Customized Millimeter Wave Channel Model for Enhancement of Next-Generation UAV-Aided Internet of Things Networks.

The success of next-generation Internet of Things (IoT) applications could be boosted with state-of-the-art communication technologies, including the operation of millimeter-wave (mmWave) bands and the implementation of three-dimensional (3D) networks. With some access points (APs) mounted on unmanned aerial vehicles (UAVs), the probability of line-of-sight (LoS) connectivity to IoT nodes could be augmented to address the high path loss at mmWave bands. Nevertheless, system optimization is essential to maintaining reliable communication in 3D IoT networks, particularly in dense urban areas with elevated buildings. This research adopts the implementation of a geometry-based stochastic channel model. The model customizes the standard clustered delay line (CDL) channel profile based on the environmental geometry of the site to obtain realistic performance and optimize system design. Simulation validation is conducted based on the actual maps of highly dense urban areas to demonstrate that the proposed approach is comprehensive. The results reveal that the use of standard channel models in the analysis introduces errors in the channel quality indicator (CQI) that can exceed 50% due to the effect of the environmental geometry on the channel profile. The results also quantify accuracy improvements in the wireless channel and network performance in terms of the CQI and downlink (DL) throughput.

Design of Hybrid Beamforming System Based on Practical Circuit Parameter of 6-Bit Millimeter-Wave Phase Shifters.

This paper addresses the design of a hybrid beamforming system considering the circuit parameter of six-bit millimeter-wave phase shifters based on the process design kit. The phase shifter design adopts 45 nm CMOS silicon on insulator (SOI) technology at 28-GHz. Various circuit topologies are utilized, and in particular, a design is presented based on switched LC components, connected in a cascode manner. The phase shifter configuration is connected in a cascading manner to get the 6-bit phase controls. Six different phase shifters are obtained, which are 180°, 90°, 45°, 22.5°, 11.25°, and 5.6°, with a minimum number of LC components. The circuit parameters of the designed phase shifters are then incorporated in a simulation model of hybrid beamforming for a multiuser MIMO system. The number of OFDM data symbols used in the simulation is ten for eight users, 16 QAM modulation schemes, -25 dB SNR, 120 simulation runs, and around 170 h runtime. Simulation results are obtained considering four and eight users, assuming accurate technology-based models of RFIC components of the phase shifter as well as ideal phase shifter parameters. The results indicate that the performance of the multiuser MIMO system is affected by the accuracy level of the phase shifter RF component models. The outcomes also reveal the performance tradeoff based on user data streams and the number of BS antennas. By optimizing the amount of parallel data streams per user, higher data transmission rates are achieved, while maintaining acceptable error vector magnitude (EVM) values. In addition, stochastic analysis is conducted to investigate the distribution of the RMS EVM. The outcomes show that the best fitting of RMS EVM distribution of the actual and ideal phase shifters agreed with the log-logistic and logistic distributions, respectively. The obtained (mean, variance) values of the actual phase shifters based on accurate library models are (46.997, 481.36), and for ideal components the values are (36.47, 10.44).

Identifying structured light modes in a desert environment using machine learning algorithms.

The unique orthogonal shapes of structured light beams have attracted researchers to use as information carriers. Structured light-based free space optical communication is subject to atmospheric propagation effects such as rain, fog, and rain, which complicate the mode demultiplexing process using conventional technology. In this context, we experimentally investigate the detection of Laguerre Gaussian and Hermite Gaussian beams under dust storm conditions using machine learning algorithms. Different algorithms are employed to detect various structured light encoding schemes including the use of a convolutional neural network (CNN), support vector machine, and k-nearest neighbor. We report an identification accuracy of 99% under a visibility level of 9 m. The CNN approach is further used to estimate the visibility range of a dusty communication channel.