Advancements in Edge Computing and Anomaly Detection

The field of edge computing and anomaly detection is moving towards more efficient and accurate methods for real-time data processing and analysis. Researchers are exploring the use of Markov chains and stochastic modeling to improve the reliability and energy efficiency of edge computing systems. Additionally, there is a growing interest in developing lightweight and interpretable anomaly detection frameworks that can detect anomalies in real-time without requiring labeled data or intensive computation. Noteworthy papers in this area include: An Efficient Anomaly Detection Framework for Wireless Sensor Networks Using Markov Process, which proposes a lightweight and interpretable anomaly detection framework based on a first-order Markov chain model. Conditional Score Learning for Quickest Change Detection in Markov Transition Kernels, which introduces a score-based CUSUM procedure for detecting changes in Markov processes with unknown transition kernels.

Sources

An Efficient Anomaly Detection Framework for Wireless Sensor Networks Using Markov Process

Lightweight Latency Prediction Scheme for Edge Applications: A Rational Modelling Approach

Stochastic Modeling for Energy-Efficient Edge Infrastructure

Conditional Score Learning for Quickest Change Detection in Markov Transition Kernels

Built with on top of