• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.40 no.5
• /
• pp.312-320
• /
• 2003

This work describes an 8b 200 MHz 0.18 urn CMOS analog-to-digital converter (ADC) based on a pipelined architecture for flat panel display applications. The proposed ABC employs an improved bootstrapping technique to obtain wider input bandwidth than the sampling tate of 200 MHz. The bootstrapuing technique improves the accuracy of the input sample-and-hold amplifier (SHA) and the fast fourier transform (FFT) analysis of the SHA outputs shows the 7.2 effective number of bits with an input sinusoidal wave frequency of 500 MHz and the sampling clock of 200 MHz at a 1.7 V supply voltage. Merged-capacitor switching (MCS) technique increases the sampling rate of the ADC by reducing the number of capacitors required in conventional ADC's by 50 % and minimizes chip area simultaneously. The simulated ADC in a 0.18 um n-well single-poly quad-metal CMOS technology shows an 8b resolution and a 73 mW power dissipation at a 200 MHz sampling clock and a 1.7 V supply voltage.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.19 no.8
• /
• pp.116-121
• /
• 2005

To realize the tap coefficient of a finite impulse response(FIR) filter or the twiddle factor of a fast Fourier transform(FFT) using a current-mode analog circuit, a high accurate current-attenuator circuit is needed This paper introduces an accuracy enhancement technique in the current-mode signal processing. First of all, the DC of set-current error in a conventional current-attenuator using a gate-ratioed orient mirror circuit is analyzed and then, the current-attenuator circuit with a negligibly small DC offset-current error is introduced. The circuit consists of N-output current mirrors connected in parallel with me another. The output current of the circuit is attenuated to 1/N of the input current. On the basis of the Kirchhoff current law, the current scale ratio is determined simply by the number of the current mirrors in the N-current mirrors connected in parallel. In the proposed current-attenuator circuit the scale accuracy is limited by the ac gain error of the current mirror. Considering that a current mirror has a negligibly small ac gain error, the attainable maximum scale accuracy is theoretically -80[dB] to the input current.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.6
• /
• pp.759-767
• /
• 2020

This paper proposes a new Recurrent neural network (RNN) based spectrum sensing technique for cognitive radio communications. The proposed technique determines the existence of primary user's signal without any prior information of the primary users. The method performs high-speed sampling by considering the whole sensing bandwidth and then converts the signal into frequency spectrum via fast Fourier transform (FFT). This spectrum signal is cut in sensing channel bandwidth and entered into the RNN to determine the channel vacancy. The performance of the proposed technique is verified through computer simulations. According to the results, the proposed one is superior to more than 2 [dB] than the existing threshold-based technique and has similar performance to that of the existing Convolutional neural network (CNN) based method. In addition, experiments are carried out in indoor environments and the results show that the proposed technique performs more than 4 [dB] better than both the conventional threshold-based and the CNN based methods.

• Korea Science and Art Forum
• /
• v.25
• /
• pp.379-388
• /
• 2016

There are some studies to get rid of slice patterns in coronal and sagittal images by FFT(Fast Fourier Transform) and Notch filter. However, the use of a notch filter is the ringing effect according to the rapid frequency change occurs in common. Therefore, this study will be to evaluate the shape and effect of the notch filter. In addition, it is proposed a new notch filter design. Results of the study are as follows. First, Emission-Notch filter was relatively naturally correcting the cutting pattern. Second, Emission-Notch filter to use as Ringing effect is reduced relatively. The next through these findings and further research progress more completely removed the Ringing effect, and expects to reproduce the data close to the actual human body information.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.5
• /
• pp.84-93
• /
• 2007

In this paper, a Radix-8 structure for high-speed FFT is propose. Main block of the proposed FFT structure is Radix-8 DIF(Decimation In Frequency) butterfly. Even throughput of the Radix-8 FFT is twice than that of the Radix-4 FFT, implementation area of the Radix-8 is larger than that of Radix-4 FFT. But, implementation area of the proposed Radix-8 FFT was reduced by using DA(Distributed Arithmetic) for multiplication. For comparison, the 64-point FFT was implemented using conventional Radix-4 butterfly and proposed Radix-8 butterfly, respectively. The Verilog-HDL coding results for the proposed FFT structure show 49.2% cell area increment comparison with those of the conventional Radix-4 FFT structure. Namely, to speed up twice, 49.2% of area cost is required. In case of same throughput, power consumption of the proposed structure is reduced by 25.4%. Due to its efficient processing scheme, the proposed FFT structure can be used in large size of FFT like OFDM Modem.

• Journal of Multimedia Information System
• /
• v.4 no.4
• /
• pp.225-232
• /
• 2017

Understanding actions in videos is an important task. It helps in finding the anomalies present in videos such as fights. Detection of fights becomes more crucial when it comes to sports. This paper focuses on finding fight scenes in Hockey sport videos using blur & radon transform and convolutional neural networks (CNNs). First, the local motion within the video frames has been extracted using blur information. Next, fast fourier and radon transform have been applied on the local motion. The video frames with fight scene have been identified using transfer learning with the help of pre-trained deep learning model VGG-Net. Finally, a comparison of the methodology has been performed using feed forward neural networks. Accuracies of 56.00% and 75.00% have been achieved using feed forward neural network and VGG16-Net, respectively.

• Proceedings of the KIPE Conference
• /
• 2020.08a
• /
• pp.1-3
• /
• 2020

본 논문은 주파수 분석 방법 중 하나인 Discrete Wavelet Transform (DWT)을 활용하여 태양광 직렬 아크 사고를 검출하는 방법에 관하여 다룬다. DWT 알고리즘은 주파수 도메인에서도 시간 축 정보를 표현할 수 있어 기존의 Fast Fourier Transform (FFT) 주파수 분석 알고리즘과 차이점이 있으며, 대용량 태양광 시스템의 직렬 아크 사고 검출에 최적화 되도록 DWT 알고리즘의 속도를 향상시켜 태양광 DC 아크 사고 안전규격인 UL1699B의 요구 조건을 만족시켰다. DWT 알고리즘의 경우 TMS320F28033 기반으로 구현 되었으며 대용량 PV 시스템 적용을 위해 로고스키코일을 전류 센서로 사용하였다. 또한, 모의 DC 직렬 아크 발생 회로를 구축하여 제작한 사고 검출기의 성능을 실제 아크 발생 조건에서 검증하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.7
• /
• pp.653-660
• /
• 2009

This paper proposes a robust audio watermarking algorithm for copyright protection and improvement of sound quality after embedding a watermark into an original sound. The proposed method computes the FFT (fast Fourier transform) of the original sound signal and divides the spectrum into a subbands. Then, it is necessary to calculate the energy of each subband and sort n subbands in descending order corresponding to its power. After calculating the energy we choose k subbands in sorted order and find p peaks in each selected subbands, and then embed a length m watermark around the p peaks. When the listeners hear the watermarked sound, they do not recognize any distortions. Furthermore, the proposed method is robust as much as Cox's method to MP3 compression, cropping, FFT echo attacks. In addition to this, the experimental results show that the proposed method is generally 10 dB higher than Cox's method in SNR (signal-to-noise ratio) aspect.

• New & Renewable Energy
• /
• v.9 no.3
• /
• pp.5-13
• /
• 2013

In the previous research, dynamic results have been analyzed in the time and frequency regions. Time and frequency region can be transformed by the Fourier transform. This transform is very useful about analyzing system behaviors. However, because of coupling, it cannot give clear results in the real system including lots of defects. In this paper, we introduced the analysis based on quefrency region to represent physical means clearly from complicated results. We simulated the drive train system which has defects, and compared between frequency and quefrency region to show its excellence. To do this process, We established mathematical model. The equation of motion was derived by the Lagrange equation and constraint equations. The constraint equation included relationships about gear mesh, flexibility of shaft. About numerical analysis, the Newmark beta method was used to get results. And FFT (Fast Fourier Transform) which converts results from time domain to frequency, qufrequency was used.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.48 no.4
• /
• pp.53-60
• /
• 2011

In order to augment the robustness of OFDM system to Rayleigh multipath fading, there exist two smart antenna algorithms, namely, Pre-FFT smart antenna and Post-FFT smart antenna. After the mathematical modeling of both smart antenna algorithms, computer simulations have been carried to compare and analyze the performance of generalized eigen problem based Pre-FFT algorithm and the performance of Wiener solution based Post-FFT algorithm. It has been shown that the Post-FFT smart antenna far outperforms the Pre-FFT smart antenna due to the computational complexities. Especially it is so when the multipath signal arrives at beyond the guard interval and a rich co-channel interferer is introduced. Performance of a subcarrier clustering method proposed to lessen the computing load has been compared to that of a typical Wiener solution based Post-FFT smart antenna. Performance comparison between MRC(Maximum Ratio Combining) diversity based Post-FFT algorithm and typical Post-FFT algorithm has also been carried.