• Proceedings of the Korea Multimedia Society Conference
• /
• 2012.05a
• /
• pp.343-344
• /
• 2012

With the increase of sensitive data and their secure transmission and storage, the use of encryption techniques has become widespread. The performance of encoding majorly depends on the computational time, so a system with less computational time suits more appropriate as compared to its contrary part. Double Random Phase Encoding (DRPE) is an algorithm with many sub functions which consumes more time when executed serially; the computation time can be significantly reduced by implementing important functions in a parallel fashion on Graphics Processing Unit (GPU). Computing convolution using Fast Fourier transform in DRPE is the most important part of the algorithm and it is shown in the paper that by performing this portion in GPU reduced the execution time of the process by substantial amount and can be compared with MATALB for performance analysis. NVIDIA graphic card GeForce 310 is used with CUDA C as a programming language.

• Journal of Broadcast Engineering
• /
• v.27 no.7
• /
• pp.1011-1020
• /
• 2022

Recent high-performance panoptic segmentation models are based on transformer architectures. However, transformer-based panoptic segmentation methods are basically slower than convolution-based methods, since the attention mechanism in the transformer requires quadratic complexity w.r.t. image resolution. Also, sine and cosine computation for positional embedding in the transformer also yields a bottleneck for computation time. To address these problems, we adopt three modules to speed up the inference runtime of the transformer-based panoptic segmentation. First, we perform channel-level reduction using depth-wise separable convolution for inputs of the transformer decoder. Second, we replace sine and cosine-based positional encoding with convolution operations, called conv-embedding. We also apply a separable self-attention to the transformer encoder to lower quadratic complexity to linear one for numbers of image pixels. As result, the proposed model achieves 44% faster frame per second than baseline on ADE20K panoptic validation dataset, when we use all three modules.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.1
• /
• pp.40-45
• /
• 2007

In this paper, aim to suggest the medical information wireless transmission system to provide the mobility of medical service by means of wireless area network which makes it possible, home or at a long range, to check and oversee the state of patients, and to can out a simulation. The proposed method converts medical information to digital data in an emergency and sends them to mobile terminals such as PDAs to make possible swift first aid. The simulation took advantage of the OFDM transmission method based on IEEE 802.1la in order to send reliable medical information in mobile wireless channel environment, and analyzed the system performance by applying convolution encoding to transmit reliable information in AWGN and 3-ray mobile multipath fading channel environment.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.9
• /
• pp.2458-2482
• /
• 2023

Medical image segmentation techniques based on convolution neural networks indulge in feature extraction triggering redundancy of parameters and unsatisfactory target localization, which outcomes in less accurate segmentation results to assist doctors in diagnosis. In this paper, we propose a multi-level semantic-rich encoding-decoding network, which consists of a Pooling-Conv-Former (PCFormer) module and a Cbam-Dilated-Transformer (CDT) module. In the PCFormer module, it is used to tackle the issue of parameter explosion in the conservative transformer and to compensate for the feature loss in the down-sampling process. In the CDT module, the Cbam attention module is adopted to highlight the feature regions by blending the intersection of attention mechanisms implicitly, and the Dilated convolution-Concat (DCC) module is designed as a parallel concatenation of multiple atrous convolution blocks to display the expanded perceptual field explicitly. In addition, MultiHead Attention-DwConv-Transformer (MDTransformer) module is utilized to evidently distinguish the target region from the background region. Extensive experiments on medical image segmentation from Glas, SIIM-ACR, ISIC and LGG demonstrated that our proposed network outperforms existing advanced methods in terms of both objective evaluation and subjective visual performance.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.2271-2272
• /
• 2008

Recently, LDPC codes received a lot of attention in 4G. LDPC codes perform good error correction at high SNR. But LDPC codes are complex design and not good at low SNR. At low SNR, convolution codes and turbo codes show more good performance than LDPC codes. The main subject presented in this study is that parallel encoding and decoding according to SNR. The system chooses convolution codes at low SNR and chooses LDPC codes at high SNR.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.1
• /
• pp.41-48
• /
• 2004

In this paper, we analyzed channel performance of PPM modulated UWB communication system in indoor radio fading environment that consider amplitude characteristic of channel. Fading channel considered various channel environments by fading index m utilizing Nakagami-m distribution model with data through an UBW radio signal experiment that announced in existing. Also, we improved performance of system that it is decreased in fading environment employing convolution encoding techniques.

• Journal of Digital Contents Society
• /
• v.7 no.3
• /
• pp.169-174
• /
• 2006

In recent years, RFID is widely used in industrial applications including factory, material flow, logistics and defense areas. In this paper, The convolutional encoding and viterbi decoding is also implemented to improve the system performance. in an FPGA chip. The used convolution code is constraint length K=3 and rate R=1/2. The length of command frame and response frame is total of 48bits consisting of SOF 8 bits, command 16 bits, CRC 16 bit, and EOF 8 bits. And also the frame error rates are measured under the channel of line-of-sight and non line-of-sight, respectively. The performances are analyzed with FSK modulation only and FSK modulation added with convolutional encoding. These two measured results are compared with that of a RFID system with ASK modulation.

• Journal of the Institute of Convergence Signal Processing
• /
• v.7 no.1
• /
• pp.17-23
• /
• 2006

In recent years, RFID has received much attention because of spread usage in industrial applications including factory, material flow, logistics and defense areas. However, there is only CRC-16 for error detection in ISO/IEC 18000-6 Protocols prepared for 860-960 MHz RFID, high error rates are expected in cases of high level of security and noisy envirionment. In this paper, we propose a usage of convolution code as a method for satisfying the high level of security requirement and system error performance.'1'he signal control function is implemented in a microprocessor with RF modulation and the convolutional encoding and Viterbi decoding are implemented in an FPGA chip.'The frame error rates are measured with and without convolution coding under the channel conditions of line-of- sight and non line-of-sight, respectively.

• Aerospace Engineering and Technology
• /
• v.8 no.1
• /
• pp.179-186
• /
• 2009

This paper describes the design and implementation for spread spectrum FTS encoder and decoder. The FTS command format is defined by 64 bit encrypted packet that contains all required information relayed between the ground and the vehicle. Encryption is accomplished using the Tripple-DES encryption algorithm in block encryption form. The proposed FTS encoder and decoder is using the Convolution Encoding and Viterbi Decoding for forward error correction. The Spread Spectrum Modulation is done using a PN code, which is 256 bit gold code. The simulation result shows that the designed FTS decoder is compatible with the designed FTS encoder.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.4
• /
• pp.156-162
• /
• 2022

Image compression plays an important role in encoding and improving various forms of images in the digital era. Recent researches have focused on the principle of deep learning as one of the most exciting machine learning methods to show that it is good scheme to analyze, classify and compress images. Various neural networks are able to adapt for image compressions, such as deep neural networks, artificial neural networks, recurrent neural networks and convolution neural networks. In this review paper, we discussed how to apply the rule of deep learning to obtain better image compression with high accuracy, low loss-ness and high visibility of the image. For those results in performance, deep learning methods are required on justified manner with distinct analysis.