanism for TCP over Wired/Wireless Heterogeneous Networks

“…The past literature proposes many solutions to solve performance degradation problem of TCP over networks including wireless links [6]- [18]. Here, we focus on the solutions with modification of the sender-side TCP, since the proposed mechanism is this paper is also an end-to-end approach.…”

“…In [12]- [18], the sender-side TCP algorithm is modi-fied so that it does not unnecessarily reduce the congestion window size. These approaches modify the congestion control mechanism of the sender TCP so that it will not shrink the congestion window in response to wireless losses.…”

“…This results in a situation in which the higher the bit error rate of the wireless link is, the more often wireless losses occur, and TCP's data transmission performance deteriorates more significantly. Many solutions to this problem have been proposed in the past literature [6]- [18]. Some of them [6]- [11] modify the functions of the base station, which is the border node of the wired and wireless networks.…”

“…Furthermore, they can not be applied when a lower-layer encryption mechanism such as IPSec [19] is utilized because they need to access TCP header at the base station. Other solutions [12]- [18] modify the sender-side TCP algorithm. They try not to slow down TCP's transmission rate when a wireless loss is detected.…”