Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing a network feature tester are disclosed. In one aspect, a method includes the actions of selecting computing devices that are configured to communicate with a wireless carrier network based on the computing devices having a common characteristic. The actions further include identifying a subset of the computing devices that are likely to communicate with a same base station of the wireless carrier network during a first period of time and a second period of time. The actions further include determining first monitoring data during the first period of time. The actions further include adjusting a characteristic of the same base station. The actions further include determining second monitoring data during the second period of time. The actions further include determining an effect that adjusting the characteristic has on the wireless carrier network.

Methods and systems for performing operations comprising: accessing one or more data objects including a data set that has been collected over a given span of time, the data set representing a plurality of parameters corresponding to resource utilization of a given server; computing first and second statistical measures based on the plurality of parameters; obtaining current resource utilization corresponding to at least a subset of the plurality of parameters; determining a first condition in which values of the current resource utilization exceed a first threshold associated with the first statistical measure; determining a second condition in which values of the data set corresponding to a time period associated with the current resource utilization exceed a second threshold associated with the second statistical measure; and triggering an anomaly detection operation in response to determining the first and second conditions.

Methods and systems for performing operations comprising: accessing one or more data objects including a data set that has been collected over a given span of time, the data set representing a plurality of parameters corresponding to resource utilization of a given server; computing first and second statistical measures based on the plurality of parameters; obtaining current resource utilization corresponding to at least a subset of the plurality of parameters; determining a first condition in which values of the current resource utilization exceed a first threshold associated with the first statistical measure; determining a second condition in which values of the data set corresponding to a time period associated with the current resource utilization exceed a second threshold associated with the second statistical measure; and triggering an anomaly detection operation in response to determining the first and second conditions.

The subject matter described herein provides systems and techniques for a network planning and optimization tool that may allow for network capacity planning using key network failures for an arbitrary pair of network topology and demands. Performing network capacity planning with key network failures, instead of using other techniques, may avoid over-building the topology of a network. In particular, key network failures may be generated from the probabilistic failures, and the impact of these failures on a network may be computed. Expected flow availability SLO or a function thereof may be computed, using this information, and used by the tool to design a robust network. With an embedded flow availability calculation and updated risk framework, the capacitated cross-layer network topologies output by the tool may meet network demands/flows with their respective SLO type at the lowest cost.

Systems and methods are disclosed to implement a time series anomaly detection system that uses configurable statistical control rules (SCRs) and a forecasting system to detect anomalies in a time series data (e.g. fluctuating values of a network activity metric). In embodiments, the system forecasts future values of the time series data along with a confidence interval based on seasonality characteristics of the data. The time series data is monitored for anomalies by comparing actual observed values in the time series with the predicted values and confidence intervals, according to the SCRs. The SCRs may be defined and tuned via a configuration interface that allows users to visually see how different SCRs perform over real data. Advantageously, the disclosed system allows users to create custom anomaly detection triggers for different types of time series data, without use of a monolithic detection model which can be difficult to tune.

The disclosure describes techniques for detecting network measurement inaccuracies through the detection of sender delays or packet drops. For example, a sender device of a test packet may determine whether the sender device is experiencing any issues in sending the test packet to a receiver device and notify a controller of the issues such that the controller may generate an indication that one or more Key Performance Indicator (KPI) measurements based on the test packets from the sender device are inaccurate and/or untrustworthy, remove the inaccurate KPI measurements, and/or adjust the inaccurate KPI measurements.

UE capability for support of machine-learning (ML) based channel environment classification may be reported by a user equipment to a base station, where the channel environment classification classifies a channel environment of a channel between the UE and a base station based on one or more of UE speed or Doppler spread, UE trajectory, frequency selectivity or delay spread, coherence bandwidth, coherence time, radio resource management (RRM) metrics, block error rate, throughput, or UE acceleration. The user equipment may receive configuration for ML based channel environment classification, including at least enabling/disabling of ML based channel environment classification. When ML based channel environment classification is enabled, UE assistance information for ML based channel environment classification, and/or an indication of the channel environment (which may be a pre-defined channel environment associated with a lookup table), may be transmitted by the user equipment to the base station.

The present disclosure is generally related to edge computing technologies (ECTs), communications networking, network slicing, and in particular, to techniques and technologies for providing flow-specific network slices. In particular, the present disclosure describes mechanisms that expand existing end-to-end architectures in order to include quality of service and monitoring mechanisms that connect network slicing technologies with infrastructure and/or network data center quality of service provider domains. The described mechanisms provide data center bridging to enable network, edge computing, and cloud computing domains.

A method includes receiving network statistics from one or more vehicles that are members of a vehicular micro cloud. The method further includes detecting a degradation in vehicle-to-vehicle (V2V) communication performance based on the network statistics. The method further includes generating a communication graph based on the network statistics. The method further includes dividing the vehicular micro cloud into subgroups based on the communication graph.

The current document is directed to methods and subsystems that instantiate and manage mobile-network computational infrastructure. The currently disclosed improved mobile-network-computational-infrastructure orchestration system employs several layers of containerized-application orchestration and management systems. For increased efficiency and security, mobile-network-specific operators are added to the containerized-application orchestration layers in order to extend the functionalities of the containerized-application orchestration layers and move virtualization-layer dependencies from the mobile-network-computational-infrastructure orchestration system down into the containerized-application orchestration layers. The improved mobile-network-computational-infrastructure orchestration system is responsible for generating, from an input mobile-network computational-infrastructure specification, one or more workload resource specifications and a node policy that are input to a containerized-application-orchestration layer. The containerized-application-orchestration layers instantiate and manage worker nodes that execute mobile-network application instances that implement VNFs and CNFs according to the one or more workload resource specifications and the node policy.