Performance Enhancement of TCP's Westwood Implementation with Improved Flow for Wired and Wireless Scenarios

Performance of Transmission Control Protocol (TCP) was analysed in presence of random drops and congestion. the analysis proved TCP WestwoodNR (TCP-WNR) was best to utilise the available bandwidth. TCP WestwoodNR protocol controls the window using end-to-end bandwidth estimation. This Bandwidth Estimation continuously estimates bandwidth at the TCP sender based on the packet rate of the connection by monitoring the ACK reception rate. The estimated connection rate is then used to compute congestion window and slow start threshold settings after a packet loss occurred. Resetting the window to match available bandwidth makes TCP-WNR more robust to Random loss as well as in Congestion. The conventional TCP over reacts to losses and leading to unnecessary window reduction. Experimental studies of TCP-WNR show significant improvements in throughput performance over Reno and SACK, particularly in wired networks. The traditional TCP has many drawbacks like unable to distinguish the losses and very much conservative to the initial start. In case of Time Out also it is very much conservative; the filter implementation was modified from conventional TCP-WNR. So the performance was improved based on this and the modified TCP-WNR was compared to original TCP-WNR. The results show the performance comparison of the two protocols.