scholarly journals Fully connected layer‐shared network architecture for massive MIMO CSI feedback

2022 ◽  
Author(s):  
Boyuan Zhang ◽  
Haozhen Li ◽  
Xin Liang ◽  
Xinyu Gu ◽  
Lin Zhang
Keyword(s):  
Network Architecture ◽  
Massive Mimo ◽  
Csi Feedback ◽  
Shared Network ◽  
Fully Connected

Sparsity Learning-based CSI Feedback for FDD Massive MIMO Systems

2020 ◽  
pp. 1-1
Author(s):  
Wenbo Zeng ◽  
Yigang He ◽  
Bing Li ◽  
Shudong Wang
Keyword(s):  
Massive Mimo ◽  
Mimo Systems ◽  
Csi Feedback

A Novel Compression CSI Feedback based on Deep Learning for FDD Massive MIMO Systems

2021 ◽  
Author(s):  
Yuting Wang ◽  
Yibin Zhang ◽  
Jinlong Sun ◽  
Guan Gui ◽  
Tomoaki Ohtsuki ◽  
...  
Keyword(s):  
Deep Learning ◽  
Massive Mimo ◽  
Mimo Systems ◽  
Csi Feedback

MRFNet: A Deep Learning-Based CSI Feedback Approach of Massive MIMO Systems

2021 ◽  
pp. 1-1
Author(s):  
Zhengyang Hu ◽  
Jianhua Guo ◽  
Guanzhang Liu ◽  
Hanying Zheng ◽  
Jiang Xue
Keyword(s):  
Deep Learning ◽  
Massive Mimo ◽  
Mimo Systems ◽  
Csi Feedback

scholarly journals Improved Water-Filling Power Allocation for Energy-Efficient Massive MIMO Downlink Transmissions

2017 ◽  
Vol 63 (1) ◽  
pp. 79-84
Author(s):  
M. K Noor Shahida ◽  
Rosdiadee Nordin ◽  
Mahamod Ismail
Keyword(s):  
Power Allocation ◽  
Network Architecture ◽  
Massive Mimo ◽  
Multiple Input Multiple Output ◽  
Base Station ◽  
Downlink Transmission ◽  
Allocation Algorithm ◽  
Mimo Antenna ◽  
Water Filling ◽  
Water Filling Algorithm

Abstract Energy Efficiency (EE) is becoming increasingly important for wireless communications and has caught more attention due to steadily rising energy costs and environmental concerns. Recently, a new network architecture known as Massive Multiple-Input Multiple-Output (MIMO) has been proposed with the remarkable potential to achieve huge gains in EE with simple linear processing. In this paper, a power allocation algorithm is proposed for EE to achieve the optimal EE in Massive MIMO. Based on the simplified expression, we develop a new algorithm to compute the optimal power allocation algorithm and it has been compared with the existing scheme from the previous literature. An improved water filling algorithm is proposed and embedded in the power allocation algorithm to maximize EE and Spectral Efficiency (SE). The numerical analysis of the simulation results indicates an improvement of 40% in EE and 50% in SE at the downlink transmission, compared to the other existing schemes. Furthermore, the results revealed that SE does not influence the EE enhancement after using the proposed algorithm as the number of Massive MIMO antenna at the Base Station (BS) increases.

scholarly journals A High-robustness and Low Resource-consumption Crowd Counting Model

2021 ◽  
Vol 15 ◽  
pp. 46-54
Author(s):  
Han Jia ◽  
Xuecheng Zou
Keyword(s):  
Network Architecture ◽  
Resource Consumption ◽  
Portable Devices ◽  
Estimation Errors ◽  
Crowd Counting ◽  
Density Map ◽  
Almost All ◽  
Fully Connected ◽  
Crowded Scenes ◽  
Counting Model

A major problem of counting high-density crowded scenes is the lack of flexibility and robustness exhibited by existing methods, and almost all recent state-of-the-art methods only show good performance in estimation errors and density map quality for select datasets. The biggest challenge faced by these methods is the analysis of similar features between the crowd and background, as well as overlaps between individuals. Hence, we propose a light and easy-to-train network for congestion cognition based on dilated convolution, which can exponentially enlarge the receptive field, preserve original resolution, and generate a high-quality density map. With the dilated convolutional layers, the counting accuracy can be enhanced as the feature map keeps its original resolution. By removing fully-connected layers, the network architecture becomes more concise, thereby reducing resource consumption significantly. The flexibility and robustness improvements of the proposed network compared to previous methods were validated using the variance of data size and different overlap levels of existing open source datasets. Experimental results showed that the proposed network is suitable for transfer learning on different datasets and enhances crowd counting in highly congested scenes. Therefore, the network is expected to have broader applications, for example in Internet of Things and portable devices.

Sign in / Sign up

Export Citation Format

Share Document