Practical Implementation of Adaptive Threshold Energy Detection using Software Defined Radio

Spectrum awareness is a fundamental characteristic of cognitive radio, and spectrum sensing is the local procedure by which the cognitive radio gains knowledge of spectrum users. Energy detection is the most widely used and widely studied form of spectrum sensing. Much of the information about energ...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on aerospace and electronic systems 2021-04, Vol.57 (2), p.1227-1241
Main Authors: Lipski, Michael V., Kompella, Sastry, Narayanan, Ram M.
Format: Article
Language:English
Subjects:
Channel estimation
Cognitive radio
Detectors
Energy
energy detection
Hardware
Parameters
Radio spectra
Sensors
Signal to noise ratio
Software
software defined radio
Software radio
Spectrum allocation
spectrum sensing
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Spectrum awareness is a fundamental characteristic of cognitive radio, and spectrum sensing is the local procedure by which the cognitive radio gains knowledge of spectrum users. Energy detection is the most widely used and widely studied form of spectrum sensing. Much of the information about energy detector performance comes from theory and simulation rather than experimental data. We acknowledge a need for empirical data that can be used to evaluate a hardware energy detector and establish expectations on the differences in performance when comparing a hardware energy detector with simulations. In this article, we build and test a real-time, adaptive threshold energy detector using a USRP software-defined radio (SDR). While several groups have built energy detectors using SDRs, we found that there is still a lack of data on the parameters and performance characteristics of SDR-based energy detectors. Our work covers in detail the construction of the SDR energy detector and includes specific hardware and software parameters as well as several practical considerations. We discuss the procedure used to benchmark the energy detector and include experimental results that show how several implementation parameters affect the detector performance. Our work also explores the use of moving average windows to formulate the detection statistic and focuses on the importance of the length of the window as well as the shape of the window.
ISSN:0018-9251
1557-9603
DOI:10.1109/TAES.2020.3040059