Signal transmission and processing for millimeter-wave and terahertz communications

Abstract

Millimeter-wave (mmWave) and Terahertz (THz) communications are envisioned as a key technology for next-generation wireless networks. The objective of the thesis is to provide a design guide on signal transmission and processing for mmWave and THz communications. Specifically, the higher-mmWave and THz bands, i.e., over 60 GHz, are investigated. First of all, a suitable system, from both communication and circuit levels, should be designed. Thus, we compare two system structures, the fully-connected structure and the array-of-subarrays structure, from system power consumption, spectral efficiency, and energy efficiency. The array-of-subarrays structure shows significant advantages over the fully-connected structure in both spectral and energy efficiency. Second, to accommodate the ultra-broad bandwidth in the mmWave and THz bands, we further study the beamforming training with time-delay components for the array-of-subarrays structure. In particular, two multi-resolution time-delay codebooks with subarray coordination and hybrid processing are proposed. Then, based on the codebooks, a hierarchical beamforming training strategy is developed to enable simultaneous training for multiple users. Third, we investigate the performance limits and optimized array design for single-user mmWave and THz communications with the array-of-subarrays structure. Based on the modified Saleh-Valenzuela (S-V) channel model, we analyze the ergodic capacity with hybrid beamforming for the system and obtain an upper bound. Last but not least, we extend the study of the array-of-subarrays structure to wideband communications. To capture the distance-frequency dependent characteristics of the THz channels, we develop a hybrid beamforming scheme with distance-aware multi-carrier transmission, including analog beamforming for user grouping and interference cancellation and digital beamforming for antenna subarray selection and adaptive power allocation. Specifically, two greedy subarray selection algorithms are proposed. In summary, the proposed design in this thesis builds the physical-layer foundation for higher-mmWave and THz communications.