Abstract: Architectures and techniques are described that can provide detection and mitigation techniques for self-conflicting behavior associated with an xApp that is executed on a near-real time radio access network intelligent controller (near-RT RIC). A first set of inputs relating to xApp data (e.g., data consumed by the xApp, parameters controlled by the xApp, xApp description of purpose or function, . . . ) can be input to a first machine learning model to determine potential self-conflicting behavior. A second set of inputs relating to xApp control data (e.g., a current control action, previous control action history, . . . ) can be input to a second machine learning model to determine other types of potential self-conflicting behavior. If self-conflicting behavior is detected, then mitigation techniques can be invoked.

Abstract: Architectures and techniques are described that can provide conflict mitigation techniques for xApp that are executed on a near-realtime radio access network intelligent controller (near-RT RIC). A first, deep learning machine learning model can be employed during registration of the xApps, which can identify potential conflicts and flag those potentially conflicting xApps. A second machine learning model can be employed during run time of the xApps, which can identify whether a control message from a given flagged xApp conflicts with a configuration applied to an E2 node by another flagged xApp based on the current network state. Conflicts can be mitigated based on a priority between the two conflicting xApps.

Abstract: Overshooting mitigation in an open radio access network (e.g., using a computerized tool), is enabled. For example, a system can comprise a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising, based on source cell data applicable to a source cell and based on neighbor cell data applicable to a neighbor cell, determining, using an overshooting classifier model, whether overshooting from the source cell to the neighbor cell has occurred, in response to determining that the overshooting has occurred, determining, using a distance recommendation model, a recommended cellular transmission distance applicable to the source cell, and based on the recommended cellular transmission distance, causing an overshooting mitigation action to be performed that has been determined to conform a cellular transmission distance, applicable to the source cell, to the recommended cellular transmission distance.

Abstract: Dynamic node key performance indicator reporting in an open radio access network (e.g., using a computerized tool), is enabled. For example, a system can comprise a processor and a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising determining resource utilization of a network node, determining a requested reporting period of a key performance indicator, applicable to the network node, requested by an extended application of a radio access network intelligent controller, and using a reporting model generated using machine learning based on past resource utilization, other than the resource utilization, and past reporting periods, other than the requested reporting period, generating a reporting period recommendation applicable to the network node and the extended application.

Abstract: Architectures and techniques are described relating to implementing a serving hub within a network architecture such as a open radio access network (O-RAN). The serving hub can, inter alia, abstract a MLOps layer, serve models, and expose models for reuse. All or a portion of the serving hub can be deployed on a near real time radio access network intelligent controller (NRT RIC) with the O-RAN architecture. Hence, xApps that execute on the NRT RIC can subscribe to MLOps services without the need to implement MLOps in the xApps.

Abstract: One example method includes a pipeline for a distributed neural network. The pipeline includes a first phase that identifies intersecting identifiers across datasets of multiple clients in a privacy preserving manner. The second phase includes a distributed neural network that includes a data receiving portion at each of the clients and an orchestrator portion at an orchestrator. The data receiving portions and the orchestrator portions communicate forward and backward passes to perform training without revealing the raw training data.

Abstract: One example method includes identifying emerging hybrid technologies. Datasets are data mined to identify transactions that are related to technology. Associations are generated based on the transactions and the associations are processed to identify emerging hybrid technologies from the transactions.

Abstract: One example method includes identifying datasets known to contain data relating to different technologies, extracting the data from the data sources, performing preprocessing on the data, clustering the data after the data has been preprocessed, and the clustering comprises generating data clusters, and mapping each of the data clusters to a phase of a Wardley Map. The mapping may be used to make a prediction about which phase a particular technology would fall under in a future time period.

Abstract: One example method includes a pipeline for a distributed neural network. The pipeline includes a first phase that identifies intersecting identifiers across datasets of multiple clients in a privacy preserving manner. The second phase includes a distributed neural network that includes a data receiving portion at each of the clients and an orchestrator portion at an orchestrator. The data receiving portions and the orchestrator portions communicate forward and backward passes to perform training without revealing the raw training data.

Abstract: A computer readable storage medium comprises executable instructions to provide a query interface for formulating continuous queries of multi-dimensional data streams. The query interface is populated to generate a plurality of continuous queries. The plurality of continuous queries are executed. The plurality of continuous queries are modified at run-time.