Seunghyeong Yoo

Generalized power iteration-based precoding (GPIP) exhibits superior performance in maximizing spectral efficiency for multi-antenna communications. Increasing the number of antennas, however, makes GPIP less practical due to its cubic complexity with respect to the number of antennas in computing matrix inversion. To resolve this, we propose efficient GPIP methods with imperfect channel knowledge. Exploiting the GPIP matrix structure, we use two low-complexity matrix inversion approaches: the matrix inversion lemma-based low-dimension matrix inversion and Gauss-Seidel iteration. These approaches decrease complexity order to the square of the number of antennas. Simulations validate the computational efficiency of the proposed methods while maintaining performance.

We propose a joint secure max-min fairness precoding and antenna selection algorithm for multi-user systems constrained by limited active antennas, low-resolution quantizers, and imperfect channel knowledge. We first formulate an optimization problem which is non-convex and non-smooth. Then, we cast it as a tractable problem, allowing us to leverage the available channel knowledge through averaging spectral efficiency and use a Lagrangian method to handle the antenna selection constraint. We identify the first-order optimality condition for a precoder and solve the condition by employing a generalized power iteration method, which achieves a superior sub-optimal solution. Simulations validate the proposed algorithm.

This paper investigates full-duplex (FD) multi-user multiple-input multiple-output (MU-MIMO) system design with coarse quantization. We first analyze the impact of self-interference (SI) on quantization in FD single-input single-output systems. The analysis elucidates that the minimum required number of analog-to-digital converter (ADC) bits is logarithmically proportional to the ratio of total received power to the received power of desired signals. Motivated by this, we design a FD MIMO beamforming method that effectively manages the SI. Dividing a spectral efficiency maximization beamforming problem into two sub-problems for alternating optimization, we address the first by optimizing the precoder: obtaining a generalized eigenvalue problem from the first-order optimality condition, where the principal eigenvector is the optimal stationary solution, and adopting a power iteration method to identify this eigenvector. Subsequently, a quantization-aware minimum mean square error combiner is computed for the derived precoder. Through numerical studies, we observe that the proposed beamformer reduces the minimum required number of ADC bits for achieving higher spectral efficiency than that of half-duplex (HD) systems, compared to FD benchmarks. The overall analysis shows that, unlike with quantized HD systems, more than 6 bits are required for the ADC to fully realize the potential of the quantized FD system.

We propose a max-min fairness (MMF) precoding algorithm for a multi-user multiple-input multiple-output (MU-MIMO) downlink system with different user weights, considering the quantization effects of low-resolution digital-to-analog converters (DACs). To find a precoder, we first formulate a problem, which jointly considers MMF and quantization. This joint problem presents some challenges that make finding a direct solution infeasible: non-smoothness and non-convexity. To address the non-smoothness, we use LogSumExp method to approximate the problem into a smooth problem. Subsequently, we derive the first-order optimality condition and cast it as a nonlinear eigenvalue problem. Lastly, we employ a generalized power iteration method to determine the superior stationary point. Simulation result demonstrates the superiority of our MMF precoding algorithm.

Satellite communications offer the potential of ensuring global coverage with a substantial capacity. In this paper, we examine the coverage performance of satellite networks by considering a distance-dependent line-of-sight (LOS) and non-LOS (NLOS) channel propagation probability, which takes into account the shadowing effect. Building upon the stochastic geometry-based network analysis used for terrestrial networks, we employ a Poisson point process to model both the satellite network and its users. Through this approach, we derive a theoretical expression for the coverage probability with beam gain under a nearest satellite association rule, providing an analytical understanding of the satellite network. The accuracy of the derived expression is confirmed through simulation results. The derived expression encompasses various network parameters such as satellite density, altitude, channel fading, pathloss, LOS probability, and beam gain, offering valuable insights into satellite networks.

Underwater acoustic (UWA) channel is one of the common systems for wireless underwater communication. However, the underwater environment poses significant challenges to reliable and efficient communication. This paper is a survey on channel modeling for UWA communications. We investigate various UWA channel modeling approaches that incorporate both large- and small-scale fading effects and provides a comprehensive summary of the key findings.

We consider a full-duplex (FD) multi-user multiple-input multiple-output (MU-MIMO) system with low-resolution quantizers at an access point (AP). In the considered FD system, there are main bottlenecks: self-interference (SI), co-channel interference (CCI), and quantization errors. In this paper, we propose a novel precoding and combining method to maximize the sum spectral efficiency (SE) by incorporating the effect of the quantization errors as well as the SI and CCI. Since the beamformers are intertwined with the quantization errors, SI, and CCI, it is highly challenging to solve the sum SE maximization problem. To address the challenges, we convert the problem into the Rayleigh quotient form. Then, we derive the first-order optimality condition with interpreting it as a generalized eigenvalue problem by leveraging the principle of the Rayleigh quotient problem. Accordingly, we adopt a power iteration method for identifying the leading eigenvector: the best local optimal precoding solution. Consequently, we propose an alternating algorithm to jointly optimize the precoder and combiner. Simulations validate the proposed algorithm.