Traffic Prediction and Random Access Control Optimization: Learning and Non-Learning-Based Approaches

  title={Traffic Prediction and Random Access Control Optimization: Learning and Non-Learning-Based Approaches},
  author={Nan Jiang and Yansha Deng and Arumugam Nallanathan},
  journal={IEEE Communications Magazine},
Random access channel (RACH) procedures in modern wireless communications are generally based on multi-channel slotted-ALOHA, which can be optimized by flexibly organizing devices' transmission and retransmission. However, due to the lack of information about the traffic generation statistics and the occurrence of random collision, optimizing RACH in an exact manner is generally challenging. In this article, we first summarize the general structure of optimization for different RACH schemes… Expand
3 Citations
Delay-aware dynamic access control for mMTC in wireless networks using deep reinforcement learning
This work proposes a delay-aware double deep reinforcement learning mechanism that can dynamically adapt two parameters of the system in order to enhance the probability of successful access using ACB, while at the same time reducing the expected delay by modifying the Random Access Opportunity (RAO) periodicity. Expand
Machine Learning for Massive Industrial Internet of Things
The QoS requirements of the typical massive non-critical and critical IIoT use cases are summarized and the corresponding machine learning solutions with its limitations and potential research directions are presented. Expand
Learning-Based Prediction, Rendering and Association Optimization for MEC-Enabled Wireless Virtual Reality (VR) Networks
Simulation results show that the proposed MEC rendering schemes and DRL algorithms substantially improve the long-term QoE of VR users and reduce the VR interaction latency compared to rendering at VR devices. Expand


A Decoupled Learning Strategy for Massive Access Optimization in Cellular IoT Networks
A decoupled learning strategy is developed to jointly and dynamically adapt the access control factors of those three access schemes, where a Recurrent Neural Network model is first employed to predict the real-time traffic values of the network environment, and multiple DRL agents are employed to cooperatively configure parameters of each RACH scheme. Expand
Reinforcement Learning-Based ACB in LTE-A Networks for Handling Massive M2M and H2H Communications
A dynamic algorithm based on reinforcement learning to adapt the barring rate parameter of Access Class Barring is proposed, which can adapt it to different traffic conditions, reducing congestion and hence the number of collisions in the RACH. Expand
Collaborative Distributed Q-Learning for RACH Congestion Minimization in Cellular IoT Networks
A novel collaborative distributed Q-learning mechanism for the resource-constrained MTC devices in order to enable them to find unique RA slots for their transmissions so that the number of possible collisions can be significantly reduced. Expand
D-ACB: Adaptive Congestion Control Algorithm for Bursty M2M Traffic in LTE Networks
An analytical model is presented to determine the expected total service time, i.e., the time used by all MTC devices to successfully access the eNodeB, and two dynamic ACB (D-ACB) algorithms for fixed and dynamic preamble allocation schemes are proposed to determined the ACB factors without a priori knowledge of the system backlog. Expand
Traffic-aware ACB scheme for massive access in machine-to-machine networks
This paper develops a Markov-Chain based traffic-load estimation scheme according to the collision status and proposes a spectrum of functions to control the barring factor varying with the estimated traffic load, thus accommodating much more M2M devices as well as lowering the access delay. Expand
Fast Adaptive S-ALOHA Scheme for Event-Driven Machine-to-Machine Communications
Simulation results demonstrate that the proposed FASA scheme achieves near-optimal performance in reducing access delay, which outperforms that of traditional additive schemes such as PB-ALOHA. Expand
Online Supervised Learning for Traffic Load Prediction in Framed-ALOHA Networks
An online learning-based adaptive traffic load prediction method based on recurrent neural networks (RNN) and specifically on the long short-term memory (LSTM) architecture is developed that considerably outperforms conventional methods and is able to adapt changing traffic statistics. Expand
PRADA: Prioritized Random Access With Dynamic Access Barring for MTC in 3GPP LTE-A Networks
The performances of LTE-A without access barring, extended access barring (EAB), and PRADA are mathematically derived and compared, showing the superior performance of the PRADA scheme. Expand
A novel class-dependent back-off scheme for Machine Type Communication in LTE systems
A novel class-dependent back-off scheme for LTE system to protect RAN from random access channel (RACH) overload and improve the throughput by 2%-5% with additional 100-200 packets access delay which is tolerable for most of the MTC applications. Expand
Reinforcement Learning for Real-Time Optimization in NB-IoT Networks
The results show that the proposed reinforcement learning-based approaches considerably outperform the conventional heuristic approaches based on load estimation (LE-URC) in terms of the number of served IoT devices and that LA-Q and DQN can be good alternatives for tabular-Q to achieve almost the same performance with much less training time. Expand