• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.9
• /
• pp.2505-2528
• /
• 2023

The field of facial expression recognition (FER) has been actively researched to improve human-computer interaction. In recent years, deep learning techniques have gained popularity for addressing FER, with numerous studies proposing end-to-end frameworks that stack or widen significant convolutional neural network layers. While this has led to improved performance, it has also resulted in larger model sizes and longer inference times. To overcome this challenge, our work introduces a novel lightweight model architecture. The architecture incorporates three key factors: Depth-wise Separable Convolution, Residual Block, and Attention Modules. By doing so, we aim to strike a balance between model size, inference speed, and accuracy in FER tasks. Through extensive experimentation on popular benchmark FER datasets, our proposed method has demonstrated promising results. Notably, it stands out due to its substantial reduction in parameter count and faster inference time, while maintaining accuracy levels comparable to other lightweight models discussed in the existing literature.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.20 no.41
• /
• pp.1-14
• /
• 1997

This study proposes a genetic algorithm using Pr(equation omitted)fer number for the traveling salesman problem(PNGATSP). Nearest neighbor nodes are mixed with randomly selected nodes at the stage of generating initial solutions. Proposed PNGATSP adopts a few ideas which are different from traditional genetic algorithms. For instance, an exponential fitness function and elitism are used and Pr(equation omitted)fer number is used for encoding TSP. Genetic operators are selected by experiments, which make a good solution among four combinations of conventional genetic operators and new genetic operators. For respective combinations, robust set of parameters is determined by the experimental designing approach. The feature of Pr(equation omitted)fer number code for TSP and the search power of GA using Pr(equation omitted)fer number is analysed. The best is a combination of OX(order crossover) and swap, which is superior to the other experimented combinations of genetic operators by 1.0%∼12.8% deviation.

• Bulletin of the Korean Mathematical Society
• /
• v.53 no.5
• /
• pp.1447-1455
• /
• 2016

In this note, we mainly discuss the Gorenstein $Pr{\ddot{u}}fer$ domains. It is shown that a domain is a Gorenstein $Pr{\ddot{u}}fer$ domain if and only if every finitely generated ideal is Gorenstein projective. It is also shown that a domain is a PID (resp., Dedekind domain, $B{\acute{e}}zout$ domain) if and only if it is a Gorenstein $Pr{\ddot{u}}fer$ UFD (resp., Krull domain, GCD domain).

• Korean Journal of Mathematics
• /
• v.18 no.2
• /
• pp.185-193
• /
• 2010

Let D be an integral domain, and let ${\bar{D}}$ be the integral closure of D. We show that if D is an almost pseudo-valuation domain (APVD), then D is a quasi-$Pr{\ddot{u}}fer$ domain if and only if D=P is a quasi-$Pr{\ddot{u}}fer$ domain for each prime ideal P of D, if and only if ${\bar{D}}$ is a valuation domain. We also show that D(X), the Nagata ring of D, is a locally APVD if and only if D is a locally APVD and ${\bar{D}}$ is a $Pr{\ddot{u}}fer$ domain.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.4A
• /
• pp.208-215
• /
• 2003

In this paper, the target FER is derived to achieve the maximum throughput in H-ARQ scheme where the FEC and ARQ are properly combined and proved through the simulation. In numerical analysis, The average upper bound on the FER(Frame Error Rate) of turbo code is derived and the throughput is expressed in terms of this bound. Then, the throughput is differentiated with respect to the FER and the optimal target PER is achieved. As a result of analysisw, the throughput is maximized at the target FER of 6% whereas the target PER is set to 5% in the current 3GPP2 specification.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.26 no.11A
• /
• pp.1848-1857
• /
• 2001

This paper proposes the ETEO and MTEO burst signal detector based on TEO algorithm. These algorithms must be used after STR and AGC operation, but are not related to phase and frequency offset. ETEO algorithm is extended version of original TEO, and MTEO algorithm is proposed for improving the output characteristics of ETEO. Also, modified ETEO and MTEO algorithm are proposed for detection of PREAMBLE64. Optimal threshold value is determined and miss and false alarm probability and FER performance are evaluated by computer simulation. Finally, this paper proposes MTEO algorithm with M=3 to guarantee the Performance that FER is less than 10$\^$-2/.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.12
• /
• pp.2534-2542
• /
• 2011

In this paper, we show that performance can be improved by using multiple antennas in the conventional orthogonal code hopping multiplexing (OCHM) scheme, which was proposed for accommodating a larger number of users with low channel activities than the number of orthogonal codewords used in code division multiple access (CDMA)-based communication systems through downlink statistical multiplexing. First, we introduce two different types of OCHM systems together with orthogonal codeword allocation strategies, and then derive their mathematical expression for log-likelihood ratio (LLR) values according to the two different schemes. Next, when a turbo encoder based on the LLR computation is used, we evaluate performance on the frame error rate (FER) for the aformentioned OCHM system. For comparison, we also show performance for the existing symbol mapping method using multiple antennas, which was used in 3GPP standards. As a result, it is shown that our OCHM system with multiple antennas based on the proposed orthogonal codeword allocation strategy leads to performance gain over the conventional system---energy required to satisfy a target FER is significantly reduced.

• Molecules and Cells
• /
• v.40 no.12
• /
• pp.976-985
• /
• 2017

Iron is an essential divalent ion for aerobic life. Life has evolved to maintain iron homeostasis for normal cellular and physiological functions and therefore imbalances in iron levels exert a wide range of consequences. Responses to iron dysregulation in blood development, however, remain elusive. Here, we found that iron homeostasis is critical for differentiation of Drosophila blood cells in the larval hematopoietic organ, called the lymph gland. Supplementation of an iron chelator, bathophenanthroline disulfate (BPS) results in an excessive differentiation of the crystal cell in the lymph gland. This phenotype is recapitulated by loss of Fer1HCH in the intestine, indicating that reduced levels of systemic iron enhances crystal cell differentiation. Detailed analysis of Fer1HCH-tagged-GFP revealed that Fer1HCH is also expressed in the hematopoietic systems. Lastly, blocking Fer1HCH expression in the mature blood cells showed marked increase in the blood differentiation of both crystal cells and plasmatocytes. Thus, our work suggests a relevance of systemic and local iron homeostasis in blood differentiation, prompting further investigation of molecular mechanisms underlying iron regulation and cell fate determination in the hematopoietic system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.178-203
• /
• 2018

In this paper, authors have been evaluated the Frame Error Rate (FER) performance of IEEE 802.11 a/g/p standard 5 GHz frequency band WLAN over Rayleigh and Rician distributed fading channels in presence of Additive White Gaussian Noise (AWGN). Orthogonal Frequency Division Multiplexing (OFDM) based transceiver is implemented by using real-time signal processing frameworks (IEEE 802.11 Blocks) in GNU Radio Companion (GRC) and Ettus USRP N200 is used to process the symbol over the wireless radio channel. The FER is calculated for each sub-carrier conventional modulation schemes used by OFDM such as BPSK, QPSK, 16, 64-QAM with different punctuated coding rates. More precise SNR is computed by modifying the SNR calculation process of YANS and NIST error rate model to estimate more accurate FER. Here, real-time signal constellations, OFDM signal spectrums etc. are also observed to find the effect of multipath propagation of signals through flat and frequency selective fading channels. To reduce the error rate due to the multipath fading effect and Doppler shifting, channel estimation (CE) and equalization techniques such as Least Square (LS) and training based adaptive Least Mean Square (LMS) algorithm are applied in the receiver. The simulation work is practically verified at GRC by turning into a pair of Software Define Radio (SDR) as a simultaneous transceiver.

• Current Research on Agriculture and Life Sciences
• /
• v.21
• /
• pp.23-29
• /
• 2003

Worldwide consumption of vegetable soybean bas been increasing recently, hence it is necessary to produce good quality of soybean in our farms. In the process of vegetable soybean production threshing and seperation work accounts for about 80% of overall labor. Therefore, developing of the vegetable soybean thresher is necessary to reduce the cost of labor. The purpose of this study is to acquire the basic informations to design of the vegetable soybeans-thresher. We make the experimental system which measure the physical properties and investigate the detachment forces. Also, We calculated the minimum speed of threshing cylinder. The result are as follows; 1. The average length of soybean stem is 68.2cm. 2. The length of soybean pods are seen as 61.3mm for 3 grain, 52.6mm fer 2 grains and 41.0mm fer 1 grain 3. The widths of soybean pods are seen as 14.1mm fer 3 grain, 13.8mm fer 2 grains and 13.4mm fer 1 grain. 4. The weights of soybean pods are seen as 4.1grams for 3 grains, 2.7grams for 2 grains and 1.4grams for 1 grain. 5. The average detachment forces of pods are seen as 1.5kgf for 3 grains, 1.2kgf for 2 grains and 0.8kgf for 1 grain respectively For 1 grain, the detachment force of pods ranges from 0.2kgf to 1.4kgf. For 2 grains, the minimum detachment force of pods is seen as 0.6kgf and the maximum one is seen as 2.5kgf. For 3 grains, the minimum detachment force of pods is seen as 0.7kgf and the maximum one is seen as 2.7kgf. 6. The minimum speed of threshing cylinder is shown 6.83m/s.