Large-Scale Network Inference: Detecting the Unknown and the Intermittent

Abstract

Accurate and timely detection of network anomalies or unusual activities, be it endogenous (caused by internal malfunctioning components) or exogenous (caused by hostile attacks and intrusions), is crucial to the functionality and survivability of tactical military networks. The objective of this research is to develop general design methodologies for large-scale network inference for anomaly detection. We aim to establish fundamental limits on sample complexityin particular, the scaling behavior of sample complexity with respect to the problem size and the detection accuracyand develop efficient algorithms that achieve or approach the fundamental limits with scalable low-complexity implementations. Our emphasis is on low-complexity deterministic strategies with implementations scalable to large networks.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 04, 2018
Accession Number
AD1082697

Entities

People

  • Qing Zhao

Organizations

  • Cornell University

Tags

Communities of Interest

  • Autonomy
  • Cyber
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Anomaly Detection
  • Change Detection
  • Data Processing
  • Denial Of Service Attack
  • Detection
  • Experimental Design
  • Information Operations
  • Information Theory
  • Intrusion Detection
  • Random Walk
  • Signal Processing
  • Students
  • Teamwork
  • Universities
  • Wireless Communications

Fields of Study

  • Computer science

Readers

  • Distributed Systems and Data Platform Development
  • Neural Network Machine Learning.
  • Sensor Fusion and Tracking Systems.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms