Title :
A CMOS Spectrum Sensor Based on Quasi-Cyclostationary Feature Detection for Cognitive Radios
Author :
Sepidband, Paria ; Entesari, Kamran
Author_Institution :
Dept. of Electr. & Comput. Eng., Texas A&M Univ., College Station, TX, USA
Abstract :
Sensitivity and sensing time are the key features for spectrum sensing in a cognitive radio (CR) which detects empty bands within VHF/UHF television broadcast bands and uses them as a secondary user. In this paper, an integrated CMOS CR spectrum sensor for a CR receiver in the 54-862 MHz band is presented. A quasi-cyclostationary feature (QCF) detector is proposed based on both energy and feature detection methods and can take advantage of both methods to reach a fast and accurate decision without the need for an analog-to-digital converter for decision making. The integrated chip has been fabricated in a standard m CMOS IBM technology and has achieved minimum detection signal-to-noise ratio of as low as -24dB and dynamic range (DR) of 39 dB.
Keywords :
CMOS integrated circuits; cognitive radio; digital video broadcasting; feature extraction; radio receivers; radio spectrum management; signal detection; CMOS CR spectrum sensor; CMOS IBM technology; CR receiver; QCF detector; VHF-UHF television broadcast bands; cognitive radios; empty band detection; energy detection method; feature detection method; integrated chip; quasicyclostationary feature detection; quasicyclostationary feature detector; secondary user; signal-to-noise ratio; size 0.18 mum; CMOS integrated circuits; Correlation; Detectors; Feature extraction; OFDM; Signal to noise ratio; Cognitive radio (CR); OFDM DVB-T signal; cyclic autocorrelation; cyclostationary feature detection; low signal-to-noise ratio; quasi-cyclostationary feature (QCF) detector; spectrum sensing;
Journal_Title :
Microwave Theory and Techniques, IEEE Transactions on
DOI :
10.1109/TMTT.2015.2496239