Abstract: According to some embodiments, an orchestrator node (15) is configured to receive an indication of a first plurality of wireless devices (22) where each wireless device (22) of the first plurality of wireless devices (22) meets a first criteria associated with a first level of monitoring of bi-directional network communications; in response to the indication of the first plurality of wireless devices (22), configure a second VSA (32) to monitor communications associated with each of the first plurality of wireless devices (22); receive, from the second VSA (32), an indication of a first subset of the first plurality of wireless devices (22) where each wireless device (22) of the first subset of the first plurality of wireless devices (22) meets a second criteria associated with a second level of monitoring of bi-directional network communications; and modify a granularity of wireless device monitoring.

Abstract: The disclosure relates to a method and system, for managing components of a fifth generation (5G) network slice. The method comprises retrieving current locations of a plurality of user equipments (UEs) connected to radio base stations (RBSs) in communication with the 5G network slice; predicting future traffic at the RBSs based on past and current locations of the plurality of UEs; and managing the components of the 5G network slice based on the predicted future traffic patterns.

Abstract: The disclosure relates to a method for training a machine learning (ML) model for image-based object detection. The method comprises obtaining the image, at least one image-class label identifying a class of object present in the image, and a plurality of indicators of candidate areas in the image. The method comprises the following steps executed iteratively. Computing a score, for each image-class label, for each of the plurality of indicators of candidate areas in the image. For each image-class label, selecting an indicator of a candidate area, based on the score associated with the indicator of the candidate area for the image-class label. Executing a training epoch of the ML model using the image as input and each image-class label and each corresponding selected indicator of the candidate area as labels for the image. Obtaining a predicted area for each image-class label.

Abstract: The disclosure relates to a method, apparatus and computer readable media, for detecting and correcting passive intermodulation (PIM). The method comprises receiving a radio frequency (RF) signal. The method comprises providing the RF signal to a first and a second PIM detectors. The method comprises, upon detection of PIM by at least one of the first and second PIM detectors, for which, based on a schedule, the detection of PIM is applicable at a time of detection of PIM, activating PIM correction.

Abstract: A method (700) for distributed machine learning, ML, using a set of N CDs. The method includes obtaining first topology information, wherein, for each CD included in the set of N CDs, the first topology information identifies other CDs included in the set of N CDs to which the CD is connected, thereby identifying a set of CD pairs. The method also includes obtaining first network state information, wherein, for each CD included in the set of N CDs, the first network state information comprises a network state vector for the CD, wherein the network state vector for the CD comprises, for each other CD to which the CD is indicated as being connected by the first topology information, a determined network state value. The method further includes determining a consensus weight matrix (W) using the obtained first network state information and the first topology information.

Abstract: There is provided a computer-implemented method for analysing one or more network attributes of a network. One or more first network attributes of the network are analysed using a first machine learning model to generate a first output comprising information about an event in the network. If an estimated confidence level for the first output is less than a confidence level threshold, the one or more first network attributes are analysed using a second machine learning model to generate a second output. The second output is indicative of one or more second network attributes of the network to analyse using the first machine learning model.

Abstract: Methods, systems and computer program products for multi-node analysis, are provided. Such methods, systems and products may comprise, or may comprise instructions operable to configure one or more processors to perform, a set of acts. Such acts may comprise receiving a performance indicator equation comprising a corresponding set of measurement parameters. Such a set of acts may also comprise, for each node from a set of nodes, receiving a set of measurement values for that node from a database and determining that node's contribution to a cluster value for a performance indicator corresponding to the performance indicator equation. Such a set of acts may also comprise reporting performance indicator information for the set of nodes wherein the reported performance indicator information is based on relative contributions of each node from the set of nodes to the cluster value for the performance indicator corresponding to the performance indicator equation.

Abstract: Security protocols and techniques are provided to reduce security risks posed by anomaly traffic in a WSN. The WSN includes a WSN base station that communicates with one or more IoT servers over a 5G network or other backbone network. The WSN is managed by a WSN Central Office. The WSN base station monitors IoT traffic based on a detection model trained at a central office and detects the presence of anomaly traffic. Responsive to the detection of the anomaly traffic, the WSBS triggers a reconfiguration of the WSN to create a special path for the anomaly traffic to minimize the impact of the anomaly traffic on the WSN. Additionally, the central office can request the 5G network to update the NSI allocation to dedicate a network slice instance (NSI) for the anomaly traffic to segregate the anomaly traffic from the normal IoT traffic in the 5G network.

Abstract: An Aluminum nitride ceramic and preparation method thereof. The aluminum nitride ceramic comprises a porous aluminum nitride matrix. A ferrite is loaded on the pore surface of the porous aluminum nitride matrix; and nano nickel particles are loaded on the surface of the ferrite. The preparation method of the aluminum nitride ceramic comprises steps: sintering the aluminum nitride ceramic by pressureless sintering method, depositing the ferrite on pore surface of porous aluminum nitride matrix by hydrothermal method, and loading nano nickel particles on the surface of the ferrite by reduction method. A micro-reactor is provided. So that the technical problems: the preheating time of the micro-reactor prepared is too long, nickel particles fall off from the surface of matrix, and nano nickel particles grow up due to quick and direct temperature rise can be solved.

Abstract: There is provided a method for detecting cyber-attacks at edge servers. The method comprises receiving network traffic. The method comprises extracting and normalizing network flow features from the network traffic to produce a data pattern for evaluation. The method comprises computing an anomaly detection score for the data pattern for evaluation. The method comprises comparing the anomaly detection score with a threshold and determining if the network traffic corresponds to normal traffic or to an anomaly. The method comprises, upon determining that the network traffic corresponds to an anomaly, mitigating malicious network traffic by triggering a mitigation strategy and sending an anomaly message to an anomaly classifier. A method for classifying cyber-attacks at a cloud server is also provided.

Abstract: An overload protection mechanism for network nodes in an edge node cluster uses two tier reinforcement learning models; one at the edge node level and one at the cluster level. The node level reinforcement learning model optimizes a routing policy for a service/application to determine whether an edge node receiving a client request shall handle the traffic for the service/application. The cluster level reinforcement learning model optimizes a policy for an application that determines which neighboring edge node shall be considered to handle a client request for the given application in case of redirection or proxy by the edge node receiving the client request.

Abstract: A method implements a network slicing controller to manage network slicing instances in an edge cloud platform. The method includes receiving at least one policy change from an artificial intelligence powered smart traffic controller (APSTC) or an artificial intelligence powered edge traffic controller (APETC), determining whether the at least one policy change is valid based on local monitoring information, and sending the at least one policy change to a common control network function in a 5G mobile network.

Abstract: An overload protection mechanism for network nodes in an edge node cluster uses two tier reinforcement learning models; one at the edge node level and one at the cluster level. The node level reinforcement learning model optimizes a routing policy for a service/application to determine whether an edge node receiving a client request shall handle the traffic for the service/application. The cluster level reinforcement learning model optimizes a policy for an application that determines which neighboring edge node shall be considered to handle a client request for the given application in case of redirection or proxy by the edge node receiving the client request.

Abstract: The present disclosure provides a system for coordinating the distribution of resource instances (e.g. Kubernetes nodes) that belong to different infrastructure providers providing resource instances at different locations. Each infrastructure provider provides one or more resource instances. The resource instances provide by an infrastructure can be spread over multiple locations. Several Kubernetes master nodes are deployed to manage the RIs spread among multiple infrastructure providers and multiple locations.

Abstract: An open edge cloud platform (OECP) enables tenants to access resources of a pool of network operator to support tenant applications. The tenant can specify a virtual location when requesting creation of a virtual network for a tenant application. The OECP creates a virtual network for the tenant application from available resources of a pool of network operators at the virtual location specified by the tenant in the request. The flexibility of selecting resources from a pool of network operators enables the tenant to access resources closer to the end user devices that will use the tenant application and thus reduce latency for applications and increase data throughput.

Abstract: The disclosure relates to an Edge Cloud Platform (ECP) and a method executed in the ECP, for providing a low-latency, distributed, multi-user application. The method comprises determining a first location of a first group of users requesting access to the multi-user application and deploying the multi-user application in a first Point of Presence (PoP) in a first Service Provider (SP) domain operative to serve the first group of users. The method comprises determining a second location of a second group of users requesting access to the multi-user application and deploying a proxy of the multi-user application in a second PoP in a second SP domain operative to serve the second group of users. The method comprises, upon determining that a Software License Agreement (SLA) exists between the first and second SPs, establishing a tunnel for linking the multi-user application and the proxy of the multi-user application, thereby providing the low-latency.

Abstract: The disclosure relates to a method and system, for managing components of a fifth generation (5G) network slice. The method comprises retrieving current locations of a plurality of user equipments (UEs) connected to radio base stations (RBSs) in communication with the 5G network slice; predicting future traffic at the RBSs based on past and current locations of the plurality of UEs; and managing the components of the 5G network slice based on the predicted future traffic patterns.

Abstract: A method, Delivery Node, DN, and Content Delivery Network, CDN, are optimized to deliver content of different categories. The DN is operative to receive a request for a content, obtain a determination of whether or not a category of content is associated with the requested content and responsive to the determination that no category of content is associated with the requested content, forward the request for the content towards an origin server and upon receiving a response from the origin server serve the content. The CDN comprises a plurality of DNs, a data processing service operative to obtain, from the DNs, and assemble, data sets into training and validation data formatted to be used for training and validating a Neural Network, NN. The CDN comprises a NN training service, operative to train and validate the NN and a configuration service, operative to configure the plurality of DNs with the trained and validated NN.

Abstract: A request router is disclosed for a multi-tier Content Delivery Network (CDN), each tier corresponding to a hierarchical layer of the CDN. The request router includes processing circuitry configured to: receive, from a first delivery node of a hierarchical layer of the CDN, a routing request to bypass the first delivery node for content delivery to a user equipment; and as a result of receiving the routing request to bypass the first delivery node for the content delivery to the user equipment, select at least a second delivery node for the content delivery to the user equipment.

Abstract: The disclosure relates to a system, method, and network node for generating at least one classification based on multiple data sources. The system comprises at least one layer comprising one or more supervised neural networks (SNN); at least one layer comprising one or more unsupervised neural networks (USNN); and at least one normalization layer. Each of the layers has inputs and outputs, the inputs of a first layer being operative to receive data from the data sources, the inputs of a layer other than the first layer being communicatively connected to the outputs of a previous layer, the outputs of a layer other than a last layer being communicatively connected to inputs of a following layer, the last layer having at least one output, and the at least one normalization layer being operative to normalize the outputs from the previous layer into normalized inputs for the following layer.