In massive multiple-input multiple-output (MIMO) system, user equipment (UE) needs to send downlink channel state information (CSI) back base station (BS). However, the feedback becomes expensive with growing complexity of CSI in MIMO system. Recently, deep learning (DL) approaches are used improve reconstruction efficiency feedback. this paper, a novel network named CRNet is proposed achieve better performance via extracting features on multiple resolutions. An advanced training scheme that...

Massive multiple-input multiple-output (MIMO) is one of the key techniques to achieve better spectrum and energy efficiency in 5G system. The channel state information (CSI) needs be fed back from user equipment base station frequency division duplexing (FDD) mode. However, overhead direct feedback unacceptable due large antenna array massive MIMO Recently, deep learning widely adopted compressed CSI task proved effective. In this paper, a novel network named aggregated reconstruction...

In massive multiple-input multiple-output (MIMO) system, channel state information (CSI) is essential for the base station (BS) to achieve high performance gain. The CSI matrix needs be estimated and fed back from user equipment (UE) in frequency division duplexing (FDD) mode. Recently, deep learning widely used compression reduce feedback overhead. However, applying neural network brings extra memory computation cost, which non-negligible especially resource limited UE. this letter, a novel...

Superpixel is generated by automatically clustering pixels in an image into hundreds of compact partitions, which widely used to perceive the object contours for its excel-lent contour adherence. Although some works use Convolution Neural Network (CNN) generate high-quality superpixel, we challenge design principles these net-works, specifically their dependence on manual labels and excess computation resources, limits flexibility compared with traditional unsupervised segmentation methods....

In massive multiple-input multiple-output (MIMO) systems under the frequency division duplexing (FDD) mode, user equipment (UE) needs to feed channel state information (CSI) back base station (BS). Though deep learning approaches have made a hit in CSI feedback problem, whether they can remain excellent actual environments be further investigated. this letter, we point out that real-time dataset application often has domain gap from training caused by time delay. To bridge gap, propose...

The channel state information (CSI) needs to be fed back from the user equipment (UE) base station (BS) in frequency division duplexing (FDD) multiple-input multiple-output (MIMO) system. Recently, neural networks are widely applied CSI compressed feedback since original overhead is too large for massive MIMO Notably, lightweight attract special attention due their practicality of deployment. However, accuracy likely harmed by network compression. In this letter, a cost free distillation...

For massive multiple-input multiple-output (MIMO) systems, downlink channel state information (CSI) compression and feedback are crucial for enhancing system performance. Deep learning (DL)-based methods have been designed proven to perform well in this task. However, the distribution of CSI real-world communication systems may differ from one observed during model training, which can undermine effectiveness DL-based due their limited generalization ability. Several proposed facilitate...

Automatic modulation recognition (AMR) has received widespread attention as a crucial aspect of non-cooperative communication. Despite this, large carrier frequency offsets (CFOs) and sample rate (SROs) caused by inaccurate parameter estimation at the receiver are harmful to accuracy, which is still be addressed. In this letter, we focus on intelligent tasks under such offsets. A novel transformer-based method named TransGroupNet designed that can extract deep features signals from...

In massive multiple-input multiple-output (MIMO) systems, the user equipment (UE) needs to feed channel state information (CSI) back base station (BS) for following beamforming. But large scale of antennas in MIMO systems causes huge feedback overhead. Deep learning (DL) based methods can compress CSI at UE and recover it BS, which reduces cost significantly. compressed must be quantized into bit streams transmission. this paper, we propose an adaptor-assisted quantization strategy bit-level...

Recently, a deep learning based error correction coding scheme is proposed to compensate for the severe distortion due one-bit quantization. However, fully-connected (FC) layers aided autoencoder too heavy, resulting in high storage cost. In this letter, novel convolutional named ECCNet introduced lighten scheme. Additionally, soft quantization function overcome gradient mismatch. The squeeze and excitation (SE) block applied further performance boosting. Simulations show that BER of...

In massive multiple-input multiple-output (MIMO) system, user equipment (UE) needs to send downlink channel state information (CSI) back base station (BS). However, the feedback becomes expensive with growing complexity of CSI in MIMO system. Recently, deep learning (DL) approaches are used improve reconstruction efficiency feedback. this paper, a novel network named CRNet is proposed achieve better performance via extracting features on multiple resolutions. An advanced training scheme that...

Hybrid precoding is introduced to millimeter wave (mmWave) multiple-input multiple-output (MIMO) System in order reduce the number of radio frequency (RF) chains. In traditional hybrid scheme, maximum independent data streams restricted by RF-chains, which limits achievable spectral efficiency (SE) rate. To further improve SE rate, generalized spatial modulation (GenSM) structure considered. However, current gradient ascent iteration (GAI) algorithm GenSM aided too complicated when antenna...

Recent work has shown that Binarized Neural Networks (BNNs) are able to greatly reduce computational costs and memory footprints, facilitating model deployment on resource-constrained devices. However, in comparison their full-precision counterparts, BNNs suffer from severe accuracy degradation. Research aiming this gap thus far largely focused specific network architectures with few or no 1 × convolutional layers, for which standard binarization methods do not well. Because convolutions...

In massive multiple-input multiple-output (MIMO) system, channel state information (CSI) is essential for the base station to achieve high performance gain. Recently, deep learning widely used in CSI compression fight against growing feedback overhead brought by MIMO frequency division duplexing system. However, applying neural network brings extra memory and computation cost, which non-negligible especially resource limited user equipment (UE). this paper, a novel binarization aided named...

In frequency division duplexing (FDD) mode, it is necessary to send the channel state information (CSI) from user equipment base station. The downlink CSI essential for massive multiple-input multiple-output (MIMO) system acquire potential gain. Recently, deep learning widely adopted MIMO feedback task and proved be effective compared with traditional compressed sensing methods. this paper, a novel network named ACRNet designed boost performance aggregation parametric RuLU activation....

In massive multiple-input multiple-output (MIMO) systems under the frequency division duplexing (FDD) mode, user equipment (UE) needs to feed channel state information (CSI) back base station (BS). Though deep learning approaches have made a hit in CSI feedback problem, whether they can remain excellent actual environments be further investigated. this letter, we point out that real-time dataset application often has domain gap from training caused by time delay. To bridge gap, propose...

Recently, deep learning aided methods have been developed for error correction coding with quantitative constraints. However, previous studies only focus on additive white Gaussian noise (AWGN) channels, which is not sufficient actual communication environments. In this paper, we propose a novel autoencoder scheme low-resolution reception under time-varying channels. Based the symbol extension of proposed and faster-than-Nyquist (FTN) technology, pilot-free transmission can be realized...