Abstract
In this study, we present a new approach for estimating the performance of the TCP network depending on a mathematical formula which converts the conceptual performance parameter to a valuable one. With respect to TCP network performance, we consider the Random Early Detection (RED) and RED-Based algorithms where both RED and RED-Based follow the Active Queue Management (AQM) approach in TCP networks. Mainly, we make a comprehensive understanding of various network parameters that continually affect the network performance. The main parameters that are used given by: network throughput, link utilization, network average delay, and packet loss rate. Also in this study, we estimate a new parameter reflecting the network performance called Network Quality Parameter (NQP).
Highlights
There are some problems affecting the network performance like congestion phenomenon in TCP networks[1]
When a network continues in stable state till the end of operation time, the en parameter equals to the product of parameter t0 which is calculated in Eq 16 by buf which is the buffer size
The buffer size equal to 46 packets: Our study aimed to introduce a new approach for estimating the TCP network performance
Summary
There are some problems affecting the network performance like congestion phenomenon in TCP networks[1]. If RED gateways mark packets by dropping them, recently seen flows, with the following extra rather than by setting a bit in the packet header RED drops packet when congestion occurs using address, destination address, etc.) Added to the list, the the marking probability in Esq. 1 & 2, where Eq 1 count of the zombie set to zero, and its timestamp is set represents the immediately marking probability of the to the arrival time of the packet. An adaptive RED (ARED) mechanism cognizant of the number of active connections can provide significant benefits in terms of decreasing packet loss and increasing network utilization.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
