Ji-Woong Choi

Hanyoung Park, Homin Jang, Ji-Woong Choi

In this paper, we propose a low-complexity blind estimator for the average noise power, average signal power, and signal-to-noise ratio (SNR) in millimeter-wave (mmWave) massive multi-antenna uplink systems. In particular, the proposed method is designed to operate using only a single received signal sample, without relying on pilot signals, iterative optimization, or multiple observations, and without requiring prior knowledge of the transmitted signal. By exploiting the inherent sparsity of mmWave channels in the beamspace domain, the estimator identifies noise-dominant components through a sorting-based procedure combined with a finite-difference criterion. This separation is further supported by the order statistics of noise power under Gaussian assumptions, enabling statistically grounded discrimination between signal and noise elements. The average noise power is estimated from the identified noise-only components, and the signal power and SNR are subsequently obtained through simple arithmetic operations. The proposed algorithm achieves low computational complexity and is well-suited for real-time implementation. To demonstrate its practical feasibility, a hardware-efficient very large-scale integration (VLSI) architecture is developed and implemented on a AMD-Xilinx Kintex UltraScale+ KCU116 Evaluation Kit, with corresponding field-programmable gate array (FPGA) results provided. The implementation exhibits low latency and sublinear scaling of hardware resource utilization with respect to the number of antennas, and enables parameter estimation within a duration shorter than a single symbol of conventional wireless systems. Simulation results verify that the proposed estimator achieves high estimation accuracy compared to existing single-sample-based methods.

Hanyoung Park, Eunho Kim, Ji-Woong Choi

In this paper, we propose a low-complexity beamspace channel denoising algorithm for millimeter-wave (mmWave) massive multi-input multi-output (MIMO) systems with low-resolution analog-to-digital converters (ADCs). The proposed method exploits the inherent sparsity of mmWave channels in the beamspace domain and formulates the denoising problem as a Bayesian binary hypothesis testing under a Bernoulli-complex Gaussian prior. To capture the distortion induced by low-resolution ADCs in a complexity-efficient manner, thermal noise and quantization noise are jointly modeled as a composite noise. Based on this modeling, a closed-form threshold value and a hard-thresholding-based denoising rule are derived to distinguish signal-dominant and noise-dominant components. The resulting algorithm avoids computationally intensive operations such as matrix inversion, iterative optimization, and parameter searching, and achieves near-linear computational complexity with respect to the number of antennas. Furthermore, a hardware-efficient very large-scale integration (VLSI) architecture is developed to enable practical deployment of the proposed algorithm, and is implemented on an AMD-Xilinx Kintex UltraScale+ KCU116 FPGA platform. The design incorporates hardware-aware simplifications and an efficient processing structure, leading to significantly lower latency and reduced hardware resource utilization compared to existing hardware implementations, along with sublinear scaling as the number of antennas increases. Extensive simulation results demonstrate that the proposed method achieves performance comparable to computationally intensive existing approaches while significantly reducing computational complexity.

Dayoung Kang, JongWon Kim, Jiho Park et al.

Public olfaction datasets are small and fragmented across single molecules and mixtures, limiting learning of generalizable odor representations. Recent works either learn single-molecule embeddings or address mixtures via similarity or pairwise label prediction, leaving representations separate and unaligned. In this work, we propose AROMMA, a framework that learns a unified embedding space for single molecules and two-molecule mixtures. Each molecule is encoded by a chemical foundation model and the mixtures are composed by an attention-based aggregator, ensuring both permutation invariance and asymmetric molecular interactions. We further align odor descriptor sets using knowledge distillation and class-aware pseudo-labeling to enrich missing mixture annotations. AROMMA achieves state-of-the-art performance in both single-molecule and molecule-pair datasets, with up to 19.1% AUROC improvement, demonstrating a robust generalization in two domains.