DETECTION OF TRAFFIC ANOMALIES IN 5G WIRELESS NETWORKS FOR AN ENERGY COMMUNICATIONS SYSTEM
Loading...
Authors
Eidson, Bryan
Subjects
machine learning
5G
microgrid
Energy Management System
EMS
Marine Corps Air Station
MCA
Long-Short Term Memory
LSTM
state vector machine
SVM
5G
microgrid
Energy Management System
EMS
Marine Corps Air Station
MCA
Long-Short Term Memory
LSTM
state vector machine
SVM
Advisors
Oriti, Giovanna
Thulasiraman, Preetha
Date of Issue
2024-12
Date
Publisher
Monterey, CA; Naval Postgraduate School
Language
Abstract
The Energy Management System (EMS) in place at Marine Corps Air Station (MCAS) Miramar is undergoing a restoration process. The aim is to establish wireless communication between the EMS facility and the backup generators and photovoltaic arrays that are distributed across the station using the Verizon 4G LTE/5G Ultra-Wideband network. This research continues previously established anomaly detection methods by utilizing an unsupervised learning algorithm known as the Long-Short Term Memory (LSTM) autoencoder. We use the LSTM on a simulated 5G data set to determine effectiveness in autonomous detection of cyber anomalies in the data. While an LSTM requires more time to detect anomalies because it references the data as a time series and not individual packets, it can perform pattern analysis and detect other attacks beyond generic distributed denial of service attacks. The 5G dataset used in this thesis includes 5G specific attacks to determine how well an LSTM autoencoder detects alternative anomalies. We show the performance of the LSTM in terms of accuracy, precision, and recall. We also compare the LSTM with a supervised state vector machine (SVM) algorithm. The SVM is used for benchmarking the LSTM performance.
Type
Thesis
Description
Series/Report No
Organization
Identifiers
NPS Report Number
Sponsors
ONR
Funding
Format
Citation
Distribution Statement
Distribution Statement A. Approved for public release: Distribution is unlimited.
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.