Heterogeneous Rank Beamforming for Industrial Communication and Sensing

Wireless communication has become an integral part of modern industrial environments, enabling real-time data exchange between pieces of industrial equipment, as well as remote control and maintenance by connecting the machines to the operators. The use of wireless communication in industrial environments has several advantages, such as reduced costs, improved productivity, and increased flexibility. However, it also poses several challenges due to the presence of obstacles, interference, and other environmental factors that can degrade the quality of wireless communication. With the increasing demand for higher data rates and reliable communication in such challenging environments, wireless communication technology has evolved significantly over the years. In recent years however, due to the diffusion of Line of Sight (LoS) oriented analog beamforming strategies, the reliability of wireless systems has degraded, and their ability to exploit the rich multipath propagation of such environments has been severely limited. On the other hand, due to the large number of antennas needed to overcome the path loss of millimeter wave (mmWave), the choice of moving towards analog beamforming is motivated by the complexity and power consumption of the devices might seem inevitable. To overcome this issue, in this thesis we propose a novel hardware architecture that allows the trade-off between bandwidth and rank of the channel. Such architecture is slightly more complex than a classical hybrid beamforming architecture but still significantly simpler, cheaper, and low power compared to fully digital beamforming. This aims to overcome the limitations of analog beamforming in at least part of the bandwidth with a limited increase in costs. We analyze the performance of Proportional Fairness (PF) resource allocation for multiple users equipped with such architecture, showing that in a multi-user environment, a large part of the communication can be performed exploiting digital beamforming, despite the users not operating with digital beamforming on the full band. We study the performance of Maximum Ratio Combining (MRC) as a robust analog beamforming beam based on digital beamforming information. Such method is particularly useful in combination with the resource allocation results mentioned above, as it can reliably fill in any gap that is not allocated with digital beamforming. Moreover, it can also reliably detect and decode control signaling to avoid unwanted disconnections from the network. Finally, we propose a method to perform multipath decomposition and time and angle of arrival estimation on the received signal that can jointly exploit the high rank and high bandwidth provided by the proposed architecture. The proposed method also has significantly lower complexity compared to existing methods, as it decomposes the 2D parameter estimation in a sequence of 1D estimations.