Seungheun Song

Seungheun Song

Auto-generated profile, based on publicly available information

Seungheun Song

Department of Electrical and Computer Engineering, University of Michigan, Ann Arbor, MI, USA

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 February 2025 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
JCJustin M. Correll
Justin M. Correll
LJLu Jie
Lu Jie
SSSeungheun Song
Seungheun Song
SLSeungjong Lee
Seungjong Lee
JZJunkang Zhu
Junkang Zhu
MFMichael P. Flynn
Michael P. Flynn
Analog compute in memory (CIM) with multilevel cell (MLC) resistive random access memory (ReRAM) promises highly dense and efficient compute support for machine learning and scientific computing. This article introduces analog to digital converter (ADC)-assisted bit-serial processing for efficient, high-throughput compute. Bit-serial digital to analog converters (DACs) and 8-bit binary-weighted multicycle sampling (BWMCS) ADCs perform analog vector-matrix multiplication (VMM) on MLC-based crossbar arrays. A direct drive ${g}_{m}$ -boosted transimpedance amplifier (TIA) enables high-speed crossbar readout. We present a system on chip (SoC) prototype consisting of four self-contained ReRAM-based CIM macros and a reduced instruction set computer-five (RISC-V) host. The test chip is fabricated in 65 nm CMOS with foundry-integrated MLC ReRAM. We trained LeNet1 for handwritten digit classification and mapped the CNN weights differentially to 3-bit MLC ReRAM across multiple CIM macros. The classification accuracy loss is 1.6% when compared to the quantization-aware trained model. The measured raw and normalized peak efficiencies are 20.7 and 662 TOPS/W, respectively. The compute density is 8.4 TOPS/mm2.

23 December 2022 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
SLSeungjong Lee
Seungjong Lee
TKTaewook Kang
Taewook Kang
SSSeungheun Song
Seungheun Song
KKKyumin Kwon
Kyumin Kwon
MFMichael P. Flynn
Michael P. Flynn
This work introduces a time-interleaved architecture to tackle the speed-resolution bottleneck of the noise-shaping (NS) successive approximation register (SAR) quantizer in a continuous-time (CT) sigma-delta modulator (SDM). A critical insight is that introducing a delay in the NS quantizer feedback loop enables complete parallelization of the NS quantizer operations. The extra time from parallelization greatly relaxes loop filtering and residue integration and enables a high quantizer resolution. Furthermore, the extra time in the feedback loop allows data weighted averaging (DWA), which eliminates the need for calibration. The prototype comprises a second-order CT front-end and a fully interleaved first-order NS quantizer. The prototype is fabricated in a 28 nm process, occupies 0.072 mm2 and consumes 6.4 mW. The peak signal-to-noise ratio (SNR), signal-to-noise-distortion ratio (SNDR), and dynamic range (DR) are 83.9, 81.6, and 84.8 dB, respectively, in a 15.625 MHz bandwidth. The corresponding Schreier SNDR figure of merit (FoM) is 175.5 dB.

21 March 2025 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
MBMatthew R. Belz
Matthew R. Belz
ZXZhengqi Xu
Zhengqi Xu
HCHsiang-Wen Chen
Hsiang-Wen Chen
SSSeungheun Song
Seungheun Song
MFMichael P. Flynn
Michael P. Flynn
An integer-N digital-sampling phase-locked loop (PLL) based on a noise-shaping SAR (NS-SAR) analog-to-digital converter (ADC) breaks the quantization noise limitation of prior ADC-based phase detectors and enables the use of a lower resolution ADC quantizer and capacitive digital to analog converter (CDAC). Additionally, this improved ADC resolution reduces the phase-detector gain requirement and improves the lock performance. The 2nd-order noise-shaping of the NS-SAR ADC improves the time-to-voltage resolution of the phase detector by reducing in-band quantization noise and ADC input-referred noise (e.g., comparator noise). Measurements verify that noise shaping reduces the in-band noise by about 7 dB. The 28 nm CMOS prototype PLL operates at 4.3 GHz and consumes 13.7 mW. The measured rms integrated jitter from 1 kHz to 100 MHz is 133 fs.

9 December 2024 in IEEE Open Journal of the Solid-State Circuits Society
Verified
Journal Article
Open access
SSSeungheun Song
Seungheun Song
TKTaewook Kang
Taewook Kang
SLSeungjong Lee
Seungjong Lee
MFMichael P. Flynn
Michael P. Flynn
This article introduces a fully dynamic SAR-assisted pipeline ADC that uses a floating ring amplifier (FLORA) and gain-enhancing Miller negative capacitance (Miller negative-C). FLORA is a fully dynamic and bias-free ring amplifier powered by reservoir capacitors. Different reservoir capacitors for auto-zero and amplification phases optimize the power consumption and dominant pole locations. Furthermore, FLORA enhances speed without needing common-mode load capacitors in each stage and does not need a switched-capacitor common-mode feedback circuit at the output. The Miller negative-C improves the accuracy of the closed-loop residue amplifier by reducing the gain error related to the product of the finite operational amplifier gain and the feedback factor. This gain error compensation scheme eliminates the need for extra circuitry or correction phases. It occupies a small area and requires little additional power consumption. The paper analyzes the stability, effective range, and settling behavior of an amplifier with Miller negative-C. The prototype ADC implemented in a 28-nm CMOS process achieves an SNDR and an SFDR of 67.9 dB and 84.3 dB, respectively, while consuming 1.72 mW at 150 MS/s from a 1 V supply. The corresponding Walden and Schreier SNDR FoMs are 5.7 fJ/conversion-step and 173 dB, respectively.

23 November 2023 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
TKTaewook Kang
Taewook Kang
SLSeungjong Lee
Seungjong Lee
SSSeungheun Song
Seungheun Song
MHMohammad Reza Haghighat
Mohammad Reza Haghighat
MFMichael P. Flynn
Michael P. Flynn
This work introduces a speech recognition frontend system that solves the problems of conventional adaptive beamforming (ABF) with: 1) low digital signal processing (DSP) power consumption (3 $\times $ lower than state-of-the-art ABF) thanks to an innovative greedy blocking matrix (GBM) employing simple calculations; 2) automatic direction-of-arrival (DOA) error compensation with direction tracking delay-and-sum beamformer aided by the GBM; 3) a multimode hybrid analog-to-digital converter (ADC) adapts to signal conditions; and 4) multimode beamforming takes advantage of high-signal signal-to-noise ratio (SNR) to reduce total power by 54%. A prototype fabricated in a 40 nm process occupies 0.94 mm2 while consuming 157 and 72 $\mu \text{W}$ in high-power and low-power modes, respectively. The proposed system improves speech recognition accuracy from 54% to 83% under noisy conditions.

26 January 2026 in IEEE Journal of Solid-State Circuits
Verified
Journal Article
Closed access
ZZZhiyuan Zhao
Zhiyuan Zhao
HCHsiang-Wen Chen
Hsiang-Wen Chen
SSSeungheun Song
Seungheun Song
TKTaewook Kang
Taewook Kang
MFMichael P. Flynn
Michael P. Flynn
The noise-shaping (NS) successive approximation register (SAR) analog-to-digital converter (ADC) architecture is very power efficient but limited in bandwidth (BW). Time-interleaving (TI) expands the NS SAR BW with low hardware cost, but the conventional noise transfer function (NTF) zero placement leads to a marginal BW improvement with increasing NTF order. This work proposes a new hybrid structure utilizing cascaded (Ca) NS to decouple the oversampling rate (OSR) from the NTF order. A 6th-order bandpass (BP) cascaded time-interleaved (CaTI) NS SAR implements complex conjugate NTF zeros and requires only three interleaved channels and an OSR of $4\times $ . The prototype, fabricated in 28-nm CMOS, provides a 60-dB signal-to-noise-distortion ratio (SNDR) over a 175-MHz BW without calibration. The core circuit occupies 0.02 mm2 and consumes 3.38 mW at 1.4 GS/s. The corresponding Schreier SNDR figure of merit (FoM) is 167 dB. The experimental results demonstrate that the proposed CaTI NS SAR is a more robust and efficient alternative to conventional TI NS SAR ADC.