Lok-won Kim

Dr. Lok-won Kim

Kyunghee University
+1

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

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

Representative Publications

View All Publications
8 March 2017 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
Lok-won KimLok-won Kim
Lok-won Kim
Although there have been many decades of research and commercial presence on high performance general purpose processors, there are still many applications that require fully customized hardware architectures for further computational acceleration. Recently, deep learning has been successfully used to learn in a wide variety of applications, but their heavy computation demand has considerably limited their practical applications. This paper proposes a fully pipelined acceleration architecture to alleviate high computational demand of an artificial neural network (ANN) which is restricted Boltzmann machine (RBM) ANNs. The implemented RBM ANN accelerator (integrating 1024 x 1024 network size, using 128 input cases per batch, and running at a 303-MHz clock frequency) integrated in a state-of-the art field-programmable gate array (FPGA) (Xilinx Virtex 7 XC7V-2000T) provides a computational performance of 301-billion connection-updates-per-second and about 193 times higher performance than a software solution running on general purpose processors. Most importantly, the architecture enables over 4 times (12 times in batch learning) higher performance compared with a previous work when both are implemented in an FPGA device (XC2VP70).

5 August 2019 in IEEE Access
Verified
Journal Article
Open access
MAMadyan Alsenwi
Madyan Alsenwi
IYIbrar Yaqoob
Ibrar Yaqoob
SPShashi Pandey
Shashi Pandey
YTYan Kyaw Tun
Yan Kyaw Tun
Anupam Kumar BairagiAnupam Kumar Bairagi
Anupam Kumar Bairagi
Choong Seon HongChoong Seon Hong
Choong Seon Hong
Long-Term Evolution in the Unlicensed Spectrum (LTE-U) is considered as an indispensable technology to mitigate the spectrum scarcity in wireless networks. Typical LTE transmissions are contention-free and centrally controlled by the Base Station (BS). However, the wireless networks that work in unlicensed bands use contention-based protocols for channel access, which raise the need to derive an efficient and fair coexistence mechanism among different radio access networks. In this paper, we propose a novel mechanism based on neural networks for the coexistence of an LTE-U BS in the unlicensed spectrum alongside with WiFi access points. Specifically, we model the problem in coexistence as a 2-Dimensions Hopfield Neural Network (2D-HNN) based optimization problem that aims to achieve fairness considering both the LTE-U data rate and the QoS requirements of WiFi networks. Using the energy function of 2D-HNNs, precise investigation of its minimization property can directly provide the solution of the optimization problem. Furthermore, the problem of allocating the unlicensed resources to LTE-U users is modeled as a 2D-HNN and its energy function is leveraged to allocate resources to LTE-U users based on their channel states. Numerical results show that the proposed algorithm allows the LTE-U BS to work efficiently in the unlicensed spectrum while protecting the WiFi networks. Moreover, more than 90% fairness among the LTE-U users is achieved when allocating the unlicensed resources to LTE-U users based on the proposed algorithm.

15 February 2022 in IEEE Access
Verified
Journal Article
Open access
HVHieu Vo
Hieu Vo
LLLinh Le
Linh Le
FAFaaiz Asim
Faaiz Asim
Lok-won KimLok-won Kim
Lok-won Kim
Choong Seon HongChoong Seon Hong
Choong Seon Hong
Deep neural networks (DNNs) have played an increasingly important role in various areas such as computer vision and voice recognition. While training and validation become gradually feasible with high-end general-purpose processors such as graphical processor units (GPU), high throughput inferences in embedded hardware platforms with low hardware resources and power consumption efficiency are still challenging. Binarized neural networks (BNNs) are emerging as a promising method to overcome these challenges by reducing bit widths of DNN data representations with many optimal prior solutions. However, accuracy degradation is a considerable problem of the BNN, compared to the same architecture with full precision, while the binary neural networks still contain significant redundancy for optimization. In this paper, to address the limitations, we implement a streaming accelerator architecture with three optimization techniques: pipelining-unrolling for streaming each layer, weight reuse for parallel computation, and MAC (multiplication-accumulation) compression. Our method first constructs streaming architecture by pipelining-unrolling method to maximize throughput. Next, the weight reuse method with the K-mean cluster is applied to reduce the complexity of the popcount operation. Finally, MAC compression reduces hardware resources used for remaining computation on MAC operations. The implemented hardware accelerator integrated into a state-of-the-art field programable gate array (FPGA) provides the maximum performance of the classification at 1531k frames per second with 98.4% accuracy for the MNIST dataset and 205K frame per second with 80.2% accuracy for the Cifar-10 dataset. Besides, the proposed design’s ratio FPS/LUTs is approximately 57 (MNIST) and 0.707 (Cifar-10), which is much lower than the state-of-the-art design with a comparable throughput and inference accuracy.

19 January 2021 in Electronics
Verified
Journal Article
Open access
PDF file
PNPhuoc Nguyen
Phuoc Nguyen
Lok-won KimLok-won Kim
Lok-won Kim
People nowadays are entering an era of rapid evolution due to the generation of massive amounts of data. Such information is produced with an enormous contribution from the use of billions of sensing devices equipped with in situ signal processing and communication capabilities which form wireless sensor networks (WSNs). As the number of small devices connected to the Internet is higher than 50 billion, the Internet of Things (IoT) devices focus on sensing accuracy, communication efficiency, and low power consumption because IoT device deployment is mainly for correct information acquisition, remote node accessing, and longer-term operation with lower battery changing requirements. Thus, recently, there have been rich activities for original research in these domains. Various sensors used by processing devices can be heterogeneous or homogeneous. Since the devices are primarily expected to operate independently in an autonomous manner, the abilities of connection, communication, and ambient energy scavenging play significant roles, especially in a large-scale deployment. This paper classifies wireless sensor nodes into two major categories based the types of the sensor array (heterogeneous/homogeneous). It also emphasizes on the utilization of ad hoc networking and energy harvesting mechanisms as a fundamental cornerstone to building a self-governing, sustainable, and perpetually-operated sensor system. We review systems representative of each category and depict trends in system development.

17 August 2020 in IEEE Access
Verified
Journal Article
Open access
MAMuhammad Salman Ali
Muhammad Salman Ali
TITauhid Bin Iqbal
Tauhid Bin Iqbal
KLKang-Ho Lee
Kang-Ho Lee
AMAbdul Muqeet
Abdul Muqeet
SLSeunghyun Lee
Seunghyun Lee
SBSung-Ho Bae
Sung-Ho Bae
Deep Learning techniques have been successfully used to solve a wide range of computer vision problems. Due to their high computation complexity, specialized hardware accelerators are being proposed to achieve high performance and efficiency for deep learning-based algorithms. However, soft errors, i.e., bit flipping errors in the layer output, are often caused due to process variation and high energy particles in these hardware systems. These can significantly reduce model accuracy. To remedy this problem, we propose new algorithms that effectively reduce the impact of errors, thus keeping high accuracy. We firstly propose to incorporate an Error Correction Layer (ECL) into neural networks where convolution is performed multiple times in each layer and majority reporting is conducted for the outputs at bit level. We found that ECL can eliminate most errors while bypassing the bit-error when the bits at the same position are corrupted multiple times under the simulated condition. In order to solve this problem, we analyze the impact of errors depending on the position of bits, thus observing that errors in most significant bit (MSB) positions tend to severely corrupt the output of the network compared to the errors in the least significant bit (LSB) positions. According to this observation, we propose a new specialized activation function, called Piece-wise Rectified Linear Unit (PwReLU), which selectively suppresses errors depending on the bit positions, resulting in an increased model resistance against the errors. Compared to existing activation functions, the proposed PwReLU outperforms with large accuracy margins of up-to 20% even with very high bit error rates (BERs). Our extensive experiments show that the proposed ECL and PwReLU work in a complementary manner, achieving comparable accuracy to the error-free networks even at a severe BER of 0.1% on CIFAR10, CIFAR100, and ImageNet.

26 October 2020 in Sensors
Verified
Journal Article
Open access
PDF file
PNPhuoc Nguyen
Phuoc Nguyen
HVHieu Vo
Hieu Vo
LLLinh Le
Linh Le
SESeokJin Eo
SeokJin Eo
Lok-won KimLok-won Kim
Lok-won Kim
Partial discharge (PD) is a major indicator of various failures in power grid systems. PD exhibits a physical occurrence where a localized electrical discharge happens in insulation materials. This phenomenon causes damage to the insulating parts and, in various circumstances, leads to complete insulation breakdown. As a consequence, it can produce more costly outcomes such as abrupt outages or lost production. Therefore, PD detection plays a vital role in preventing insulation failure. In this work, we propose a system that utilizes heterogeneous sensors for the PD detection along with multi-sensor interface, real-time advanced denoise processing, flexible system operation, and Bluetooth-low-energy (BLE)-based ad hoc communication. Among the variety of heterogeneous sensors, several are developed by the application of nanomaterials and nanotechnology, thus outperforming the regular types. The proposed system successfully identifies the presence of PD from sensor signals using a microprocessor-based processing system and effectively performs an advanced denoising technique based on the wavelet transform through field-programmable-gate-array (FPGA)-based programmable logics. With the development of the system, we aim to achieve a solution with low cost, high flexibility and efficiency, and ease of deployment for the monitoring of power grid systems.