Journal de génie informatique et des technologies de l'information

The Effect of Changing the Weight Parameter Boundaries In Random Early Detection Method

Mohsen Hosamo

Random Early Detection method (RED) uses a single linear drop function to calculate the probability of marking the packet ( pb) and uses the lower queue threshold ( QL), Higher queue threshold (QH), the maximum probability of dropping the packets at QH ( Maxdrop), the low-pass filter time constant (wQ ), and the Average size of the queue (Qavg) to adjust the queue performance. The two key components of the RED algorithm are the evaluation of the Qavg and the pb, that leads to the success of RED method. In this paper we are showing the effect of changing the low-pass filter time constant (wQ) in the dropping method of RED.