Search

Abstract Fast and accurate anomaly detection is critical in telemetry systems because it helps operators take appropriate actions in response to abnormal behaviours. However, recent techniques are accurate but not fast enough to deal with real-time data. There is a need to reduce the anomaly detection time, which motivates us to propose two new algorithms called AnDePeD (Anomaly Detector on Periodic Data) and AnDePed Pro. The novelty of the proposed algorithms lies in exploiting the periodic nature of data in anomaly detection. Our proposed algorithms apply a variational mode decomposition technique to find and extract periodic components from the original data before using Long Short-Term Memory neural networks to detect anomalies in the remainder time series. Furthermore, our methods include advanced techniques to eliminate prediction errors and automatically tune operational parameters. Extensive numerical results show that the proposed algorithms achieve comparable performance in terms of Precision, Recall, F-score, and MCC metrics while outperforming most of the state-of-the-art anomaly detection approaches in terms of initialisation delay and detection delay, which is favourable for practical applications.

The fifth-generation networks User Plane Function (UPF) provides necessary procedures to connect end-devices and data networks. A Protocol Data Unit (PDU) session between specific user equipment (UE) and the UPF is established before data transmission in 5G networks. The UPF software can be packed into either a virtual machine (VM) or a container image and instantiated in service providers’ cloud infrastructure. Each UPF instance may handle several concurrent PDU sessions from various users. Operators should control the number of UPF instances to save resource consumption in the provision of Quality of Service (QoS). This paper presents a queueing model for a scenario where a threshold-based algorithm controls the number of UPF instances depending on users’ traffic. We derive a method to compute the steady-state probabilities and performance measures efficiently. We investigate the performance of the UPF scaling algorithm in various scenarios with the assumption of specific hardware. Numerical results show that scaling UPF instances can lead to the good utilization of the system resource.

In the 5G core and the upcoming 6G core, the User Plane Function (UPF) is responsible for the transportation of data from and to subscribers in Protocol Data Unit (PDU) sessions. The UPF is generally implemented in software and packed into either a virtual machine or container that can be launched as a UPF instance with a specific resource requirement in a cluster. To save resource consumption needed for UPF instances, the number of initiated UPF instances should depend on the number of PDU sessions required by customers, which is often controlled by a scaling algorithm. In this paper, we investigate the application of Deep Reinforcement Learning (DRL) for scaling UPF instances that are packed in the containers of the Kubernetes container-orchestration framework. We propose an approach with the formulation of a threshold-based reward function and adapt the proximal policy optimization (PPO) algorithm. Also, we apply a support vector machine (SVM) classifier to cope with a problem when the agent suggests an unwanted action due to the stochastic policy. Extensive numerical results show that our approach outperforms Kubernetes’s built-in Horizontal Pod Autoscaler (HPA). DRL could save 2.7–3.8% of the average number of Pods, while SVM could achieve 0.7–4.5% saving compared to HPA.

The impact of the traffic characteristics of secondary users (SUs) on the performance of cognitive radio networks (CRNs) should be understood for designing operation rules. This paper focuses on multi-type burst services and network congestion problem in CRNs and evaluates the QoS of SUs based on multiple cross-layer considerations. A two-state Markov-modulate Bernoulli process (MMBP-2) is adopted to model packet flows of SUs with different burst degrees in CRNs. We propose a two-dimensional discrete queuing model to consider spectrum access, burstiness of traffic, network congestion, channel environment, user activity and finite buffer. We construct an iterative algorithm to compute the steady-state distribution of the proposed queuing model and determine the performance metrics like the throughput, the average delay, the average queue length and the total packet loss probability. Numerical analysis evaluates the QoS of SUs under different burst environments.

Data throughput is an important metric used in the performance evaluation of the next generation cellular networks such as long-term evolution (LTE) and LTE-advanced. To evaluate the performance of these networks, Monte Carlo simulation schemes are usually used. Such simulations do not provide the throughput of intermediate call state; instead it gives the overall performance of this network. The authors propose a hybrid model consisting of both analysis and simulation. The benefit of this model is that the throughput of any possible call state in the system can be evaluated. Here, the probability of possible call distribution is first obtained by analysis, which is used as input to the event-driven-based simulator to calculate the throughput of a call state. Comparison is made between throughput obtained from the author's hybrid model with that obtained from event-driven-based simulation. Numerical results are presented and show good agreement between both the proposed hybrid model and the simulation. The maximum difference of relative throughput between their hybrid model and the simulation is found in the interval of (0.04 and 1.06%) over a range of call arrival rates, mean holding times and number of resource blocks in the system.