Automation & AI/ML Platforms
Automation platforms provide operational efficiency and deliver superior customer experience through network automation - one of the key tenets of running networks at scale. It focuses on delivering automation and insights in various aspects of network planning, optimization and operations. JPL’s unique analytics applications helps to leverage untapped data into actionable insights and real-time decisions, ensuring a superior customer experience.
Adaptive Troubleshooting and Operations Management Platform (ATOM)
ATOM Platform is a cloud native data lake platform tailored for carriers to enable smarter operations through Machine learning as a Service (MLaaS). It leverages machine learning to detect anomalous network patterns and create reports and alerts based on these patterns for proactive root cause analysis and resolution before the network symptoms start affecting operations. The platform provides operational insights, data binding, correlation and automated analysis using AI / ML algorithms. It bring organizational transparency with Automated SLA management capabilities in the workflow engine, orchestrating the operational tasks between the systems.
Jio Cognitive Platform
Jio Cognitive Platform (JCP) is an AI driven platform for end-to-end telecom network management. It is built on top of a state-of-the-art Big Data architecture touching upon every aspect of the network – from the core to the radio and even the business functions of a telecom service provider. JCP offers features which helps to automate coverage simulation, TAC Audits and proactively improve coverage simulations with live customer data. Jio Cognitive platform deploys multiple algorithms to support closed loop automation for full fledged life cycle of planning, deploying and maintaining operations of radio access network.
Core Network Operations Platform (CNOPS)
5G CN-OPS is Core Network Operations platform, which automates the process of upgrades of the network elements. With a view to provide best customer experience, JPL's CN-OPS is deeply integrated with AI, data native, and automation technologies based on the technical advantages in the 5G core network and in-depth understanding of O&M services which further reduces the time taken for upgrades in production and thus allows speedy roll out of the services. In addition, network automation results in simpler management tasks through automated updates, which also reduces risk of error occurrence.
Integrated Virtual Probe and Real-Time Intelligence solution provides telecom operators with unprecedented insights into network performance and customer behaviour. With this solution, MNOs can reduce hardware costs, network complexity and the performance issues typically associated with legacy probe solutions, while providing real-time network intelligence. The solution has the capability to deploy vProbes embedded in the network functions. This makes it possible to probe geographically distributed Virtualized Network Functions (VNFs)/Cloud Native Functions (CNFs) in a scalable and cost-effective manner. vProbe is a probing agent which collects probing data (Streaming Data Record-SDR) from the network nodes. Streaming these records in real-time from the probe into the data analytics engine, enables operators to eliminate data silos, rapidly develop valuable reporting tools, and make informed business decisions.
Correlation Engine uses the learning models and machine learning algorithms to correlate the alarms with the clear codes or infrastructure events received from other systems. It provides operational insights, and data binding and data correlation feature. It also constantly monitors and compares the collected data with the baseline behaviour to detect any deviations. Correlation Engine (CRE) provides an environment to correlate user defined workflow steps. Based on the data received over message broker, CRE executes the corresponding workflow steps and fetch the network data inserted by normalization layer. This correlated data will help in troubleshooting activities. Using this correlated data, JPL's ATOM (Adaptive Troubleshooting Operations & Management) Analytics Platform, is able to detect anomalous network patterns and create reports and alerts based on these patterns. On any violation, the pre-defined remediation action is triggered in order to maintain network consistency.
xProbes- RAN Probing Solution
RAN Probes or xProbes are generic xApps hosted by RICs running on gNodeB for collecting call traces and feeding them to upstream AI/ML engines at real time. The analytics platform ATOM enables the use of xProbes in the 5G network which are connected through a real-time conductor fabric & leverages the existing machine learning engines of the platform. It also helps in anomaly detection on the collected data based on the historical models, perform forecasting on the available data or normalize it to use it for different debugging, reporting and correlation purposes. JPL’s RAN Probes are capable to be configured and scale through software commands, making it dynamically adjustable. RAN Probes have deployment flexibility which allows multi-level aggregation during execution.
RAN Probes also supports detection of anomalies or defaulters by detecting deviation in terms of success % and failure %. It also includes root cause diagnosis to find and resolve the issue detected from depth and provide a concrete resolution or feedback on the same.
SON (Self Organizing Network)
Self Organizing Network (SON) is an automation technology designed to make the planning, configuration, management, optimization and healing of mobile Radio Access Networks simpler and faster. It is one of the solutions to make the system more efficient with Quality of Experience (QoE). SON has the capability to remain aware of current status and the ongoing changes, and supports the ability to do necessary analysis to determine optimal network parameter values, to implement the network adjustment, and also to provide network maintenance in an optimal and timely fashion. It is an automated adaptive network, capable of performing a set of functions with minimum human intervention. It reduces OPEX, as some functions are automatized and may only require minimum human intervention, and, also, reduces CAPEX, as better usage of the resources can be done and, therefore, the service may improve without the need of deploying more infrastructures.