http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Rubina Akter,Mohtasin Golam,Jae-Min Lee(이재민),Dong-Seong Kim(김동성) 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2
Data security, data integrity and access control are the major challenge of the C4I (command, control, communications, computer and intelligence) system. Motivated by this issue, the paper has introduced a block-chain assisted intelligent framework to authenticate and localize the unauthorized object in the surveillance area. The proposed scheme has used blockchain to deny the third party access, data falsification and intruder attacks and informs to the central control server (CCS). Concurrently, CCS received the radio frequency signal transmitted by the antenna array element to estimate the direction of arrival (DoA) to localize the unauthorized object. Hence, we design a signal model for the received signal which is processed through the convolution neural network (CNN). We also propose a CNN model, where number of layers and filters are adjusted to integrate large number of dataset and extracted features pools to output layer. According to the simulation results, the proposed algorithm outperforms to estimate DoA with their respective sources and generate spatial pseudo-spectrum with high signalto- noise ratio.
Low-Complexity Convolution Neural Network for Accurate Arrival-of-Angle Estimation in Low SNR
Rubina Akter,Van-Sang Doan,Jae-Min Lee,Dong-Seong Kim 한국통신학회 2022 한국통신학회 학술대회논문집 Vol.2022 No.2
This paper presents a low-complexity and robustly applied convolution neural network for angle of arrival (AOA) estimation for object localization. The deep network named LRCANet is cleverly designed using asymmetric convolution kernels specified in multiple residual blocks to successively learn the multi-scale feature maps of a time-variant signal correlation and the spatial relations of different antennas. According to the empirical results, LRCANet outperforms other existing models with a classification rate of 96.35% and an angle error of 0.2<SUP>0</SUP> RMSE at +5 dB signal-to-noise ratio.
Blockchain Technology for Industrial Internet of Things Based on Artificial Intelligence
Rubina Akter,Sanjay Bhardwaj,Jae Min Lee,Dong-Seong Kim 한국통신학회 2019 한국통신학회 학술대회논문집 Vol.2019 No.6
This paper proposes an energy efficient and highly secured blockchain based industrial internet of things (IIoT) system, which is distributed in nature and offer a new direction for the development of IIoT network. Typical IIoT model is based on the centralized architectural scheme, however it is fragile in terms of power consumption and secured transaction, where blockchain can be a possible solution. In this paper, we give the solution of data security and efficient power consuming mechanism through modified proof-of-work (PoW) mechanism and proper data integration mechanism. This paper analyses the dissimilar impact of IoT technology in the area of artificial intelligence. Finally the simulation results show that blockchain based IIoT scheme is more efficient than traditional static architecture in terms of throughput, latency, and security.
A Machine Learning-based Energy Prediction Scheme for Energy-Harvesting Base Station
Mohtasin Golam,Rubina Akter,Jae-Min Lee,Dong-Seong Kim 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2
Base station (BS) deployed in forest area for monitoring forest ecology faces certain challenges of limited communication. Worst more, electricity supply in forest area is always difficult and insufficient. Solar energy harvesting base station (EH-BS) can be a potential solution for this case but harvesting solar energy is challenging. To mitigate this challenge, this paper presents a novel solar energy prediction model named (SEP-Mod). The prediction model is created based on long short term memory (LSTM) neural network to learn the energy inflow autonomously for EH-BS. A dataset of four month provided by NASA HI-SEAS weather station is used to evaluate the proposed model. The results shows that the proposed model has been able to achieve nearly optimal performance and outperform existing models.