BLBaozhu Li

Baozhu Li

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Baozhu Li has a total of 22 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Multi-Scale Multi-Physics Imaging of Free-Space Targets With Electromagnetic and Acoustic Data

    by Zhichao Lin, Marco Salucci, Maokun Li, Baozhu Li, Andrea Massa

    A new methodology for imaging two-dimensional (2-D) free-space targets jointly processing multi-physics (MP) data collected with microwave and ultrasound sensors is presented. The proposed strategy combines reciprocal regularization constraints, which enforce structural similarity between the two underlying physics, with an iterative multi-scale (MS) meta-level inversion scheme to effectively tackle both the non-linearity and the ill-posedness of the MP inverse scattering problem (ISP) at hand. A set of representative numerical results is reported to assess, also in a comparative fashion, the imaging capabilities of the proposed MS-MP technique.


  • Open Access

    MT-GCNN: Multi-Task Learning with Gated Convolution for Multiple Transmitters Localization in Urban Scenarios

    by Wenyu Wang, Lei Zhu, Zhen Huang, Baozhu Li, Lu Yu, Kaixin Cheng

    With the advance of the Internet of things (IoT), localization is essential in varied services. In urban scenarios, multiple transmitters localization is faced with challenges such as nonline-of-sight (NLOS) propagation and limited deployment of sensors. To this end, this paper proposes the MT-GCNN (Multi-Task Gated Convolutional Neural Network), a novel multiple transmitters localization scheme based on deep multi-task learning, to learn the NLOS propagation features and achieve the localization. The multi-task learning network decomposes the problem into a coarse localization task and a fine correction task. In particular, the MT-GCNN uses an improved gated convolution module to extract features from sparse sensing data more effectively. In the training stage, a joint loss function is proposed to optimize the two branches of tasks. In the testing stage, the well-trained MT-GCNN model predicts the classified grids and corresponding biases jointly to improve the overall performance of localization. In the urban scenarios challenged by NLOS propagation and sparse deployment of sensors, numerical simulations demonstrate that the proposed MT-GCNN framework has more accurate and robust performance than other algorithms.


  • Open Access

    Traffic Steering in Large-scale Public Cloud

    by Zhangfeng Hu, Siqing Sun, Ping Yin, Yanjun Li, Qiuzheng Ren, Baozhu Li, Xiong Li

    More and more complex services composed of a series of sequentially arranged middleboxes which are mainly used to meet the requirements of advanced services such as security services, auditing services, monitoring services, personalized enterprise services, and so forth, are increasingly deployed in cloud data centers of public cloud. SFC (Service Function Chaining) is a technique that facilitates the enforcement of complex services and differentiated traffic forwarding policies, dynamically steering the traffic through an ordered list of service functions. Flow table-based traffic steering scheme is commonly adopted in SDN-enabled scenarios, which consumes too many flow entries and is unsuitable for large-scale public clouds in steering traffic between VNFs (Virtual Network Function) inside of VPC (Virtual Private Cloud). Legacy PBR (Policy-based Routing) based schemes which are widely used in traditional physical networks cannot fulfill the requirements of fully distributed routing architectures of large-scale public clouds. In this paper, we present a PBR and unsymmetrical NAT (Network Address Translation) converged scheme to structure SFC in a fully distributed routing architecture. The scheme uses distributed PBR rules to steer traffic between an ordered list of VNFs located on different nodes while performing NAT on different nodes for ingress/egress traffic of a specific flow to avoid asymmetry of packet headers which may lead to failures of communication. The proposed scheme brings no overhead in data transmission, eliminates extra configurations on each middle box of the chain, and is scalable to support the scenarios of large-scale public cloud.

  • Open Access

    Deep Learning-Based Localization with Urban Electromagnetic and Geographic Information

    by Wenyu Wang, Baozhu Li, Zhen Huang, Lei Zhu

    There is a growing demand for localization of illegal signal sources, aiming to guarantee the security of urban electromagnetic environment. The performance of traditional localization methods is limited due to the non-line-of-sight (NLOS) propagation and sparse layouts of sensors. In this paper, a deep learning-based localization method is proposed to overcome these issues in urban scenarios. Firstly, a model of electromagnetic wave propagation considered with geographic information is proposed to prepare reliable datasets for intelligent cognition of urban electromagnetic environment. Then, this paper improves an hourglass neural network which consists of downsampling and upsampling layers to learn the propagation features from sensing data. The core modules of VGG and ResNet are, respectively, utilized as feature extractors in downsampling. Moreover, this paper proposes a weighted loss function to expand the attention on position features, in order to improve the performance of localization with sparse layouts of sensors. Representative numerical results are discussed to assess the proposed method. ResNet-based extractor performs more efficiently than VGG-based extractor, and the proposed weighted loss function increases the localization accuracy by more than 50%. Additionally, the established geographic model supports qualitative and quantitative evaluation of the robustness with varied degree of NLOS propagation. Compared with other deep learning-based algorithms, the proposed method presents the more robust and superior performance under severe NLOS propagation and sparse sensing conditions.


  • Open Access

    Deep Learning-Based Multiple Co-Channel Sources Localization Using Bernoulli Heatmap

    by Meiyan Lin, Yonghui Huang, Baozhu Li, Zhen Huang, Zihan Zhang, Wenjie Zhao

    Multiple sources localization (MSL) has received considerable attention in scenarios of commercial, industrial, and defense areas. In this paper, a novel deep learning-based approach with observations of received signal strength (RSS) is proposed for the localization of multiple co-channel sources. The proposed method, named MSLocNet, formulates the MSL problem as a Bernoulli heatmap regression problem, solved by a fully convolutional network (FCN). The proposed MSLocNet enables simultaneous localization of variable numbers of sources, and exhibits better localization performance. Simulations, under complex environments with shadow fading, are conducted to validate the improved localization accuracy of the proposed method over other benchmark schemes. Moreover, experiments are carried out in a real environment to verify the feasibility of the proposed method.


  • Compact Dual-Passband Three-Dimensional FSS with Good Angular Stability and Both-Side Fast Roll-Off Characteristics

    by Zhengyong Yu, Baozhu Li, Shenggao Ding, Wanchun Tang

    A compact dual-passband three-dimensional (3D) frequency selective surface (FSS) is proposed based on multiple square coaxial waveguides (SCWs), which exhibits good angular stability and both-side fast roll-off characteristics. The unit cell of the proposed 3D FSS is composed of one parallel plate waveguide (PPW) propagation path and two SCW propagation paths. By etching a centered annular slot, each SCW path forms two identical short SCWs. Each short SCW inherently generates one square slot resonance. In each SCW path, on the account of electromagnetic coupling between two square slot resonators provided by two short SCWs, the square slot resonant mode will split into even-/odd-resonant modes. Accordingly, each SCW path can provide a flat second-order passband with two transmission poles. Due to the reflection and out of phase of electromagnetic waves, four transmission zeros located at both sides of the passbands are introduced for high frequency selectivity, realizing both-side fast roll-off performances. In order to explain the operating principle, the electric-field distributions at transmission-zero/pole frequencies are investigated. Finally, an FSS prototype is fabricated and measured, and the results exhibit good angular stability for both TE and TM polarizations under incident angles from 0° to 60°. In addition, the proposed 3D FSS has a compact unit cell.