In Wireless Sensor Networks (WSNs), most of the issues related to management of topology, managing energy, estimation of bandwidth, calculating packet loss etc. are addressed. Several researches have addressed queue management issues, congrolling the congestion that is incorporated with Soft Computing (SC) techniques. Very few works have focused on mitigating or avoding the congestion, delay estimation and the required bandwidth for transmitting data packets in the sensor network. This paper presents an Active Queue Management (AQM) model for estimating congestion using priority Random Early Detection (RED) to mitigate the congestion. The proposed method integrates AQM model with fuzzy subsystem which controls the admission of packets into the network and the same is compared with several existing works and is simulated using Matlab Simulator. The results outperforms the existing methods in terms of packet loss probability, energy consumption, queue length, end-to-end delay and delay between intermediate nodes.