Research
My dissertation work considers using an automotive millimeter-wave radar as a vibrometer. Vibrometers broadly fall into two categories, those that use direct contact to measure displacements and those who acquire displacement measurements from some standoff distance. The latter tends to come in the form of a laser Doppler vibrometer (LDV) which exploits the Doppler effect to get displacement information from whatever surface is being observed by the device. Unfortunately, LDVs can be expensive. With automotive radars becoming cheaper, operating at higher frequencies (shorter wavelengths result in better velocity resolution), and making use of MIMO (providing opportunity for adequately narrow main lobes), my dissertation work looks at how well these radars could do at supplementing traditional LDVs and also expanding the application space of standoff vibrometry.
Though my primary research throughout my Ph.D. has been in radar signal processing, I have also grown more interested in a handful of other technical areas. My current research interests include other areas of signal processing, communication systems, coding theory, engineering education, technology ethics and policy, physics-motivated machine learning, and computational social science. Through courses, fellowship projects, and even an additional amount of work with other Duke community members, I have been able to explore these different interests.
Below you will find brief descriptions and citations for my publicly accessible signal processing and coding theory work. (Do note that in my first ~2 years of graduate school, I was primarily working on a DARPA project whose results were not permitted to be published.) Most of my work in engineering education has been described in the 'My Teaching' section of this site. Similarly, because my interest in technology ethics has lead me to also pursue an M.A. in Tech Ethics and Science Policy, my perspective, motivation, and Master's practicum are described in the 'My Perspective' section of this site. My other interests are intertwined in my work and play a part in where I desire to seek future opportunities but do not necessarily have separate space on this site.
Publications
01
Power Spectra of Constrained Codes with Level-Based Signaling: Overcoming Finite-Length Challenges
Abstract: In various practical systems, certain data patterns are prone to errors if written or transmitted. In magnetic recording and communication over transmission lines, data patterns causing consecutive transitions that are not sufficiently separated are prone to errors. In Flash memory with two levels per cell, data patterns causing high–low–high charge levels on adjacent cells are prone to errors. Constrained codes are used to eliminate error-prone patterns, and they can also achieve other goals. Recently, we introduced efficient binary symmetric lexicographically-ordered constrained (LOCO) codes and asymmetric LOCO (A-LOCO) codes to increase density in magnetic recording systems and lifetime in Flash systems by eliminating the relevant detrimental patterns. Due to their application, LOCO and A-LOCO codes are associated with level-based signaling. Studying the power spectrum of a random signal with certain properties is principal for any storage or transmission system. It reveals important properties such as the average signal power at DC, the bandwidth of the signal, and whether there are discrete power components at certain frequencies. In this paper, we first modify a framework from the literature in order to introduce a method to derive the power spectrum of a sequence of constrained data associated with level-based signaling. We apply our method to infinitely long sequences satisfying symmetric and asymmetric constraints. Next, we show how to generalize the method such that it works for a stream of finite-length codewords as well, thus demonstrating how to overcome the associated finite-length challenges. We use the generalized method to devise closed forms for the spectra of finite-length LOCO and A-LOCO codes from their transition diagrams. Our LOCO and A-LOCO spectral derivations can be performed for any code length and can be extended to other constrained codes. We plot these power spectra, and discuss various important spectral properties for both LOCO and A-LOCO codes. We also briefly discuss an alternative method for deriving the power spectrum and introduce an idea towards reaching the spectra of self-clocked codes.
​
​J. Centers, X. Tan, A. Hareedy and R. Calderbank, "Power Spectra of Constrained Codes With Level-Based Signaling: Overcoming Finite-Length Challenges," in IEEE Transactions on Communications, vol. 69, no. 8, pp. 4971-4986, Aug. 2021, doi: 10.1109/TCOMM.2021.3073179.
02
Vibrational Radar Backscatter Communications Theory and Bound
Abstract: Vibrational radar backscatter communication (VRBC) utilizes millimeter-wave radar vibrometry to receive message signals from vibrating surfaces. So long as multiple transponding surfaces are separable by the system radar in angle and/or range, VRBC can simultaneously detect, isolate, and decode messages from these multiple sources. For anything-to-vehicle communications, VRBC is therefore a scalable, low-latency approach which leverages existing automotive radars to increase situational awareness without requiring any additional use of the RF spectrum. In this paper, we describe the modeling and processing of VRBC signals. We show that knowledge of surface resonances and coding constraints can be used to improve system performance by incorporating them into a Markov model. We provide insight into the rate capabilities of VRBC by providing a theoretical upper bound that incorporates the presence of intersymbol interference (ISI). Simulation results compare our upper bound to achievable communication rates using single symbol vs sequence detection maximum likelihood methods for a given symbol length, set of symbol excitations, and VRBC radar pulse repetition interval (PRI).
​
J. Centers and J. Krolik, "Vibrational Radar Backscatter Communications Theory and Bound," in IEEE Transactions on Radar Systems, doi: 10.1109/TRS.2023.3268050.
Workshops & Conferences
1 / Non-Volatile Memory Workshop 2021
In collaboration with a Duke undergraduate, post doctoral researcher, and professor not in the lab that my primary research falls under, we presented on some work we completed that covers a method for obtaining the power spectra of level-based constrained codes. This workshop submission was nominated for Most Memorable Paper but did not win that title.
​
Links to Presentations on YouTube:
​
J. Centers, X. Tan, A. Hareedy and R. Calderbank, "Power spectra of finite-length constrained codes with level-based signaling", Proc. Annu. Non-Volatile Memories Workshop, pp. 1-2, Mar. 2021.
2 / Asilomar Conference on Signals, Systems, and Computers 2021
​I presented our initial work on vibrational radar backscatter communications (VRBC) at Asilomar. VRBC makes use of a millimeter-wave radar as a vibrometer to detect vibrational displacement on a transponder whose vibrations contain a modulated message. This conference paper/presentation introduced the signal model for VRBC and provided initial results as well as discussions on tradeoffs characteristic of such a system.
​
Link to Presentation on YouTube
​
J. Centers and J. Krolik, "Vibrational Radar Backscatter Communications," 2021 55th Asilomar Conference on Signals, Systems, and Computers, 2021, pp. 1086-1090, doi: 10.1109/IEEECONF53345.2021.9723226.
3 / IEEE's Sensor Array and Multichannel Signal Processing Workshop 2022
I shared our follow-up work on vibrational radar backscatter communications (VRBC) at SAM 2022. This primarily entailed providing clarity into how VRBC could be better implemented when large amounts of intersymbol interference (ISI) is present. To account for ISI, we modified VRBC to be modeled as a Markov model in which each symbol interval is defined by ISI and the current communication symbol. We then used the Viterbi algorithm to best minimize errors due to ISI.
​
This workshop was held in person, so there is no video link available to the presentation.
​
J. Centers and J. Krolik, "Vibrational Radar Backscatter Communication using Resonant Transponding Surfaces," 2022 IEEE 12th Sensor Array and Multichannel Signal Processing Workshop (SAM), 2022, pp. 71-75, doi: 10.1109/SAM53842.2022.9827869.
4 / IEEE's International Conference on Acoustics, Speech, and Signal Processing 2023
I will be sharing our detailed work on vibrational radar backscatter communications (VRBC) range processing methods at ICASSP 2023. This work focuses on why we veer from conventional FMCW range-Doppler processing methods and instead utilize the full de-chirped return in message decoding. In doing so, we describe how to maintain the ability to separate out multiple transponders in range by proposing the use of bandpass filters rather than range bin selection after an FFT in fast-time.
​
This workshop will be held in person, but a video link summarizing this work will be made available shortly before the conference in June of 2023.
​
[coming soon]