Artificial Intelligence-Powered Mobile Edge Computing-Based Anomaly Detection in Cellular Networks

Abstract

Escalating cell outages and congestion-treated as anomalies-cost a substantial revenue loss to the cellular operators and severely affect subscriber quality of experience. Stateof-the-art literature applies feed-forward deep neural network at core network (CN) for the detection of above problems in a single cell; however, the solution is impractical as it will overload the CN that monitors thousands of cells at a time. Inspired from mobile edge computing and breakthroughs of deep convolutional neural networks (CNNs) in computer vision research, we split the network into several 100-cell regions each monitored by an edge server; and propose a framework that pre-processes raw call detail records having user activities to create an image-like volume, fed to a CNN model. The framework outputs a multilabeled vector identifying anomalous cell(s). Our results suggest that our solution can detect anomalies with up to 96% accuracy, and is scalable and expandable for industrial Internet of things environment.

Anomaly Detection in Multi Tiered Cellular Networks Using LSTM and 1D-CNN

“…Achieved maximum True Positive Rates can be considered rather successful when compared to earlier works. For example, in [13] researchers have reached similar maximum accuracies in detection of anomalies in a cellular network of macro BSs for two states. Therefore, in this study, attaining that high TPRs on detection of FAP anomalies, which is far more difficult than detection of macro BS anomalies due to reasons such as sparse user statistics and vertical handovers, stands rather promising in favor of LSTM and 1D-CNN approaches intended for future work [3].…”

“…In [12] Recurrent Neural Networks were comparatively analyzed along with traditional Support Vector Machines in terms of COD performance. Another study was introduced in [13] where anomaly detection in cellular networks was studied by using Convolutional Neural Networks (CNN). On the other hand, some researchers have presented studies about COD in only small cells so as to focus on its difficulties when compared to relatively easier detection of macro cell anomalies.…”

Network Public Opinion Prediction and Control Based on Edge Computing and Artificial Intelligence New Paradigm

“…This model is called the opinion exchange model [14]. A theoretical framework is constructed to discuss the conditions under which the common opinions among members are affected by the society over time, the formation of network contact methods defined by the environment, and the attributes of individual agents and time-related factors, as well as the influence of joint effect on the dissemination of public opinion [15]. Ignore the costs of resource is put forward to maximize the expected utility of fully rational decision makers the best decisions, puts forward a thermodynamic inspired the standardization of the limited rational decision, with limited rational decision problems can describe famous variational principle and reflect the thinking of the people in the spread of public opinion and decision, from another aspect to think [16] individual Internet users to the dissemination of public opinion or not.…”

“…Ignore the costs of resource is put forward to maximize the expected utility of fully rational decision makers the best decisions, puts forward a thermodynamic inspired the standardization of the limited rational decision, with limited rational decision problems can describe famous variational principle and reflect the thinking of the people in the spread of public opinion and decision, from another aspect to think [16] individual Internet users to the dissemination of public opinion or not. Starting from the two dimensions of political participation, interaction, and democratic supervision of the subjects of online public opinion, the influence of online public opinion on online democracy is theoretically analyzed and studied [17,18]. From the perspective of mastering the initiative of online public opinion, this paper puts forward that if we want to grasp the law correctly, we need to collect and analyze the online public opinion scientifically to master the initiative of online public opinion, so as to guide the online public opinion correctly.…”

Energy-Efficient Resource Provisioning Strategy for Reduced Power Consumption in Edge Computing

“…The edge computing architecture is proposed to solve the problem of large cloud computing delay [7][8][9]. Specifically, edge computing handles tasks that require high latency on the near-user side and puts tasks that could only be processed on the cloud computing platform into the edge server [10,11]. Since the devices on the edge layer are limited power and computing capacities, poor resource allocation for the tasks will result in high latency and power consumption.…”