Junil Choi

Hwanjin Kim, Junil Choi, David J. Love

Accurate channel knowledge is critical in massive multiple-input multiple-output (MIMO), which motivates the use of channel prediction. Machine learning techniques for channel prediction hold much promise, but current schemes are limited in their ability to adapt to changes in the environment because they require large training overheads. To accurately predict wireless channels for new environments with reduced training overhead, we propose a fast adaptive channel prediction technique based on a meta-learning algorithm for massive MIMO communications. We exploit the model-agnostic meta-learning (MAML) algorithm to achieve quick adaptation with a small amount of labeled data. Also, to improve the prediction accuracy, we adopt the denoising process for the training data by using deep image prior (DIP). Numerical results show that the proposed MAML-based channel predictor can improve the prediction accuracy with only a few fine-tuning samples. The DIP-based denoising process gives an additional gain in channel prediction, especially in low signal-to-noise ratio regimes.

Hyun Ryu, Junil Choi

Modulation classification (MC) is the first step performed at the receiver side unless the modulation type is explicitly indicated by the transmitter. Machine learning techniques have been widely used for MC recently. In this paper, we propose a novel MC technique dubbed as Joint Equalization and Modulation Classification based on Constellation Network (EMC2-Net). Unlike prior works that considered the constellation points as an image, the proposed EMC2-Net directly uses a set of 2D constellation points to perform MC. In order to obtain clear and concrete constellation despite multipath fading channels, the proposed EMC2-Net consists of equalizer and classifier having separate and explainable roles via novel three-phase training and noise-curriculum pretraining. Numerical results with linear modulation types under different channel models show that the proposed EMC2-Net achieves the performance of state-of-the-art MC techniques with significantly less complexity.

Jihoon Cha, Hwanjin Kim, Junil Choi

In this paper, we consider people counting and localization systems exploiting channel state information (CSI) measured from commodity WiFi network interface cards (NICs). While CSI has useful information of amplitude and phase to describe signal propagation in a measurement environment of interest, CSI measurement suffers from offsets due to various uncertainties. Moreover, an uncontrollable external environment where other WiFi devices communicate each other induces interfering signals, resulting in erroneous CSI captured at a receiver. In this paper, preprocessing of CSI is first proposed for offset removal, and it guarantees low-latency operation without any filtering process. Afterwards, we design people counting and localization models based on pre-training. To be adaptive to different measurement environments, meta-learning-based people counting and localization models are also proposed. Numerical results show that the proposed meta-learning-based people counting and localization models can achieve high sensing accuracy, compared to other learning schemes that follow simple training and test procedures.

Yongjun Kim, Hyeongjun Park, Hwanjin Kim et al.

Federated learning (FL) and federated distillation (FD) are distributed learning paradigms that train UE models with enhanced privacy, each offering different trade-offs between noise robustness and learning speed. To mitigate their respective weaknesses, we propose a hybrid federated learning (HFL) framework in which each user equipment (UE) transmits either gradients or logits, and the base station (BS) selects the per-round weights of FL and FD updates. We derive convergence of HFL framework and introduce two methods to exploit degrees of freedom (DoF) in HFL, which are (i) adaptive UE clustering via Jenks optimization and (ii) adaptive weight selection via a damped Newton method. Numerical results show that HFL achieves superior test accuracy at low SNR when both DoF are exploited.

Yongjun Kim, Sejin Seo, Jihong Park et al.

In this work, we compare emergent communication (EC) built upon multi-agent deep reinforcement learning (MADRL) and language-oriented semantic communication (LSC) empowered by a pre-trained large language model (LLM) using human language. In a multi-agent remote navigation task, with multimodal input data comprising location and channel maps, it is shown that EC incurs high training cost and struggles when using multimodal data, whereas LSC yields high inference computing cost due to the LLM's large size. To address their respective bottlenecks, we propose a novel framework of language-guided EC (LEC) by guiding the EC training using LSC via knowledge distillation (KD). Simulations corroborate that LEC achieves faster travel time while avoiding areas with poor channel conditions, as well as speeding up the MADRL training convergence by up to 61.8% compared to EC.

Hwanjin Kim, Sucheol Kim, Hyeongtaek Lee et al.

This paper focuses on channel prediction techniques for massive multiple-input multiple-output (MIMO) systems. Previous channel predictors are based on theoretical channel models, which would be deviated from realistic channels. In this paper, we develop and compare a vector Kalman filter (VKF)-based channel predictor and a machine learning (ML)-based channel predictor using the realistic channels from the spatial channel model (SCM), which has been adopted in the 3GPP standard for years. First, we propose a low-complexity mobility estimator based on the spatial average using a large number of antennas in massive MIMO. The mobility estimate can be used to determine the complexity order of developed predictors. The VKF-based channel predictor developed in this paper exploits the autoregressive (AR) parameters estimated from the SCM channels based on the Yule-Walker equations. Then, the ML-based channel predictor using the linear minimum mean square error (LMMSE)-based noise pre-processed data is developed. Numerical results reveal that both channel predictors have substantial gain over the outdated channel in terms of the channel prediction accuracy and data rate. The ML-based predictor has larger overall computational complexity than the VKF-based predictor, but once trained, the operational complexity of ML-based predictor becomes smaller than that of VKF-based predictor.