• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.2
• /
• pp.670-683
• /
• 2016

Network coding, as a new technique to improve the throughput, is studied combined with multi-relay model in this paper to address the challenges of long distance and power limit in deep space communication. First, an amplify-and-forward relaying approach based on analog network coding (AFNC) is proposed in multi-relay network to improve the capacity for deep space communication system, where multiple relays are introduced to overcome the long distance link loss. The design of amplification coefficients is mathematically formulated as the optimization problem of maximizing SNR under sum-power constraint over relays. Then for a dual-hop relay network with a single source, the optimal amplification coefficients are derived when the multiple relays introduce non-coherent noise. Through theoretic analysis and simulation, it is shown that our approach can achieve the maximum transmission rate and perform better over single link transmission for deep space communication.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.38 no.5
• /
• pp.499-506
• /
• 2010

Many countries of the world have been launched the competition of space development and Korea also has a plan for the launch of Lunar orbiter in 2020 and Lunar lander in 2025 for Lunar explorations. For the success of the planned Lunar exploration, we need to enhance the required deep space communication technologies. To achieve our goals, we should develop space communications system and Korean DSN (deep space network) based on experiences and technologies through cooperation with the advanced countries in the field of deep space exploration. In this paper, we investigate overseas DSNs and deep space communication systems, and present the link margin and other technical requirements for successful DSN deployment. In addition, we propose a best strategy to secure domestic ground stations for the Korean Lunar exploration missions.

• Journal of Astronomy and Space Sciences
• /
• v.40 no.2
• /
• pp.79-88
• /
• 2023

On Aug. 4, 2022, at 23:08:48 (UTC), the Korea Pathfinder Lunar Orbiter (KPLO), also known as Danuri, was launched using a SpaceX Falcon 9 launch vehicle. Currently, KPLO is successfully conducting its science mission around the Moon. The National Aeronautics and Space Administration (NASA)'s Deep Space Network (DSN) was utilized for the successful flight operation of KPLO. A great deal of joint effort was made between the Korea Aerospace Research Institute (KARI) and NASA DSN team since the beginning of KPLO ground system design for the success of the mission. The efficient utilization and management of NASA DSN in deep space exploration are critical not only for the spacecraft's telemetry and command but also for tracking the flight dynamics (FD) operation. In this work, the top-level DSN interface architecture, detailed workflows, DSN support levels, and practical lessons learned from the joint team's efforts are presented for KPLO's successful FD operation. Due to the significant joint team's efforts, KPLO is currently performing its mission smoothly in the lunar mission orbit. Through KPLO cooperative operation experience with DSN, a more reliable and efficient partnership is expected not only for Korea's own deep space exploration mission but also for the KARI-NASA DSN joint support on other deep space missions in the future.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.1019-1029
• /
• 2020

In this paper, we propose a novel knowledge distillation algorithm to create an compressed deep ensemble network coupled with the combined use of local and global features of face images. In order to transfer the capability of high-level recognition performances of the ensemble deep networks to a single deep network, the probability for class prediction, which is the softmax output of the ensemble network, is used as soft target for training a single deep network. By applying the knowledge distillation algorithm, the local feature informations obtained by training the deep ensemble network using facial subregions of the face image as input are transmitted to a single deep network to create a so-called compressed ensemble DCNN. The experimental results demonstrate that our proposed compressed ensemble deep network can maintain the recognition performance of the complex ensemble deep networks and is superior to the recognition performance of a single deep network. In addition, our proposed method can significantly reduce the storage(memory) space and execution time, compared to the conventional ensemble deep networks developed for face recognition.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.2
• /
• pp.46.2-46.2
• /
• 2021

We present solar event auto detection using deep-learning-based object detection algorithms and DeepSDO event dataset. DeepSDO event dataset is a new detection dataset with bounding boxed as ground-truth for three solar event (coronal holes, sunspots and prominences) features using Solar Dynamics Observatory data. To access the reliability of DeepSDO event dataset, we compared to HEK data. We train two representative object detection models, the Single Shot MultiBox Detector (SSD) and the Faster Region-based Convolutional Neural Network (R-CNN) with DeepSDO event dataset. We compared the performance of the two models for three solar events and this study demonstrates that deep learning-based object detection can successfully detect multiple types of solar events. In addition, we provide DeepSDO event dataset for further achievements event detection in solar physics.

• Journal of Astronomy and Space Sciences
• /
• v.31 no.1
• /
• pp.51-61
• /
• 2014

In this work, the preliminary analysis on both the tracking schedule and measurements characteristics for the spacecraft on the phase of lunar transfer and capture is performed. To analyze both the tracking schedule and measurements characteristics, lunar transfer and capture phases' optimized trajectories are directly adapted from former research, and eleven ground tracking facilities (three Deep Space Network sties, seven Near Earth Network sites, one Daejeon site) are assumed to support the mission. Under these conceptual mission scenarios, detailed tracking schedules and expected measurement characteristics during critical maneuvers (Trans Lunar Injection, Lunar Orbit Insertion and Apoapsis Adjustment Maneuver), especially for the Deajeon station, are successfully analyzed. The orders of predicted measurements' variances during lunar capture phase according to critical maneuvers are found to be within the order of mm/s for the range and micro-deg/s for the angular measurements rates which are in good agreement with the recommended values of typical measurement modeling accuracies for Deep Space Networks. Although preliminary navigation accuracy guidelines are provided through this work, it is expected to give more practical insights into preparing the Korea's future lunar mission, especially for developing flight dynamics subsystem.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.45-55
• /
• 2021

This paper has proposed an algorithm that detecting for dense small vehicle in large image efficiently. It is consisted of two Ensemble Deep-Learning Network algorithms based on Coarse to Fine method. The system can detect vehicle exactly on selected sub image. In the Coarse step, it can make Voting Space using the result of various Deep-Learning Network individually. To select sub-region, it makes Voting Map by to combine each Voting Space. In the Fine step, the sub-region selected in the Coarse step is transferred to final Deep-Learning Network. The sub-region can be defined by using dynamic windows. In this paper, pre-defined mapping table has used to define dynamic windows for perspective road image. Identity judgment of vehicle moving on each sub-region is determined by closest center point of bottom of the detected vehicle's box information. And it is tracked by vehicle's box information on the continuous images. The proposed algorithm has evaluated for performance of detection and cost in real time using day and night images captured by CCTV on the road.

• Journal of Advanced Information Technology and Convergence
• /
• v.9 no.1
• /
• pp.127-134
• /
• 2019

In terms of deep Reinforcement Learning (RL), exploration can be worked stochastically in the action of a state space. On the other hands, exploitation can be done the proportion of well generalization behaviors. The balance of exploration and exploitation is extremely important for better results. The randomly selected action with ε-greedy for exploration has been regarded as a de facto method. There is an alternative method to add noise parameters into a neural network for richer exploration. However, it is not easy to predict or detect over-fitting with the stochastically exploration in the perturbed neural network. Moreover, the well-trained agents in RL do not necessarily prevent or detect over-fitting in the neural network. Therefore, we suggest a novel design of a deep RL by the balance of the exploration with drop-out to reduce over-fitting in the perturbed neural networks.

• Aerospace Engineering and Technology
• /
• v.13 no.1
• /
• pp.120-128
• /
• 2014

Adventure of human race for space exploration is toward outer region of solar system. Recently the main targets of space explorations are becoming mainly the Mars, Venus and asteroids instead of the Moon which was the popular place human wants to explore. There are several technical challenges as spacecraft goes far and far away from the Earth, and among them communication protocol is one of the most challenging problems. In this paper, several international technical trends regarding deep space communication protocol technologies in an aspect of software implementation is presented. It is expected that these references are helpful for the development of the lunar orbiter pathfinder which is planned to be launched in 2017.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.44 no.3
• /
• pp.218-227
• /
• 2016

Korean government plans to launch a lunar orbiter and a lander to the Moon by 2020. Before launch these two proves, an experimental lunar orbiter will be launched by 2018 to obtain key space technologies for the lunar exploration. Several payloads equipped in experimental lunar orbiter will monitor the surface of the Moon and will gather science data. Lunar orbiter sends telemetry and receives tele-command from ground using S-band while science data is sent to ground stations using X-band when the visibility is available. Korean deep space network will be mainly used for S and X-band communication with lunar orbiter. Deep Space Network or Universal Space Network can also be used for the S-band during trans-lunar phase when korean deep space network is not available and will be used for the S-band in normal mission orbit as a backup. This paper analyzes a visibility condition based on the combination of various ground antennas and its mask angles according to mission scenario to predict the number of contacts per day and to build an operational scenario for the lunar orbiter.