An Exponential Active Queue Management Method Based on Random Early Detection

Abstract

Congestion is a key topic in computer networks that has been studied extensively by scholars due to its direct impact on a network’s performance. One of the extensively investigated congestion control techniques is random early detection (RED). To sustain RED’s performance to obtain the desired results, scholars usually tune the input parameters, especially the maximum packet dropping probability, into specific value(s). Unfortunately, setting up this parameter into these values leads to good, yet biased, performance results. In this paper, the RED-Exponential Technique (RED_E) is proposed to deal with this issue by dropping arriving packets in an exponential manner without utilizing the maximum packet dropping probability. Simulation tests aiming to contrast E_RED with other Active Queue Management (AQM) methods were conducted using different evaluation performance metrics including mean queue length (mql), throughput (T), average queuing delay (D), overflow packet loss probability (PL), and packet dropping probability (DP). The reported results showed that E_RED offered a marginally higher satisfactory performance with reference to mql and D than that found in common AQM methods in cases of heavy congestion. Moreover, RED_E compares well with the considered AQM methods with reference to the above evaluation performance measures using minimum threshold position (min threshold) at a router buffer.

Survey on: A variety of AQM algorithm schemas and intelligent techniques developed for congestion control

“…When the new packet arrives with a size that exceeds the available capacity in a virtual queue, this packet will mark/drop to the actual queue; otherwise, it adds and then updates the virtual queue. The RED-exponential technique "RED-E" [43] was proposed as a modified RED algorithm. The RED-E uses non leaner behavior to drop packets to manage heterogeneous services flow types.…”

Random early detection-quadratic linear: an enhanced active queue management algorithm

“…A lot of AQM algorithms which leverages its simple model continues to evolve in literature [10]. Examples include: gentle RED (GRED) by [15], smart RED (SmRED) by [16], nonlinear RED (NLRED) by [11], SmRED-i by [17], Qlearning-based RED (QRED) by [18], modified RED (MRED) by [19], virtual queue RED (VQ-RED) by [20], hazard estimate red (HERED) by [21], RED-exponential (RED_E) by [22], RED with reconfigurable maximum dropping probability (RRMDP) by [23], delay-controller RED (DcRED) by [24], three-section RED (TRED) by [25], multi RED by [26], adaptive drop-tail by [27], just to mention a few.…”

“…Abu-Shareha [24] developed the delay-controller RED (DcRED) algorithm which calculates a delay parameter for each packet as it arrives the router in order to estimate the dropping probability. In Exponential RED (RED_E) algorithm proposed by Abdel-Jaber [22], the maxP parameter was eliminated and utilizes an exponential drop function when ave varies from minTh to maxTh. The RED_E's drop probability function is given by:  ISSN: 2302-9285…”

Applying reinforcement learning for random early detaction algorithm in adaptive queue management systems

“…The trade-off between queuing delay and throughput is investigated in this study using an AQM (RED) integrated deep reinforcement learning framework for effective network control [13]. Deep Qnetwork (DQN) is used to create our application [14]. The key Q-network and the target network, for example, are both equipped with experience replay.…”