Performance Analysis of the Circular Folding Cooperative Power Spectral Density Split Cancellation Algorithm for Spectrum Sensing Under Errors at the Quantized Report Channel

Abstract

The circular folding cooperative power spectral density split cancellation (CF-CPSC) algorithm was recently proposed for detecting idle bands in centralized cooperative spectrum sensing for cognitive radio (CR) applications. This algorithm has low complexity and is robust under nonuniform noise. This paper analyzes the performance of the CF-CPSC under quantized and erroneous report channel transmissions. Two approaches has been investigated regarding the CF-CPSC test statistic: i) it is completely computed at the fusion center (FC), and ii) it is partially calculated at each CR for saving report channel resources. In both cases, the information to be reported by the CRs are uniformly quantized and then submitted to errors. Results show that the CF-CPSC is also robust to the quantization and report errors. The second approach is more sensitive to the quantization and report channel errors, but less channel resources are used in comparison with the first approach. The smaller resource usage is achieved by the second approach even with error correcting codes, which increases the amount of data through the report channel. The resource usage achieved by the first approach is larger, even with no error correcting codes.