Bin Li

The University of Tokyo

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Chenglin Zhao

Beijing University of Posts and Telecommunications

106

Weisi Guo

Cranfield University

Zheng Zhou

Ningbo University

Zhuangkun Wei

Durham University

Weixia Zou

Beijing University of Posts and Telecommunications

Arumugam Nallanathan

Queen Mary University of London

Wanyi Gu

Guangzhou University

Mengwei Sun

Cranfield University

Ziping Wei

Beijing University of Posts and Telecommunications

Shanguo Huang

Beijing University of Posts and Telecommunications

Cooperative Institutions

Chinese Academy of Sciences

164

Beijing University of Posts and Telecommunications

146

Tsinghua University

Harbin Institute of Technology

Beijing Institute of Technology

University of Chinese Academy of Sciences

Shanghai Jiao Tong University

Peking University

University of Science and Technology of China

Southeast University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Deep Learning Based Digital Signal Modulation Recognition

Lecture notes in electrical engineering (2015)

Junqiang Fu Chenglin Zhao Bin Li Peng Xiao

Modulation (music)

SIGNAL (programming language)

10.1007/978-3-319-08991-1_100

Cite

Citations (16)

Hamming–Luby rateless codes for molecular erasure channels

Nano Communication Networks (2019)

Zhuangkun Wei Bin Li Wenxiu Hu Weisi Guo Chenglin Zhao

Online codes

Fountain code

Raptor code

Erasure

Luby transform code

10.1016/j.nancom.2019.100280

Cite

Citations (4)

Towards All-in-one Pre-training via Maximizing Multi-modal Mutual Information

arXiv (Cornell University) (2022)

Weijie Su Xizhou Zhu Chenxin Tao Lewei Lu Bin Li

To effectively exploit the potential of large-scale models, various pre-training strategies supported by massive data from different sources are proposed, including supervised pre-training, weakly-supervised pre-training, and self-supervised pre-training. It has been proved that combining multiple pre-training strategies and data from various modalities/sources can greatly boost the training of large-scale models. However, current works adopt a multi-stage pre-training system, where the complex pipeline may increase the uncertainty and instability of the pre-training. It is thus desirable that these strategies can be integrated in a single-stage manner. In this paper, we first propose a general multi-modal mutual information formula as a unified optimization target and demonstrate that all existing approaches are special cases of our framework. Under this unified perspective, we propose an all-in-one single-stage pre-training approach, named Maximizing Multi-modal Mutual Information Pre-training (M3I Pre-training). Our approach achieves better performance than previous pre-training methods on various vision benchmarks, including ImageNet classification, COCO object detection, LVIS long-tailed object detection, and ADE20k semantic segmentation. Notably, we successfully pre-train a billion-level parameter image backbone and achieve state-of-the-art performance on various benchmarks. Code shall be released at https://github.com/OpenGVLab/M3I-Pretraining.

Code (set theory)

Modalities

10.48550/arxiv.2211.09807

Cite

Citations (2)

Analysis of atmospheric effects on the continuous variable quantum key distribution

Chinese Physics B (2022)

Tao Liu Shuo Zhao Ivan B. Djordjević Shuyu Liu Sijia Wang

Atmospheric effects have significant influence on the performance of a free-space optical continuous variable quantum key distribution (CVQKD) system. In this paper, we investigate how the transmittance, excess noise and interruption probability caused by atmospheric effects affect the secret-key rate (SKR) of the CVQKD. Three signal wavelengths, two weather conditions, two detection schemes, and two types of attacks are considered in our investigation. An expression aims at calculating the interruption probability is proposed based on the Kolmogorov spectrum model. The results show that a signal using long working wavelength can propagate much further than that of using short wavelength. Moreover, as the wavelength increases, the influence of interruption probability on the SKR becomes more significant, especially within a certain transmission distance. Therefore, interruption probability must be considered for CVQKD by using long-signal wavelengths. Furthermore, different detection schemes used by the receiver will result in different transmission distances when subjected to individual attacks and collective attacks, respectively.

Quantum key distribution

SIGNAL (programming language)

Continuous variable

10.1088/1674-1056/ac6b1f

Cite

Citations (0)

Federated Learning for Radar Gesture Recognition Based on Spike Timing-Dependent Plasticity

IEEE Transactions on Aerospace and Electronic Systems (2024)

Mengjun Zhang Bin Li Hongfu Liu Chenglin Zhao

Radar-based gesture recognition combined machine learning methods can achieve excellent performance and has been utilized in a wide variety of applications. However, large-scale radar signal processing suffers from the risk of privacy leakage and the problem of limited computing resources. In this paper, we propose a federated learning framework based on the spiking neural network to solve above limitations in radar gesture recognition. First, we build a distributed federated learning system with good privacy protection to process radar data collaboratively, which only exchanges the model parameters between clients to train the model, without the need for transferring the data itself. Second, we train the spiking neural network using the spike timing dependent plasticity with biological interpretability, which is more compatible with the biological characteristics and can significantly reduce the energy consumption. Our creative combination of low-power spiking neural network and federated learning thus address the issue of resource constraint of edge nodes in federated learning. Furthermore, we introduce the weight pruning technique to the federated learning training process to reduce the model's communication cost. We conduct experiments on the 8GHz radar gesture dataset and Google Soli dataset to test the proposed model's validity. We evaluate the model's performance across various participating devices and examine the spiking neural network's capacity to recognize gestures under independent identical distribution and non-independent identical distribution strategies in federated learning. We also analyze the energy consumption and communication cost of the proposed model, demonstrating the spike timing dependent plasticity-based algorithm is more energy efficient than the traditional backpropagation as well as the backpropagation through time algorithms.

Spike-timing-dependent plasticity

10.1109/taes.2024.3353148

Cite

Citations (0)

An Improved Energy-Efficient Clustering Algorithm Based on MECA and PEGASIS for WSNs

Jin Wang Jiayi Cao Yiquan Cao Bin Li Sungyoung Lee

Wireless sensor networks (WSNs) with one static sink often suffer from energy hole problem, which means energy consumption of certain sensors near the sink or on critical path is much faster than other sensor nodes. Finally, network partition and isolated nodes are caused. To solve this problem, we propose an Energy-Efficient Strategy Enhancing Lifetime for WSNs based on MECA and PEGASIS. This algorithm aims at minimizing and balancing energy consumption for all sensor nodes and cluster heads. We divide the network into several equal clusters and the intra-cluster routing exploits multi-hop routing to save energy. The cluster heads will connect a chain such as PEGASIS, which can avoid distant cluster head from sending data to the sink node. Simulation results show that our proposed routing algorithm consumes much less than traditional routing algorithm like LEACH.

Sink (geography)

Network partition

Routing algorithm

10.1109/cbd.2015.49

Cite

Citations (5)

Propagation Delay Characteristics Within Multi-beam Antenna Coverage of LEO Satellite Systems

International Conference on Mechanic Automation and Control Engineering (2012)

Ming Ma Xi Luan Bin Li Shubo Ren Jian Wu

In this paper a calculation model of propagation delays of satellite beams is introduced according to the analysis of propagation delay characteristic in the LEO multi-beam satellite communications system. Based on actual parameters and beam coverage patterns of the Iridium and Globalstar system, the maximum propagation delay differentials in several typical beams from the center to the edge of the satellite's coverage are calculated. Analysis of these results can provide reference for analyzing the feasibility of LEO multi-beam satellite system compatible with uplink synchronization in LTE, and thereby make theoretical preparation for further study of the satellite system compatible with LTE.

Communications satellite

10.1109/mace.2012.485

Cite

Citations (0)

A Survey about Routing Protocols with Mobile Sink for Wireless Sensor Network

International Journal of Future Generation Communication and Networking (2014)

Jin Wang Xiaoqin Yang Zhongqi Zhang Bin Li Jeong‐Uk Kim

Mobile sink node properly used in routing protocols can improve network performance. Thus we investigate the state-of-the-art mobile sink based query-based and location-based routing protocols. The latter strategy can be further classified into backbone-based and rendezvous-based routing protocols. In this paper, we first describe the main principles of the most representative routing strategies with sink mobility support, and highlight their advantages and disadvantages. Descriptions and comparisons of several typical routing protocols are given to deepen the understanding.

Sink (geography)

Mobile wireless

10.14257/ijfgcn.2014.7.5.18

Cite

Citations (19)

An Efficient ISAR Imaging Deception Jamming Method of Chirp Modulation Signal

Yanwei Zhao Yunfo Liu Bin Li Xiaoyuan Ma Wenshu Wei

Inverse Synthetic aperture radar (ISAR) imaging deception jamming method of Chirp Modulation Signal is a powerful tool in electronic war. In this paper, an Efficient All-Digital ISAR raw signal deception model for air targets is presented. The proposed procedure is based on an angle-frequency domain analysis: a proper analytical reformulation of the ISAR raw signal expression is presented. It is shown that this reformulation allows us to design a very efficient simulation scheme which applies the joint analysis method of wideband frequency-angle domain. Accordingly, the problem is solved that the target wide frequency band scattering field can't be effectively simulation in time domain and the computational load of ISAR raw signal deception is dramatically reduced.

Chirp

SIGNAL (programming language)

10.1109/nemo49486.2020.9343651

Cite

Citations (0)

Elevation angle research in three-dimension channel model using ray-tracing

Xiyu Wang Bin Li Xi Yuan Jianwu Dou Yong Li

By adopting elevation domain beamforming and spatial domain multiplexing, 3 Dimensional (3D) Multiple Input and Multiple Output (MIMO) antenna system is the key technology in providing high throughput in modern communication systems. However, the conventional channel modeling method mainly focuses on 2 dimensional (2D) MIMO channels which neglect the vertical plane. It inspires scholars to do research on the characteristics of 3D channel models' elevation parameters. In this paper, we propose an automatic measurement approach for 3D MIMO channel by using ray-tracing algorithm. This method delivers the characteristics of the elevation angle for low-rise floors in Outdoor to Indoor (O2I) scenario, which have accurately matched the classical experimental results. In addition, further investigation for high-rise floors is successfully made by this proposed methodology as well.

Elevation (ballistics)

Spatial multiplexing

10.1109/ursigass.2014.6929281

Cite

Citations (9)