Welcome to David's website
David G. M. Mitchell received the Ph.D. degree in Electrical Engineering from the University of Edinburgh, United Kingdom, in 2009. He is currently an Assistant Professor in the Klipsch School of Electrical and Computer Engineering at the New Mexico State University, USA. He previously held Visiting Assistant Professor and Post-Doctoral Research Associate positions in the Department of Electrical Engineering at the University of Notre Dame, USA. He is a Senior Member of the IEEE and his research interests are in the area of digital communications, with emphasis on error control coding and information theory. Dr. Mitchell has published over 40 peer-reviewed IEEE journal and conference articles gathering more than 1000 citations. He has received two best paper awards and is the recipient of the 2019 NMSU Early Career Award for Exceptional Achievements in Creative Scholarly Activity. He was also the recipient of the 2019 New Mexico EPSCoR Mentor Award. Dr. Mitchell serves as an Associate Editor for the IEEE Transactions on Information Theory.
- 05/24/21: Our new paper "Spatially Coupled Generalized LDPC Codes: Asymptotic Analysis and Finite Length Scaling" has been published in the latest IEEE Transactions on Information Theory.
In this paper, we show that by generalizing the code constraints, so-called spatially coupled generalized LDPC (SC-GLDPC) codes have capacity approaching performance at moderate block/constraint lengths and no observable error floors (see solid curves on right versus non spatially coupled on left). Such codes are an attractive choice for applications requiring excellent performance throughout the entire range of decoded error rates under strict latency and power constraints.
- 04/15/20: Min Zhu presented (online) our research paper "Adaptive Doping of Spatially Coupled
LDPC Codes" at the 2020 IEEE Information Theory Workshop (ITW).
- 01/26/21: Our new book chapter "Low‐Density Parity‐Check (LDPC) Codes for 5G Communications" has been published as part of the Wiley 5G Ref: The Essential 5G Reference Online, Wiley & Sons.
In this article, we describe the fundamental advances in low‐density parity‐check (LDPC) codes over the last two decades with special emphasis on the class of LDPC codes selected for the 5G new radio standard. We conclude the article by presenting three of the more promising extensions of LDPC codes known to date (generalized LDPC codes, nonbinary LDPC codes, and spatially coupled LDPC codes), which could potentially replace conventional LDPC codes in future communication standards.
- 12/10/20: Yanfang Liu presented (online) our research paper "On the Design of Generalized LDPC codes with Component BCJR Decoding" at the 2020 IEEE Global Communications Conference (GLOBECOM).
- 11/20/20: Our new paper "Error Propagation Mitigation in Sliding Window Decoding of Braided Convolutional Codes" has been published in the latest IEEE Transactions on Communications.
Braided Convolutional Codes can achieve capacity approaching performance, but suffer from infrequent but severe error propagation in a streaming environment (see the blue curve). In this paper, we show that such behavior can be detected and mitigated by a "resynchronization" of the decoder, yielding orders of magnitude better error control performance (red curve).
- 11/13/20: Congratulations to Ahmad Golmohammadi for defending his PhD thesis "Spatially Coupled Codes for Source Compression". Well done Dr. Golmohammadi, you have done some excellent work!
- 10/27/20: Min Zhu presented our research paper "Decoder Error Propagation Mitigation for Spatially Coupled LDPC Codes" at the 2020 International Symposium on Information Theory and Its Applications (online).
- 09/20/20: Our new paper "Efficient Implementation of a Threshold Modified Min-Sum Algorithm for LDPC Decoders" has been published in the latest IEEE Transactions on Circuits and Systems.
The proposed architecture introduces a novel lookup table based threshold attenuation technique, called threshold attenuated MSA (TAMSA). The TAMSA implementation (schematic above) is shown to improve bit error rate performance compared to conventional methods with no extra circuit power or circuit area compared to conventional AMSA, and only 0.07% extra leaf cells compared to conventional MSA.
- 10/27/20: Yanfang Liu presented (online) our research paper "Efficient Implementation of a Threshold Modified Min-Sum Algorithm for LDPC Decoders" at the 2020 IEEE International Symposium on Integrated Circuits and Systems (ISICAS).
- 07/01/20: Excited to start work as an Associate Editor for the IEEE Transactions on Information Theory!
- 06/30/20: Congratulations to Andrew Cummins for defending (by zoom) his MSEE thesis "Iterative Threshold Decoding for Spatially Couped, Turbo-like Codes". Great job Andrew!
- 06/25/20: Had a great time this week teaching an "Introduction to Probability" Workshop to undergraduate students from New Mexico regional universities, community colleges, and tribal colleges as part of the New Mexico EPSCoR STEM Advancement Program (STEMAP).
- 06/20/20: Our new paper "Designing Protograph-Based Quasi-Cyclic Spatially Coupled LDPC Codes With Large Girth" has been published in the latest IEEE Transactions on Communications.
In this paper, we introduce a systematic multi-stage graph lifting process that yields QC-SC-LDPC codes with large girth. Simulation results show the design leads to improved decoding performance, particularly in the error floor, compared to random constructions.
- 06/22/20: Attending the 2020 IEEE International Symposium on Information Theory (online) to present our research paper "A Novel Design of Spatially Coupled LDPC Codes for Sliding Window Decoding".
- 05/29/20: Honored to have my research selected as a National Science Foundation highlight of the New Mexico Smart Grid Center! An excerpt:
"...experts predict that the number of networked devices will be three times the global population by 2023, an increase that will challenge the carrying capacity of communications infrastructure on a global scale. Professor David Mitchell and his students at New Mexico State University address the data deluge by improving the performance of LDPC codes inherent in most next-generation wireless technology (5G) communications. Their algorithms demonstrate previously unachievable gains in computational efficiency that lessen the energy costs of digital information exchange and enable efficient monitoring and management of our nation’s power resources in the future."
Thanks to Brittney Van Der Werff and NM EPSCoR for the wonderful write-up and image (note the convolutional Tanner graphs protecting the information!)
- 04/15/20: Our new paper “A Threshold-Based Min-Sum Algorithm to Lower the Error Floors of Quantized LDPC Decoders” has been published in the latest IEEE Transactions on Communications.