Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Seung-Hwan Kim has a total of 7 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Hybrid Deep Learning Model for Automatic Modulation Classification

    by Seung-Hwan Kim, Chang-Bae Moon, Jae-Woo Kim, Dong-Seong Kim

    Automatic modulation classification (AMC) is one of the major challenges for cognitive radio (CR), which can enhance the spectrum utilization efficiency. In this study, a hybrid signal and image-based deep learning model is designed for AMC in CR. A convolutional neural network (CNN) is applied in both the deep learning models. The signal-based CNN (SBCNN) is designed with the optimal filter size for the prediction accuracy, which is used as a pre-training deep learning network to extract features with size $24\times 1$ . The features extracted by SBCNN are converted into heat map images, which showed RGB images in the scale range of −30 to +30. Finally, the images are utilized for training and testing the image-based CNN (IBCNN). The dataset used for the experiment is DeepSig: RADIOML2018.01A, which is the latest version. For the IBCNN, the prediction accuracy is 1.96%, 7.99%, and 4.63% higher at signal-to-noise ratio (SNR) 10 dB, and 3.26%, 6.4%, and 4.13% higher at SNR 0 dB as compared to conventional models: ECNN, SCGNet, and LCNN, respectively.


  • Open Access

    Precise-Point-Positioning Estimations for Recreational Drones Using Optimized Cubature-Extended Kalman Filtering

    by Williams-Paul Nwadiugwu, Seung-Hwan Kim, Dong-Seong Kim

    In modern-day multi-dimensional recreational drones (UAVs), the global navigation satellite system (GNSS) units in- use are commonly fraught with precise-point-positioning (PPP) data errors or inaccuracies, hence, necessitating this work. These data inaccuracies, occasioned by the system’s drawbacks such as sudden GPS lock or jamming, embedded device misalignment, drone’s limited communication coverage, signaling and detection, all contributes to the system’s PPP computation complexity. To mitigate PPP complexity, an intelligent and robust accurate continuous-discrete (ACD) based hybrid cubature-extended Kalman filter (C-EKF) computation model for an integrated GNSS unit is corroborated in this article. More precisely, time updates to the state and parameter sub-vectors for the GNSS unit is accomplished using the third-degree spherical-radial cubature rule. The system’s testbed simulation is then conducted using tightly-coupled units of (i) ring laser gyroscope (RLG) and (ii) micro-electro-mechanical system (MEMS) variants of the inertial measurement unit (IMU) to ascertain the PPP cooperative tendencies. Optimized performance comparisons of the proposed hybrid C-EKF over the existing cubature Kalman filter (CKF) and extended Kalman filter (EKF) models with-respect-to (w.r.t) its probabilistic outages, Yaw error-differences and ergodic capacities are demonstrated and presented.


  • Open Access

    Corrections to “Deep Learning-Based Robust Automatic Modulation Classification for Cognitive Radio Networks”

    by Seung-Hwan Kim, Jae-Woo Kim, Williams-Paul Nwadiugwu, Dong-Seong Kim

    In the above article [1] , in Table 4 , the operator * was used to represent frame sizes, which are 2 * 128 and 4 * 128. The correct operator should be a multiplication operator $\times $ for $2 \times 128$ and $4 \times 128$ .

  • Open Access

    Deep Learning-Based Robust Automatic Modulation Classification for Cognitive Radio Networks

    by Seung-Hwan Kim, Jae-Woo Kim, Williams-Paul Nwadiugwu, Dong-Seong Kim

    In this paper, a novel deep learning-based robust automatic modulation classification (AMC) method is proposed for cognitive radio networks. Generally, as network input of AMC convolutional neural networks (CNNs) images or complex signals are utilized in time domain or frequency domain. In terms of the image that contains RGB(Red, Green, Blue) levels the input size may be larger than the complex signal, which represents the increase of computational complexity. In terms of the complex signal it is normally used as $2 \times N$ size for the input, which is divided into in-phase and quadrature-phase (IQ) components. In this paper, the input size is extended as $4 \times N$ size by copying IQ components and concatenating in reverse order to improve the classification accuracy. Since the increase in the amount of computation complexity due to the extended input size, the proposed CNN archiecture is designed to reduce the size from $4 \times N$ to $2 \times N$ by an average pooling layer, which can enhence the classification accuracy as well. The simulation results show that the classification accuracy of the proposed model is higher than the conventional models in the almost signal-to-noise ratio (SNR) range.


  • Open Access

    Design and Implementation Scheme of QSFP28 Optical Transceiver for Long-Reach Transmission Using PAM4 Modulation

    by Jae-Woo Kim, Dong-Seong Kim, Seung-Hwan Kim, Sang-Moon Shin

    A quad, small form-factor pluggable 28 Gbps optical transceiver design scheme is proposed. It is capable of transmitting 50 Gbps of data up to a distance of 40 km using modulation signals with a level-four pulse-amplitude. The proposed scheme is designed using a combination of electro-absorption-modulated lasers, transmitter optical sub-assembly, low-cost positive-intrinsic-native photodiodes, and receiver optical sub-assembly to achieve standard performance and low cost. Moreover, the hardware and firmware design schemes to implement the optical transceiver are presented. The results confirm the effectiveness of the proposed scheme and the performance of the manufactured optical transceiver, thereby confirming its applicability to real industrial sites.


  • Open Access

    Lightweight Deep Learning Model for Automatic Modulation Classification in Cognitive Radio Networks

    by Seung-Hwan Kim, Jae-Woo Kim, Van-Sang Doan, Dong-Seong Kim

    Automatic modulation classification (AMC) used in cognitive radio networks is an important class of methods apt to utilize spectrum resources efficiently. However, conventional likelihood-based approaches have high computational complexity. Thus, this paper proposes a novel convolutional neural network architecture for AMC. A bottleneck and asymmetric convolution structure are employed in the proposed model, which can reduce the computational complexity. The skip connection technique is used to solve the vanishing gradient problem and improve the classification accuracy. The dataset DeepSig:RadioML, which is composed of 24 modulation classes, is used for the performance analysis. Simulation results show that the classification accuracy performance of the proposed model is outstanding in the signal-to-noise ratio (SNR) range from $- 4\,\,dB$ to $20~dB$ compared with MCNet that is the best model in the conventional models, where the proposed model achieves 5.52% and 5.92% improvement regarding classification accuracy at the SNRs of $0~dB$ and $10~dB$ , respectively. In terms of the computational complexity, the proposed model not only saves the trainable parameters by more than 67% but also reduces the prediction time for a signal by more than 54.4% compared with those of MCNet.


  • Open Access

    Energy Consumption Analysis of Beamforming and Cooperative Schemes for Aircraft Wireless Sensor Networks

    by Seung-Hwan Kim, Jae-Woo Kim, Dong-Seong Kim

    In this paper, the eight schemes for aircraft wireless sensor networks are investigated, which are single-hop array beamforming schemes (including analog beamforming (ABF), and digital beamforming (DBF)), non-cooperative schemes (including single-hop and multi-hop schemes), cooperative schemes (including amplify and forward (AF), decode and forward (DF)), and incremental cooperative schemes (incremental decode and forward (IDF), and incremental amplify and forward (IAF)). To set up the aircraft wireless communication environment, we design the aircraft channel model by referring to the experimental parameters of the ITU (International Telecommunication Union)-R M.2283, which is composed of path loss, shadowing fading, and multi-path fading channel responses. To evaluate the performance, the conditions energy consumption and throughput analysis are performed. Through simulation results, the incremental cooperative scheme outperformed by 66.8% better at spectral efficiency 2 than the DBF scheme in terms of the energy consumption metric. Whereas, in terms of throughput metric, overall SNR (signal-to-noise ratio) ranged from −20 to 30 dB the beamforming scheme had the best performance in which the beamforming scheme at SNR 0 dB achieved 85.4% better than the multi-hop scheme. Finally, in terms of normalized throughput metric in low SNR range between −20 and 1 dB the ABF scheme had the best performance over the others in which the ABF at SNR 0 dB achieved 75.4% better than the multi-hop scheme. Whereas, in high SNR range between 2 and 30 dB the IDF scheme had the best performance in which the IDF at SNR 10 dB achieved 62.2% better than the multi-hop scheme.


  • Open Access

    The Firmware Design and Implementation Scheme for C Form-Factor Pluggable Optical Transceiver

    by Jae-Woo Kim, Dong-Seong Kim, Seung-Hwan Kim, Sang-Moon Shin

    The demand for integrated telecommunication network infrastructure has increased, and 100 Gbps optical transceivers are a critical part of this infrastructure. In this paper, an efficient firmware design scheme is proposed for a 100 Gbps C form-factor pluggable (CFP) optical transceiver based on the multi-source agreement standard for optical transceivers. In the proposed method, a field programmable gate array (FPGA) approach is used to integrate the CFP communication interface and register structure, and a micro controller unit (MCU) is employed to implement the operation of the CFP optical transceiver. This paper also proposed monitoring techniques using high-order polynomials for accurate optical power monitoring of CFP optical transceivers. To ensure that the implemented firmware satisfies the proposed design scheme, an actual testbed was constructed and the performance of the firmware was evaluated. The results demonstrate that the proposed design scheme not only satisfies standard items but also achieved an average monitoring accuracy of more than 90%. In addition, the proposed scheme can be applied to 200/400G CFP optical transceivers in the future. By implementing the CFP firmware using the proposed method, manufacturers can ensure quality while reducing cost and development time.