Reinforcement Learning for Active Queue Management in Mobile All-IP Networks

Vučević, Nemanja; Perez-Romero, J.; Sallent, O.; Agustí, R.

doi:10.1109/pimrc.2007.4394713

Cited by 11 publications

(10 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each randomly derived value the distribution and the mean value are indicated. The described traffic model has previously been used for similar evaluations in [20] and [21].…”

Section: A Traffic Sourcesmentioning

confidence: 99%

See 1 more Smart Citation

Reinforcement Learning for Load Management in DiffServ-MPLS Mobile Networks

Vučević

Perez-Romero

Sallent

et al. 2009

VTC Spring 2009 - IEEE 69th Vehicular Technology Conference

Self Cite

View full text Add to dashboard Cite

Abstract-Cognitive networks are envisaged to provide optimized resource usage in future. While heterogeneity and resource scarcity draw research attention to the wireless part, the rest of the network (mobile backhaul) is rarely considered for these improvements. The future of next generation wireless networks is probable to be all-IP, where a common flexible infrastructure is looking for dynamic autonomous solutions that cognition may provide.This work proposes a novel solution, where the introduction of reinforcement learning over multiprotocol label switching (MPLS) in a differentiated services (DiffServ) mobile backhaul should provide autonomous network adaptation aiming at enhanced QoS capabilities. The proposed solution enables intelligent traffic routing by means of distributed reinforcement learning agents that base decisions on edge-gained experience.

show abstract

“…For each randomly derived value the distribution and the mean value are indicated. The described traffic model has previously been used for similar evaluations in [20] and [21].…”

Section: A Traffic Sourcesmentioning

confidence: 99%

“…Thus, the proposed solution affects only network extremes. Use of such a domain edge experience for autonomous network adapting by means of RL in scheduling [20] or queue management [21] has been applied to achieve end-to-end goals.…”

Section: Introductionmentioning

confidence: 99%

Reinforcement Learning for Load Management in DiffServ-MPLS Mobile Networks

Vučević

Perez-Romero

Sallent

et al. 2009

VTC Spring 2009 - IEEE 69th Vehicular Technology Conference

Self Cite

View full text Add to dashboard Cite

show abstract

“…Kim et al [16] considered the fairness of customer services and improved the efficiency of the network while reducing the difference of customers' waiting time. Vučević et al [17] and Zhou et al [18] used a reinforcement learning (RL) algorithm to optimize queue management that allocates the data packets to the queues.…”

Section: Introductionmentioning

confidence: 99%

Online Self-Organizing Network Control with Time Averaged Weighted Throughput Objective

Zhang

Ли

Yan

2018

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

We study an online multisource multisink queueing network control problem characterized with self-organizing network structure and self-organizing job routing. We decompose the self-organizing queueing network control problem into a series of interrelated Markov Decision Processes and construct a control decision model for them based on the coupled reinforcement learning (RL) architecture. To maximize the mean time averaged weighted throughput of the jobs through the network, we propose a reinforcement learning algorithm with time averaged reward to deal with the control decision model and obtain a control policy integrating the jobs routing selection strategy and the jobs sequencing strategy. Computational experiments verify the learning ability and the effectiveness of the proposed reinforcement learning algorithm applied in the investigated self-organizing network control problem.

show abstract

“…9 of 3GPP to support speech services in LTE systems [13] [14]. Basically, ECN is an AQM (Active Queue Management) scheme [15] by which the network can notify about the congestion to its senders and receivers, so that the senders can reduce their transmission rates, before the packets are forced to drop or the end-to-end delay occurs. By interacting (for example) with the ECN-policer [16] [17], the PCRF can acquire the network congestion information and based on this information a few necessary measures (such as rate reduction, packet dropping) can be taken to provide operators with better QoS decision policy.…”

Section: Introductionmentioning

confidence: 99%

Comparison of Policy Realization Strategies for LTE Networks

Mir

Nuaymi

2012

2012 IEEE Vehicular Technology Conference (VTC Fall)

View full text Add to dashboard Cite

International audienceEfficient radio resource usage and quality of service (QoS) provision are important criterions in order to choose the right QoS policy for the LTE operators. In our previous work, we have proposed to add a new database named as BSLRC (Base Station Load and Radio Conditions) to optimize the operator policy with regard to the radio efficiency [1]. Some concrete voice capacity estimations were also proposed. In this paper, we extend our earlier work by addressing some novel concerns. As a first step, we start by calculating the signaling load due to the use of BSLRC. Then, for voice services the BSLRC method is compared with the explicit congestion notification (already proposed for LTE systems). Both BSLRC and explicit congestion notification (ECN) share the same objective to optimize the choice of codec rate with regard to radio use efficiency. We also propose some capacity estimates for video services as an extension to our previously proposed voice capacity estimations over LTE systems

show abstract

Reinforcement Learning for Active Queue Management in Mobile All-IP Networks

Cited by 11 publications

References 7 publications

Reinforcement Learning for Load Management in DiffServ-MPLS Mobile Networks

Reinforcement Learning for Load Management in DiffServ-MPLS Mobile Networks

Online Self-Organizing Network Control with Time Averaged Weighted Throughput Objective

Comparison of Policy Realization Strategies for LTE Networks

Contact Info

Product

Resources

About