MmWave Codebook Selection via Multinomial Thompson Sampling
We study the codebook selection problem through a multi-armed bandit (MAB) formulation in mmWave networks with rapidly-varying channels. We develop multiple novel Thompson Sampling-based algorithms for our setting given different codebook structures with theoretical guarantees on regret.
Publication: Yi Zhang, Soumya Basu, Sanjay Shakkottai, and Robert W. Heath Jr. 2021. MmWave Codebook Selection in Rapidly-Varying Channels via Multinomial Thompson Sampling. In Mobihoc '21: The Twenty-Second ACM International Symposium on Mobile Ad Hoc Networking and Computing, July 26, 2021, Shanghai, China. ACM, New York, NY, USA. 10 pages. (Best Paper Award Runners-up) [Link][Full version][Slides][Talk]
MmWave Beam Alignment Algorithm Designs and Prototyping
This projects focuses on the application of side-information and the non-coherent property of low-cost mmWave systems. An Side-information-Aided Non-coherent Beam Alignment (SANBA) is designed and prototyped with Matlab, USRP and phased arrays.
Publication: Yi Zhang, Kartik Patel, Sanjay Shakkottai, and Robert W. Heath Jr.. 2019. Side-information-aided Non-coherent Beam Alignment Design for Millimeter Wave Systems. In MobiHoc '19: The Twentieth ACM International Symposium on Mobile Ad Hoc Networking and Computing, July 02-05, 2019, Catania, Italy. ACM, New York, NY, USA, 10 pages. [GitHub]
DeepWiPHY: Deep learning-based receiver design and dataset for IEEE 802.11ax systems
In this work, we develop DeepWiPHY, a deep learning-based architecture to replace the channel estimation, common phase error (CPE) correction, sampling rate offset (SRO) correction, and equalization modules of IEEE 802.11ax based orthogonal frequency division multiplexing (OFDM) receivers. We first train DeepWiPHY with a synthetic dataset, which is generated using representative indoor channel models and includes typical radio frequency (RF) impairments that are the source of nonlinearity in wireless systems. To further train and evaluate DeepWiPHY with real-world data, we develop a passive sniffing-based data collection testbed composed of Universal Software Radio Peripherals (USRPs) and commercially available IEEE 802.11ax products. The comprehensive evaluation of DeepWiPHY with synthetic and real-world datasets (110 million synthetic OFDM symbols and 14 million real-world OFDM symbols) confirms that, even without fine-tuning the neural network's architecture parameters, DeepWiPHY achieves comparable performance to or outperforms the conventional WLAN receivers, in terms of both bit error rate (BER) and packet error rate (PER), under a wide range of channel models, signal-to-noise (SNR) levels, and modulation schemes.
Publication: Yi Zhang, A. Doshi, R. Liston, W. Tan, X. Zhu, J. G. Andrews, and R. W. Heath, "DeepWiPHY: Deep learning-based receiver design and dataset for IEEE 802.11ax systems,'' IEEE Transactions on Wireless Communications, vol. 20, no. 3, pp. 1596-1611, Mar. 2021. [Link][Code and Dataset]
Sept. 2015 - Present, Institute of Information Engineering, School of Electronic and Information Engineering, Xi'an Jiaotong University, China.
Research Assistant supervised by Prof. Hui-Ming Wang.
As an emerging technology to enhance spectrum sharing, NOMA has been widely recognized as a promising technology for 5G wireless systems. Different from conventional orthogonal multiple access (OMA), NOMA realizes the simultaneous transmission of multiple data streams via power domain division. At a transmitter, a superposition of different messages is broadcasted. At the receivers, the successive interference cancellation (SIC) technique is used to realize multi-user detection.
Much of my master's work focuses on the algorithm design and convex optimization in NOMA from the perspective of physical layer security, energy efficiency, and spectrum sharing:
Jan. 2014 - Apr. 2014, Ecole Centrale de Nantes, Nantes, France
Graduate Project supervised by Prof. Guillaume Morreau.
I conducted my graduate project under the supervision of Prof. Guillaume Morreau in Ecole Centrale de Nantes. In the project, a real-time program which realized 3D object visualization, detection, and segmentation has been developed. This project was based on Microsoft Kinect and Point Cloud Library (PCL). Thereinto, the registration, segmentation, and polygon mesh reconstruction for objects are realized, respectively, by the iterative closest points algorithm, the color-based region growing algorithm and the normal estimation method. Here is the demo: