Aira Naavik Anomaly Detection rApp
Aira Technologies Inc | Network Healing
The Naavik AI-based Anomaly Detection rApp provides automated early detection, prediction, and decision-making capabilities tailored for mobile RAN environments.
Leveraging advanced AI and GenAI models, this solution utilizes a data-driven causality graph to analyze and interpret multidimensional data sources in real time. The rApp enables rapid identification of correlations and causations within critical RAN KPIs, which is essential for effective root cause analysis (RCA). By automatically detecting anomalies and predicting performance trends, Naavik empowers operators to optimize network performance, reduce downtime, and streamline operational efficiency through precise and actionable insights.
Utilizes AI/ML models trained on historical network data to detect and identify anomalies within the network
Naavik Anomaly Detection rApp incorporates a KPI Causality Graph Generator module, which applies a data-driven methodology to identify relationships between a specific Key Performance Indicator (Target KPI) and related Source KPIs. The solution utilizes a model-based approach, fine-tuned for telecom environments, to predict both target and source KPI behaviors. This enables highly accurate anomaly detection, ensuring reliable identification of performance issues within the network.
Enhance network performance and optimize the subscriber experience.
The rApp enables operators to configure varying sensitivity levels for detecting anomalies across different KPIs, allowing for proactive identification of potential issues. This early detection empowers operators to respond swiftly, addressing problems before they escalate and minimizing any potential impact on subscribers' quality of experience.
Accelerate the resolution of network issues
By offering a proactive approach to detecting network anomalies, the rApp significantly reduces issue resolution time. With real-time anomaly detection, the operational team is immediately alerted, enabling faster response and minimizing the impact of network disruptions.