Machine Learning Adaptive Computational Capacity Prediction for Dynamic Resource Management in C-RAN

Abstract

Efficient computational resource management in 5G Cloud Radio Access Network (C-RAN) environments is a challenging problem because it has to account simultaneously for throughput, latency, power efficiency, and optimization tradeoffs. The assumption of a fixed computational capacity at the baseband unit (BBU) pools may result in underutilized or oversubscribed resources, thus affecting the overall Quality of Service (QoS). As resources are virtualized at the BBU pools, they could be dynamically instantiated according to the required computational capacity (RCC). In this paper, a new strategy for Dynamic Resource Management with Adaptive Computational capacity (DRM-AC) using machine learning (ML) techniques is proposed. Three ML algorithms have been tested to select the best predicting approach: support vector machine (SVM), time-delay neural network (TDNN), and long short-term memory (LSTM). DRM-AC reduces the average of unused resources by 96 %, but there is still QoS degradation when RCC is higher than the predicted computational capacity (PCC). To further improve, two new strategies are proposed and tested in a realistic scenario: DRM-AC with pre-filtering (DRM-AC-PF) and DRM-AC with error shifting (DRM-ACES), reducing the average of unsatisfied resources by 98 % and 99.9 % compared to the DRM-AC, respectively.

ML-Based Radio Resource Management in 5G and Beyond Networks: A Survey

“…Guerra-Gómez et. al [85] propose a dynamic resource management scheme, based on the prediction of the total system's capacity. They use three different ML algorithms: SVM, DNN, and LSTM.…”

“…DL methods, due to their ability to mine deep data and label associations through multiple complex hidden layers (ANNs, DNNs, CNNs), are mainly used in user, subcarrier, power allocation and CSI prediction tasks [86], [87]. The multiparameter nature of the RRM problem and the complex channel feature associations render DL approaches as the most efficient way to deal with the total RRM problem [83], [84], [85].…”

“…In other words, accuracy is the sum of True Positive (TP) and True Negative (TN) predictions, divided by the number of the total predictions (TP + TN + False Positive (FP) + False Negative (FN)). Then, F1-score is given by the following formula: [83], [84], [85], [86], [87] Unsupervised Learning RN or BS selection unsupervised k-NN, k-Means clustering variations [88] Unsupervised Learning user grouping, clustering, handover management k-NN, k-Means, Agg-GNN, f-test [89], [90], [91], [92] Reinforcement Learning subcarrier allocation, power control, frequency selection MDP, DRL, Water-Filling, WMMSE [93]. [94], [95], [96], [97] Reinforcement Learning minimization of difference between requested and active KPIs (throughput, SNIR, CSI)…”

A Survey on Applications of Deep Learning in Cloud Radio Access Network

“…The authors of [109] proposed a technique called dynamic resource management with adaptive computational capacity (DRM-AC) utilizing supervised ML and DL approaches. Their main goal was to minimize the underutilized resources in the BBU pool of C-RAN while maintaining QoS.…”

“…In [109], for the prediction of the required computational capacity in the proposed DRM-AC technique, real-world traffic data were used. The data was divided into 80% training and 20% test dataset.…”