Hyunjeong Kwon

Hyunjeong Kwon

Auto-generated profile, based on publicly available information

Hyunjeong Kwon

Electrical Engineering Department, Pohang University of Science and Technology, Pohang, South Korea

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
2 February 2021 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
HKHyunjeong Kwon
Hyunjeong Kwon
DKDaeyeon Kim
Daeyeon Kim
YKYoung Hwan Kim
Young Hwan Kim
SKSeokhyeong Kang
Seokhyeong Kang
Under process, voltage, and temperature variations, SRAM cell stability largely fluctuates from the nominal value. In the design step, SRAM cell optimization while ignoring the fluctuation induces the yield loss for the stability. Variation-aware optimization of an SRAM cell can prevent the yield loss problem by considering the mean and variance of SRAM cell stability when finding optimal design parameters. This paper proposes a novel SRAM optimization method that uses a deep neural network (DNN). Multiple DNNs from ensemble techniques represent the mean and variance of SRAM cell stability for the nominal design parameters. Subsequent sensitivity analysis of DNN extracts the ${K}$ design parameters that have the most dominant effects on the mean and variance of SRAM cell stability. Then multidimensional optimization is used to find the optimal values of these ${K}$ parameters to maximize the mean stability while minimizing its variance. The proposed method achieved an average of 2% error compared to MC simulation. The proposed optimization method takes only 561 s to provide the most optimal design parameter values of an SRAM cell.

17 September 2018 in IEEE Access
Verified
Journal Article
Open access
HKHyunjeong Kwon
Hyunjeong Kwon
MWMingyu Woo
Mingyu Woo
YKYoung Hwan Kim
Young Hwan Kim
SKSeokhyeong Kang
Seokhyeong Kang
In the design process of advanced semiconductor devices, statistical leakage analysis has emerged as a major step due to uncertainties in the leakage current caused by the process variations. In this paper, a novel statistical leakage analysis which uses Gaussian mixture model (GMM) as the density function of leakage current is proposed. To estimate the probability density function, our proposed method clusters the rapidly converged leakage data using the GMM. The GMM can represent any distributions, so it is suitable to estimate the leakage distribution, which varies as the technology node or operating condition changes. In addition, our proposed method (SLA-GMM) defines a terminating condition that guarantees the convergence of the leakage data and prevents the underfitting or overfitting in the GMM modeling process. With sequential addition, SLA-GMM significantly reduced the error that can occur during the addition process. In studies with a goodness-of-fit test, SLA-GMM achieved up to 98% and 94% improvements in the Chi-square static and the K-S static compared with the previous method based on an analytic model.

1 February 2021
Verified
Conference Paper
Closed access
Sung-Hoon KimSung-Hoon Kim
Sung-Hoon Kim
HKHyunjeong Kwon
Hyunjeong Kwon
EKEunji Kwon
Eunji Kwon
YCYoungchang Choi
Youngchang Choi
TOTae-Hyun Oh
Tae-Hyun Oh
SKSeokhyeong Kang
Seokhyeong Kang
In this work, we present a differentiable neural architecture search (NAS) method that takes into account two competing objectives, quality of result (QoR) and quality of service (QoS) with hardware design constraints. NAS research has recently received a lot of attention due to its ability to automatically find architecture candidates that can outperform handcrafted ones. However, the NAS approach which complies with actual HW design constraints has been under-explored. A naive NAS approach for this would be to optimize a combination of two criteria of QoR and QoS, but the simple extension of the prior art often yields degenerated architectures, and suffers from a sensitive hyperparameter tuning. In this work, we propose a multi-objective differential neural architecture search, called MDARTS. MDARTS has an affordable search time and can find Pareto frontier of QoR versus QoS. We also identify the problematic gap between all the existing differentiable NAS results and those final post-processed architectures, where soft connections are binarized. This gap leads to performance degradation when the model is deployed. To mitigate this gap, we propose a separation loss that discourages indefinite connections of components by implicitly minimizing entropy.

11 October 2018 in IEEE Access
Verified
Journal Article
Open access
HKHyunjeong Kwon
Hyunjeong Kwon
JKJae Hoon Kim
Jae Hoon Kim
SKSeokhyeong Kang
Seokhyeong Kang
YKYoung Hwan Kim
Young Hwan Kim
We propose SoftCorner, which is a novel approach to estimating the worst delay (3σ delay) of a VLSI system. SoftCorner is a modification of deterministic static timing analysis to overcome its tendency to give pessimistic estimates of the system delays when it uses 3σ corner delays for logic gates. The basic idea of SoftCorner is to use corner delays that are relaxed to < 3σ. To select the relaxed gate delays, SoftCorner uses the delay information obtained from the K most-critical paths and then determines the degree of relaxation, and constructs a model that represents the probability that the delay of each logic gate is selected. This probability model is used to guide selection of gate delays; the worst system delay with relaxed delay criterion can be obtained by running the selection module until the mean of the results converges. SoftCorner also can estimate the system delay at an arbitrary percentile. In experiments, SoftCorner estimated the system delay of benchmark circuits at 99.87th, 95th, and 85th percentiles with an average error < 2% compared to Monte-Carlo (MC) simulation, and produced the estimates 5.89 × 104 times faster than MC simulation on average.

19 February 2020 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
HKHyunjeong Kwon
Hyunjeong Kwon
SLSung-Yun Lee
Sung-Yun Lee
YKYoung Hwan Kim
Young Hwan Kim
SKSeokhyeong Kang
Seokhyeong Kang
Variation-aware leakage analysis becomes an essen-tial design process as the technology node continuously shrinks. This paper proposes a novel additive statistical leakage analysis method that uses exponential mixture model (EMM) to estimate the leakage distribution. Using a few leakage data for sub-blocks of an input circuit, we estimate any shape of leakage distribution regardless of new process nodes or operating conditions. Leakage distribution of an input circuit can be obtained by adding the leak-age distributions of the sub-blocks. The proposed addition step se-quentially adds the leakage distributions of sub-blocks that are ex-pressed as EMMs. Before the addition step, we improve the accu-racy by handling linear dependence among leakage simulation data of sub-blocks. In addition, we propose a method to reduce the number of components of an EMM to prevent exponential in-crease in runtime and memory during the addition process. The proposed method achieved 43.6 times improvement in goodness-of-fit of the estimated cumulative density functions compared to the best results of other analytic model-based methods.

1 October 2016
Verified
Conference Paper
Closed access
HKHyun-Jeong Kwon
Hyun-Jeong Kwon
YKYoung Hwan Kim
Young Hwan Kim
We propose an analytic method to calculate the probability of timing violation of F/F-controlled paths by considering timing variations. We first characterize the timing characteristics of F/Fs and path delays using a generalized canonical delay model. The probability of setup-time violation is then calculated by considering the correlation between F/Fs and logic paths. For the calculation, we used conditional tightness probability equation which reflects the nonlinearity of the process parameters. In experiments, the proposed method exhibited the error less than 7 % with respect to Monte-Carlo (MC) simulation results. Compared to the benchmark method that does not consider the variations of the timing characteristics of F/Fs, the proposed method improved the accuracy by more than 18% on average.