부산대학교 도서관

The coherent optical processing technique, referred to as k -space tomography, enables simultaneous detection of radio waves’ frequency and angle of arrival. As with the majority of signals, the spatial and spectral distribution of RF signals in the electromagnetic environment is sparse in nature. This inherent characteristic provides an ideal opportunity to merge this imaging modality with compressive sensing techniques. Compressive sensing has set a foundation for data acquisition of sparse signals from a small amount of data. By adapting specific aspects from this methodology, and applying them to k -space tomography, compressive k -space tomography is developed. Fundamental performance limits of this compressive sensing application are defined through numerical simulations, and these responses are subsequently utilized in experimental spatial-spectrum utilization mapping for impending 5G bands between 28 and 40 GHz. As a result, reconstruction data size and processing time are reduced consistently by a factor of over 100, without loss of information, for sufficiently sparse RF scenes.