• Journal of Electrical Engineering and Technology
• /
• 제3권2호
• /
• pp.285-291
• /
• 2008

Active Queue Management(AQM) has been widely used for congestion avoidance in Transmission Control Protocol(TCP) networks. Although numerous AQM schemes have been proposed to regulate a queue size close to a reference level, most of them are incapable of adequately adapting to TCP network dynamics due to TCP's non-linearity and time-varying stochastic properties. To alleviate these problems, we introduce an AQM technique based on a dynamic neural network using the Back-Propagation(BP) algorithm. The dynamic neural network is designed to perform as a robust adaptive feedback controller for TCP dynamics after an adequate training period. We evaluate the performances of the proposed neural network AQM approach using simulation experiments. The proposed approach yields superior performance with faster transient time, larger throughput, and higher link utilization compared to two existing schemes: Random Early Detection(RED) and Proportional-Integral(PI)-based AQM. The neural AQM outperformed PI control and RED, especially in transient state and TCP dynamics variation.

• 한국통신학회논문지
• /
• 제29권1B호
• /
• pp.29-40
• /
• 2004

본 논문에서는 TCP 흐름제어 기법을 전제로 하여, 가까운 미래에 발생할 가능성이 있는 인터넷 입구에서의 혼잡을 미리 예측함으로써 TCP 흐름제어의 효과를 배가시킬 수 있는 QR-AQM이라는 버퍼관리 기법을 제안한다. QR-AQM 버퍼관리기법은 라우터로 전송되어 오는 단위시간당 TCP 흐름량을 추정하여 이를 이용함으로써 RED기법과 같이 평균버퍼 수위에만 의존하는 조기혼잡제어기법에 비해 미래에 네트워크에 일어날 혼잡에 보다 민첩하게 대처 할 수 있다. 모의 실험을 통한 성능 검증결과 인터넷 입구에서 OR-AQM 기법은 Adaptive RED 기법과 비교하여 간단한 변수의 설정만으로 버퍼를 목적한 수준으로 유지함이 가능함으로, 다양한 응용프로그램의 요구에 맞추어 직관적이고 쉽게 최적의 버퍼수위를 설정할 수 있다. 또한 QR-AQM 기법은 시간에 따른 버퍼수위의 변화가 목적한 버퍼수위를 기준으로 작은 변동폭을 유지함으로써, 패킷들이 라우터의 버퍼를 지나면서 겪게되는 지연시간의 등락현상을 효과적으로 억제할 수 있음을 확인하였다.

• International Journal of Control, Automation, and Systems
• /
• 제6권4호
• /
• pp.526-533
• /
• 2008

We propose a wavelet neural network (WNN) control method for active queue management (AQM) in an end-to-end TCP network, which is trained by adaptive learning rates (ALRs). In the TCP network, AQM is important to regulate the queue length by passing or dropping the packets at the intermediate routers. RED, PI, and PID algorithms have been used for AQM. But these algorithms show weaknesses in the detection and control of congestion under dynamically changing network situations. In our method, the WNN controller using ALRs is designed to overcome these problems. It adaptively controls the dropping probability of the packets and is trained by gradient-descent algorithm. We apply Lyapunov theorem to verify the stability of the WNN controller using ALRs. Simulations are carried out to demonstrate the effectiveness of the proposed method.

• 한국위성정보통신학회논문지
• /
• 제12권4호
• /
• pp.131-140
• /
• 2017

Multipath TCP(MPTCP)는 다중 인터페이스를 이용하여 데이터를 동시에 전송하는 전송계층의 프로토콜이다. MPTCP는 비슷한 특성의 망을 사용하는 네트워크 환경에서는 기존 TCP에 비해 우월한 성능을 보이지만 이종 망을 사용하는 MPTCP의 다중경로 중 특정 경로에서 버퍼블로트 현상이 일어나게 되면 급격한 지연시간의 증가로 인해 느린 경로로 전송 된 패킷이 MPTCP 수신버퍼에 도착하지 않게 되어 다른 경로로 수신한 패킷이 수신버퍼에서 대기하는 Head-of-Line(HoL) blocking 현상을 초래하여 기존 TCP 대비 더욱 악화된 성능을 보이기도 한다. 따라서 본 논문에서는 이러한 버퍼블로트 문제를 해결하기 위해 Active Queue Management(AQM) 기법 중 Adaptive Random Early Detection(ARED), Controlled Delay(CoDel), Proportional Integral Controller Enhanced(PIE)를 네트워크에 적용하여 이종 망에서 MPTCP 성능 개선을 목표로 실험을 진행하였다. 성능 향상을 살펴보기 위해 Linux 기반의 유선 테스트베드를 구축하고 기존 드롭 테일과 AQM을 비교분석하여 AQM 기반의 TCP가 bufferbloat가 있는 다중경로 MPTCP의 성능을 향상시키는 것을 살펴보았다.

• 한국인터넷방송통신학회논문지
• /
• 제16권1호
• /
• pp.129-137
• /
• 2016

전송계층 프로토콜처럼 전송제어 프로토콜은 안정적인 데이터 전송 서비스를 제공한다. 다양한 네트워크에서 TCP의 성능을 저해하는 일부 심각한 문제가 있다. TCP 네트워크 환경에서 중요한 문제는 빠른 전송 속도로 인해 또는 동시에 네트워크로 접속하는 다수의 새로운 접속으로 인하여 발생하는 혼잡이다. 그러므로 라우터에서 큐의 크기는 패킷 하락에 기인하여 증가한다. 손실된 패킷의 재전송과 감소된 처리량은 많은 비용을 발생시킨다. RED처럼 AQM과 ECN은 패킷 하락 보다는 패킷 마킹에 사용된다. IP 패킷 헤더에서 ECN 비트는 불필요한 패킷 하락을 피하기 위한 혼잡 표시로 추가할 수 있다. 제안하는 ECN과 AQM 메커니즘은 NS2 시뮬레이터의 도움으로 구현할 수 있으며, 그 성능은 다른 TCP 변종에서 테스트할 수 있다.

• Journal of Communications and Networks
• /
• 제8권1호
• /
• pp.93-105
• /
• 2006

The domination of the Internet by TCP-based services has spawned many efforts to provide high network utilization with low loss and delay in a simple and scalable manner. Active queue management (AQM) algorithms attempt to achieve these goals by regulating queues at bottleneck links to provide useful feedback to TCP sources. While many AQM algorithms have been proposed, most suffer from instability, require careful configuration of nonintuitive control parameters, or are not practical because of slow response to dynamic traffic changes. In this paper, we propose a new AQM algorithm, hybrid random early detection (HRED), that combines the more effective elements of recent algorithms with a random early detection (RED) core. HRED maps instantaneous queue length to a drop probability, automatically adjusting the slope and intercept of the mapping function to account for changes in traffic load and to keep queue length within the desired operating range. We demonstrate that straightforward selection of HRED parameters results in stable operation under steady load and rapid adaptation to changes in load. Simulation and implementation tests confirm this stability, and indicate that overall performances of HRED are substantially better than those of earlier AQM algorithms. Finally, HRED control parameters provide several intuitive approaches to trading between required memory, queue stability, and response time.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.767-771
• /
• 2004

In this paper the performance of some control theory-based Active Queue Management (AQM) methods are improved and investigated through simulation. The simulation scenario is considered to be closely realistic, in which the traffic is mainly due to http sessions. The results show the best performance of PID controller with respect to the others in good queue regulation as well as high link utilization and low delay metrics.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2005년도 추계종합학술대회
• /
• pp.159-162
• /
• 2005

Random Early Detection (RED), one of the most well-known Active Queue Management (AQM), has been designed to substitute Tail Drop and is nowadays widely implemented in commercially available routers. RED algorithm provides high throughput and low delay as well as a solution of global synchronization. However RED is sensitive to parameters setting, so the performance of RED, significantly depends on the fixed parameters. To solve this problem, the Adaptive RED (ARED) algorithm is suggested by S. Floyd. But, ARED also uses fixed parameters like target-queue length; it is hard to respond to bursty traffic actively. In this paper, we proposed AQM algorithm based on the variation of current queue length in order to improve adaptability about burst traffic. We measured performance of proposed algorithm through a throughput, marking-drop rate and bias phenomenon. In experimentation, we raised a packet throughput as reduced packet drop rate, and we confirmed to reduce a bias phenomenon about bursty traffic.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 제37회 하계학술대회 논문집 D
• /
• pp.1807-1808
• /
• 2006

In this paper, we present the wavelet neural network (WNN) controller as an active queue management(AQM) in end-to-end TCP network. AQM is important to regulate the queue length and short round trip time by passing or dropping the packets at the intermediate routers. As the role of AQM, the WNN controller adaptively controls the dropping probability of the TCP network and is trained by gradient-descent algorithm. We illustrate our result that WNN controller is superior to PI controller via simulations.

• Journal of information and communication convergence engineering
• /
• 제6권1호
• /
• pp.19-23
• /
• 2008

The AQM (Active Queue Management) starts dropping packets earlier to notify traffic sources about the incipient stage of congestion. The AQM improves fairness between response flow (like TCP) and non-response flow (like UDP), and it can provide high throughput and link efficiency. In this paper, we suggest the QVARED (Queue Variation Adaptive RED) algorithm to respond to bursty traffic more actively. It is possible to provide more smoothness of average queue length and the maximum packet drop probability compared to RED and ARED (Adaptive RED). Therefore, it is highly adaptable to new congestion condition. Our simulation results show that the drop rate of QVARED is decreased by 80% and 40% compare to those of RED and ARED, respectively. This results in shorter end-to-end delay by decreasing the number of retransmitted packets. Also, the QVARED reduces a bias effect over 18% than that of drop-tail method; therefore packets are transmitted stably in the bursty traffic condition.