Browsing by Author "Muhammad, Khan"

Now showing 1 - 3 of 3

A dimension-reduction based multilayer perception method for supporting the medical decision making
(2020-03-01)
Lee, Shin-Jye
;
Tseng, Ching-Hsun
;
Lin, G. T-R
;
Yang, Yun
;
Yang, Po
Due to the rapid development of Medical IoT recently, how to effectively apply these huge amounts of IoT data to enhance the reliability of the clinical decision making has become an increasing issue in the medical field. These data usually comprise high-complicated features with tremendous volume, and it implies that the simple inference models may less powerful to be practiced. In deep learning, multilayer perceptron (MLP) is a kind of feed-forward artificial neural network, and it is one of the high-performance methods about stochastic scheme, fitness approximation, and regression analysis. To process these high uncertain data, the proposed work based on MLP structure in particular integrates the boosting scheme and dimension-reduction process. In this proposed work, the advanced ReLU-based activation function is used. Also, the weight initialization is applied to improve the stable prediction and convergence. After the improved dimension-reduction process is introduced, the proposed method can effectively learn the hidden information from the reformative data and the precise labels also can be recognized by stacking a small amount of neural network layers with paying few extra cost. The proposed work shows a possible path of embedding dimension reduction in deep learning structure with minor price. In addition to the prediction issue, the proposed method can also be applied to assess risk and forecast trend among different information systems. (C) 2019 Elsevier B.V. All rights reserved.
Elastic and cost-effective data carrier architecture for smart contract in blockchain
(2019-11-01)
Liu, Xiaolong
;
Muhammad, Khan
;
Lloret, Jaime
;
Chen, Yu-Wen
;
Yuan, Shyan-Ming
Smart contract, which could help developer deploy decentralized and secure blockchain application, is one of the most promising technologies for modern Internet of things (IoT) ecosystem today. However, Ethereum smart contract lacks of ability to communicate with outside IoT environment. To enable smart contracts to fetch off-chain data, this paper proposes a data carrier architecture that is cost-effective and elastic for blockchain-enabled IoT environment. Three components, namely Mission Manager, Task Publisher and Worker, are presented in the data carrier architecture to interact with contract developer, smart contract, Ethereum node and off-chain data sources. Selective solutions are also proposed for filtering smart contract event and decoding event log to fit different requirements. The evaluation results and discussions show the proposed system will decrease about 20USD deployment cost in average for every smart contract, and it is more efficient and elastic compared with Oraclize Oracle data carrier service. (C) 2019 Elsevier B.V. All rights reserved.
Joint Data Hiding and Compression Scheme Based on Modified BTC and Image Inpainting
(2019-01-01)
Liu, Xiaolong
;
Lin, Chia-Chen
;
Muhammad, Khan
;
Al-Turjman, Fadi
;
Yuan, Shyan-Ming
Seamlessly integrating image compression technique and secret data hiding technique into a single procedure for security and efficient data transmission is a novel research issue in modern, decentralized digital communication environments. The first joint data hiding and compression (JDHC) scheme on block truncation coding (BTC) compression domain is presented in this paper. In the compressing and embedding procedure, for the complex blocks, modified block truncation coding (MBTC) is utilized to embed secret data and compress blocks simultaneously, while further controlling the visual distortion that is caused during data embedding. For the smooth blocks, according to the current embedding bit, either image inpainting or block search order coding (BSOC) is used to embed secret data and compress blocks simultaneously with maintaining acceptable compression performance. According to the image compression codes that are provided as output, image decompression and secret bits extraction procedures can be conducted simultaneously. Experimental results indicate that outstanding compression bit rate can be achieved by the proposed JDHC scheme with satisfactory visual quality and hiding capacity.