Qi Zheng

PhD. Candidate
Computer Engineering Laboratory
Computer Science and Engineering
University of Michigan, Ann Arbor

I am a PhD candidate in the Department of Electrical Engineering and Computer Science at University of Michigan, Ann Arbor. I am working with Professor Trevor Mudge. I received my B.S.E degree in Electrical Engineering from Harbin Institute of Technology, China, in 2010, and my M.S.E degree in Computer Engineering from University of Michigan, Ann Arbor, in 2012.
[Curriculum Vitae]

Research

My research interests include energy-efficient computer architecture for high-throughput processors, and power-efficient computer system designs for wireless communication, computer vision and image processing. Here is a list of my ongoing and past research projects.

  • Design Future Radio Access Networks on General Purpose Cloud Platforms
    Investigate the cloud server design with GPUs for cloud radio access network (C-RAN), a new cloud service for future wireless communication systems.

  • Leveraging Inter-warp Progress Divergence to Improve GPU Energy Efficiency
    Study the causes and influences of inter-warp progress divergence during GPU execution and use it to reduce the energy consumption in the GPU register file with advanced low-leakage memory technologies.

  • Power Efficient Revolution for Embedded Computing Technology
    Explore a hardware-software co-design solution to achieve 75GFLOPS/w power efficiency for the next-generation embedded computing systems for computer vision and image processing applications.

  • Architect an LTE Base Station on GPGPU
    Study the cost, power consumption and performance of building a baseband subsystem of an LTE base station with commercial GPUs.

  • WiBench: a Configurable Wireless Benchmark
    Design WiBench, a configurable wireless benchmark of different protocols for communication processor design, including WCDMA, LTE and 802.11a.

Publications

  • Using Graphics Processing Units in an LTE Base Station
    Q. Zheng, Y. Chen, H. Lee, R. Dreslinski, C. Chakrabarti, A. Anastasopoulos, S. Mahlke, and T. Mudge.
    Journal of Signal Processing Systems 78, no. 1 (2015): 35-47. (Invitated Paper)
    [Paper: pdf]

  • A hybrid approach to offloading mobile image classification
    J. Hauswald, T. Manville, Q. Zheng, R. Dreslinski, C. Chakrabarti, and T. Mudge.
    In 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), May 2014, pp. 8375-8379.
    [Paper: pdf]

  • An Architecture for Low-Power High-Performance Embedded Computing
    R.G. Dreslinski, Q. Zheng, R.P. Higgins, J. Hauswald, D. Blaauw, T. Mudge, C. Chakrabarti, J. Ballast, W. Snapp.
    In Thirty ninth Annual GOMACTech Conference (GOMAC), April 2014, pp. 423-426.
    [Paper: pdf]

  • Architecting an LTE base station with graphics processing units
    Q. Zheng, Y. Chen, R. Dreslinski, C. Chakrabarti, A. Anastasopoulos, S. Mahlke, and T. Mudge.
    In 2013 IEEE Workshop on Signal Processing Systems (SiPS), October 2013, pp. 219-224.
    [Paper: pdf] [Presentation: pdf]

  • WiBench: An open source kernel suite for benchmarking wireless systems
    Q. Zheng, Y. Chen, R. Dreslinski, C. Chakrabarti, A. Anastasopoulos, S. Mahlke, and T. Mudge.
    In 2013 IEEE International Symposium on Workload Characterization (IISWC), September 2013, pp. 123-132.
    [Paper: pdf] [Presentation: pdf]

  • Parallelization techniques for implementing trellis algorithms on graphics processors
    Q. Zheng, Y. Chen, R. Dreslinski, C. Chakrabarti, A. Anastasopoulos, S. Mahlke, and T. Mudge.
    In 2013 IEEE International Symposium on Circuits and Systems (ISCAS), May 2013, pp. 1220-1223.
    [Paper: pdf] [Presentation: pdf]

  • M5 based edge architecture modeling
    P. Gou, Q. Li, Y. Jin, Q. Zheng, B. Yang, M. Yu, and J. Wang.
    In 2010 IEEE International Conference on Computer Design (ICCD), October 2010, pp. 289-296.
    [Paper: pdf]

WiBench

WiBench is an open-source configurable wireless benchmark suite that includes the key signal processing kernels of the baseband system in many mainstream wireless protocols, include 802.11, WCDMA and LTE. It was designed by me and Yajing Chen at University of Michigan, Ann Arbor.

WiBench kernels are self-contained configurable blocks that can be tuned to meet the different system requirements. Several standard channel models are also included to study system performance, such as the bit error rate. The suite also contains an LTE uplink system as a representative example of a wireless system that can be built using these kernels. WiBench is provided in C++ to make it easier for computer architects to profile and analyze the system. A MATLAB version is also included for debugging purposes.

The download of WiBench is available through this link. More detailed information can be found on the WiBench website.

Here is a full list of what is in WiBench:
Category Benchmark
Kernels Channel coding/decoding
Rate matching
Scrambling/Descrambling
Constellation mapping/demapping
MIMO detection
FFT/IFFT
Sub-carrier mapping/demapping
Channel Estimation
Channel models Gaussian Random Channel model (GRC)
Extended Pedestrian A model (EPA)
Extended Vehicular A model (EVA)
Extended Typical Urban model (ETU)
Applications LTE Uplink

Miscellaneous

  • Interned at Qualcomm Research and Mediatek.
  • Awarded Dwight F. Benton Fellowship and Chinese National Fellowship.
  • Mainly use C/C++, Python and MATLAB.

  • Speak Mandarain, English and Spanish.
  • Like Coke and Cheetos.
  • Fan of Football Manager and Civilization.
  • Forza Inter!

Contact

Email:    qizheng at umich dot edu
Addr:     4856 Beyster Bldg
                  2260 Hayward Street
                  Ann Arbor, MI 48109-2121