Search

In frequency division duplex mode, the user equipment sends downlink channel state information (CSI) to the base station for feedback. However, high-dimensional CSI can cause a large feedback overhead. Although convolution-based deep learning methods help compress and recover CSI, the redundant features among the CSI feature maps extracted by the convolution operator cause efficiency decay. This letter applies the Ghost module, which generates feature maps from a handful of primary features, to reduce redundancy and improve feedback efficiency. Additionally, a lightweight neural network, called GCRNet, is proposed based on the Ghost module. Compared with CLNet, GCRNet reduces complexity by an average of 22.15% while maintaining comparable performance.

Channel state information (CSI) plays a critical role in achieving the potential benefits of massive multiple input multiple output (MIMO) systems. In frequency division duplex (FDD) massive MIMO systems, the base station (BS) relies on sustained and accurate CSI feedback from users. However, due to the large number of antennas and users being served in massive MIMO systems, feedback overhead can become a bottleneck. In this paper, we propose a model-driven deep learning method for CSI feedback, called learnable optimization and regularization algorithm (LORA). Instead of using $l_{1}$ -norm as the regularization term, LORA introduces a learnable regularization module that adapts to characteristics of CSI automatically. The conventional Iterative Shrinkage-Thresholding Algorithm (ISTA) is unfolded into a neural network, which can learn both the optimization process and the regularization term by end-to-end training. We show that LORA improves the CSI feedback accuracy and speed. Besides, a novel learnable quantization method and the corresponding training scheme are proposed, and it is shown that LORA can operate successfully at different bit rates, providing flexibility in terms of the CSI feedback overhead. Various realistic scenarios are considered to demonstrate the effectiveness and robustness of LORA through numerical simulations.

In order to break through the development bottleneck of modern wireless communication networks, a critical issue is the out-of-date channel state information (CSI) in high mobility scenarios. In general, non-stationary CSI has statistical properties which vary with time, implying that the data distribution changes continuously over time. This temporal distribution shift behavior undermines the accurate channel prediction and it is still an open problem in the related literature. In this paper, a hypernetwork based framework is proposed for non-stationary channel prediction. The framework aims to dynamically update the neural network (NN) parameters as the wireless channel changes to automatically adapt to various input CSI distributions. Based on this framework, we focus on low-complexity hypernetwork design and present a deep learning (DL) based channel prediction method, termed as LPCNet, which improves the CSI prediction accuracy with acceptable complexity. Moreover, to maximize the achievable downlink spectral efficiency (SE), a joint channel prediction and beamforming (BF) method is developed, termed as JLPCNet, which seeks to predict the BF vector. Our numerical results showcase the effectiveness and flexibility of the proposed framework, and demonstrate the superior performance of LPCNet and JLPCNet in various scenarios for fixed and varying user speeds.

In massive multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) systems, a challenging problem is how to predict channel state information (CSI) (i.e., channel prediction) accurately in mobility scenarios. However, a practical obstacle is caused by CSI non-stationary and nonlinear dynamics in temporal domain. In this paper, we propose a spatio-temporal neural network (STNN) to achieve better performance by carefully taking into account the spatio-temporal characteristics of CSI. Specifically, STNN uses its encoder and decoder modules to capture the spatial correlation and temporal dependence of CSI. Further, the differencing-attention module is designed to deal with the non-stationary and nonlinear temporal dynamics and realize adaptive feature refinement for more accurate multi-step prediction. Additionally, an advanced training scheme is adopted to reduce the discrepancy between STNN training and testing. Evaluated on a realistic channel model with enhanced mobility and spherical waves, experimental results show that STNN can effectively improve the accuracy of prediction and perform well with respect to different signal to noise ratios (SNRs). Visualization and testing for unit root illustrate STNN is able to learn CSI time-varying patterns by alleviating series non-stationarity.

Lung adenocarcinoma (LUAD) is the leading cause of cancer-related deaths worldwide. Traditional RNA sequencing data fails to detect the exact cellular and molecular changes in tumor cells as they make up only a small proportion of tumor tissue. 10× genomics single-cell RNA sequencing (10× scRNA-seq) and gene expression data of LUAD patients was obtained from the Department of Thoracic Surgery, Zhongshan Hospital, Fudan University, ArrayExpress, TCGA, and GEO databases. Differentially expressed genes (DEGs) were identified in LUAD and alveolar cells (DEGs-scRNA-cancer_cell), tumor- and normal tissue-derived cells (DEGs-scRNA-sample), and normal and LUAD patients (DEGs-Bulk). Flow cytometry and qRT-PCR were performed to validate the significantly differentially expressed ligand–receptor pairs. We selected 159,219 cells and 594 samples in the scRNA-seq data and traditional RNA sequencing, respectively. A total of 1042 DEGs-scRNA-cancer_cell, 788 DEGs-scRNA-sample, and 2510 DEGs-Bulk were identified in this study. We also identified 57 DEGs that were only detected in DEGs-scRNA-cancer_cell (only-DEGs-scRNA-cancer_cell). To explore the relationship between only-DEGs-scRNA-cancer_cell and survival in LUAD, 14 and 22 only-DEGs-scRNA-cancer_cell, which were closely related with survival in TCGA and GEO cohorts were identified. Functional enrichment analyses showed these DEGs-scRNA-cancer_cells were mainly related to cell proliferation and immunoregulation. Our study detected and compared DEGs at different levels and revealed genes that may regulate tumor development. Our results provide a potential new protocol to determine the contribution of DEGs to cancer progression and to help identify potential therapeutic targets.

ABSTRACTThe main objective of this study is to develop a two-stage stochastic programming framework for lot-sizing and scheduling the production activities at a kitting facility to support a manufacturing plant. The novelty of the study lies in modeling multiple uncertainties using two-stage stochastic programming to solve a kitting specific production planning problem in a manufacturing setting. The demand for the kits and the yield of the kitting workers are the two sources of uncertainties considered in this study. The first-stage decisions include the baseline production schedule and the workforce requirement, while the second stage makes recourse decisions on overtime production. The proposed decision-making framework is validated on a multi-period, multi-product case study involving a kitting facility supporting a manufacturing plant producing braking equipment. The main conclusion of the study is that uncertainties have significant impacts on kitting planning decisions and that the proposed two-stage stochastic programming model was robust in determining optimal production plans under uncertainty.

Channel estimation (CE) is one of the crucial and fundamental elements of signal processing, especially considering the requirement of high accuracy in future wireless communication systems. Most traditional CE algorithms are explored under the assumption of Gaussian white noise, which limits the algorithms performance in real wireless communication situations. In this work, a novel self-adaptive CE algorithm based on the maximum entropy principle (MEP) was studied, which analyzes the statistical components of an arbitrary noise environment. In addition, an MEP channel-based signal estimation algorithm was studied. Furthermore, the statistical characteristics of channels were considered the regularization terms in the objective function for providing prior information and further increasing the accuracy. It was found that the proposed algorithm not only provides accurate CE but also reduces pilot consumption by using estimated signal data as pseudo pilots. The superior features of the proposed method concerning CE accuracy, pilot consumption, and robustness were confirmed through Monte Carlo simulations.

Production planning and scheduling focus on efficient use of resources and are widely used in the manufacturing industry, especially when the system operates in an uncertain environment. The goal of this paper is to provide a two-stage stochastic programming framework for a multi-period, multi-product, lot-sizing and scheduling problem considering uncertainties in both demand and the quality of raw materials. The objectives are to determine the number of units to be produced and the production sequence so that the total production costs are minimised. The decisions made in the first stage include the basic production plan along with the production quantities and sequences, which are later updated with recourse decisions on overtime production made in the second-stage. To demonstrate the proposed decision-making framework, a case study for a manufacturing facility producing braking equipment for the automotive industry was conducted. The results show that the stochastic model is more effective in production planning under the uncertainties considered. The managerial insights derived from this study will facilitate the decision making for determining optimal production quantities and sequences under uncertainties.