The ability to dynamically discover portions of unused radio spectrum (spectrum holes) is an important ability of cognitive radio systems. Spectrum holes present a potential opportunity for wireless communication. Detection of holes and signals allows cognitive radios to dynamically access and share the spectrum with minimal interference. This work steps through the design, implementation, and analysis of a spectrum holes detector for cognitive radios. Energy detection and cyclostationary detection algorithms for detecting spectrum holes are compared through computer simulations. Ultimately an energy detection algorithm is proposed which performs better than the cyclostationary detection algorithm and requires no a-priori knowledge of noise power. The energy detection algorithm is implemented on the bladeRF x115 software-defined radio for wideband detection, leveraging on-board FPGA hardware and field-programmable analog hardware to scan a gigahertz-order range of frequencies and discover spectrum holes in real time. Resource utilization and requirements of the implementation are analyzed, and a utilization of 8.8% of the FPGA's logic resources is reported. Experiments are performed on the implementation to measure its detection performance and demonstrate its ability to detect holes over a wide bandwidth with reasonable latency.
Library of Congress Subject Headings
Cognitive radio networks--Data processing; Signal detection
Computer Engineering (MS)
Department, Program, or Center
Computer Engineering (KGCOE)
Frasch, Ian J., "Algorithmic Framework and Implementation of Spectrum Holes Detection for Cognitive Radios" (2017). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus