| Popis: |
This paper proposes a hybrid wideband millimeter-wave (mm-Wave) beamforming technique for achievable rate maximization in single-user, multi-relay, multi-input multi-output (MIMO) systems. The underlying network architecture is of great practical importance from the perspective of hybrid precoding for enhancing coverage area, transmission range, link quality, and spectral efficiency. In hybrid transceivers, the analog processing component in the radio frequency (RF) section of the system essentially needs to follow an element-wise constant amplitude constraint, which makes the optimization problem non-convex and hence mathematically intractable. Therefore, it is quite a challenging task to obtain the global optimal solution. Moreover, the common analog precoder and combiner are required for different frequency-selective channels in broadband systems, which adds another layer of design complexity. To address this problem, the proposed algorithm transforms the original complicated optimization problem into a relatively simple form. Leveraging this, the complexity of the optimization problem is reduced through decoupled design using the norm preservation property of the unitary transform. Moreover, the analog beamforming design is decomposed into two sub-problems by exploiting statistical independence among the first hop and second hop channels. Each sub-problem is solved to harvest the array gain through RF precoding and combining using an alternative optimization framework. Finally, the frequency-selective baseband processing components are obtained by minimizing interference among transmitted data streams. The spectral efficiency of the proposed scheme is evaluated by changing system parameters under imperfect channel and phase noise (PN). Simulation results demonstrate that the achieved performance is close to full-complexity precoding and outperforms traditional algorithms. |
