• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.04b
• /
• pp.1365-1368
• /
• 2002

OFDM은 제4세대 변조기술로 일컬어지는 방식이다. 이는 최근 유럽에서 디지털 오디오 방송(DAB)과 디지털 비디오 방송(DVB)에 표준이 되었으며, IEEE 802.11a 무선 LAN 에서도 이 방식을 채택했고, ADSL, VDSL 등에서도 사용되어지고 있다. 본 논문에서는 이러한 OFDM 방식의 핵심이라고 할 수 있는 64포인트 FFT(Fast Fourier Transform)하드웨어 프로세서의 여러 가지의 구현된 예를 비교 분석하고, 가장 효율적인 방법인 Radix-2 SDF(Singlepath Delay Feedback)[1] 방법을 개선하여 새로운 구조를 제안하였다. 동일한 속도 성능을 가지는 여러구조 중에서 적은 수의 지연소자를 활용하여 FFT 크기를 작게 한 것이 SDF 방식으로 가장 널리 사용되고 있다. 본 논문에서는 SDF 방식이 내부적으로 4개의 복소곱셈기를 필요로 하는데 비해 2개의 복소곱셈기만을 사용하는 구조로 변형하고 컨트롤을 조절하여 새로운 구조를 설계하였다. 구현한 결과, FFT에서 전체 구조의 약 80%를 차지하는 복소곱셈기의 수를 절반으로 줄여 FFT 하드웨어 크기를 SDF 방식의 60% 정도로 줄일 수 있게 되었고, 이러한 구현방식은 64포인트 FFT만이 아닌 더 큰 크기의 FFT를 구현함에 있어서도 동일하게 적용할 수 있으며 현재 국내외에 발표된 논문 중 성능 대 면적비가 가장 우수한 구조이다.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.12 no.4
• /
• pp.503-512
• /
• 2001

A new adaptation algorithm for the digital predistorter is presented. The proposed technique employs the real-time transmitted signals through HPA for the table update. Hence the proposed algorithm does not depend on modulation format and the signal transmission of HPA is not affected during the adaptation process. The experimental results are presented to verify the proposed method and we obtained about 8.4 dB of ACPR improvement.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.9C
• /
• pp.817-824
• /
• 2006

The crest factor properties of 16-QAM modulated one- and two-code assisted multi-carrier code-division multiple-access (MC-CDMA) signals employing complementary pair as spreading sequences are characterized. It is shown that a set of relationship between the two 16-QAM symbols entirely characterize the power envelope waveforms of the signals. There exists 60 different sets of relationship, which results in 16 different crest factors as a result of various equivalent transforms on the corresponding message symbols. It is also shown that the individual crest factor corresponding to each message combination is always bounded by 3dB.

• Korean Journal of Optics and Photonics
• /
• v.17 no.2
• /
• pp.191-197
• /
• 2006

We analyze the self-pulsation(SP) characteristics due to mode beating of two modes emitted in a multi-section complex-coupled (CC) DFB laser composed of two DFB sections and a phase control section between them. SP frequency due to mode beating of the two modes is determined by the difference of grating periods in the two CC DFB regions. As the difference of grating periods in the two CC DFB regions increases, the SP frequency increases from very low frequency to the THz region. In the case of a mode which is not located in the stop band of the other DFB region, the mode propagates into the other DFB region without a high reflection, so that output powers emitted in a multi-section CC DFB laser have high modulation indexes due to the large interaction between the two modes.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.12 no.7
• /
• pp.1086-1093
• /
• 2001

A new linearization technique for RF high-power amplifiers(HPAs) using n-th order error signal generator (ESGn) is proposed. The n-th order ESG generates an error signal based on the complex envelope transfer characteristics of the HPA, which is combined at the output of the HPA. Therefore, the higher-order nonlinearlities are not affected by the ESG$\_$n/ and the stability of the linearized system is guaranteed due to the inherent open-loop configuration. Moreover, the output delay loss can be avoided, because the error signal is generated with the input signal of the HPA. The IMD(intermodulation distortion) improvement obtained applying the ESG$\_$7/ to 5 W class A HPA in cellular band demonstrates the feasibility of the proposed postdistortion system.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.25 no.6
• /
• pp.700-708
• /
• 2014

Jet engine modulation(JEM) is a frequency modulation phenomenon of the radar signal induced by electromagnetic scattering from a rotating jet engine turbine. Although JEM can be used as a representative radar target recognition method by providing unique information on the target, its recognition performance may be degraded in the observation range of weakly present JEM. Hence, this paper presents a method for extracting the JEM component by decomposing the radar signal into intrisic mode functions(IMFs) via complex empirical mode decomposition(CEMD) and by combining them based on signal eccentricity. Its application to various signals demonstrated that the proposed method improved the clarity of JEM analysis and could extend the effective observation range of JEM.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.18 no.1
• /
• pp.20-25
• /
• 2008

Acoustic holography exhibits the spatial distribution of sound pressure in time or frequency domain. The obtained picture often contains far more than what we need in practice. For example. when we need to know only the locations and overall propagation pattern of sound sources. a method to show only what we need has to be introduced. One way of obtaining the necessary information is to use envelope in space. The spatial envelope is a spatially slowly-varying amplitude of acoustic waves which contains the information of sources' location. A spatial modulation method has been theoretically developed to get a spatial envelope. By applying the spatial envelope. not only the necessary information is obtained but also computation time is reduced during the process of holography. The spatial envelope is verified as an effective visualization scheme in time domain by being applied to complicated sound fields.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.17 no.8
• /
• pp.693-700
• /
• 2007

We can predict spatial acoustic pressure distribution on the plane of interest by using acoustic holography. However, the information embedded in the distribution plot is usually much more than what we need: for example, source locations and their overall propagation pattern. One possible candidate to solve the problem is complex envelope analysis. Complex envelope analysis extracts slowly-varying envelope signal from a band signal. We have attempted to extend this method to space domain so that we can have spatial information that we need. We have to modulate two dimensional data for obtaining spatial envelope. Although spatial modulation basically follows the same concept that is used in time domain, the algorithm for the spatial modulation turns out to be different from temporal modulation. We briefly describe temporal complex envelope analysis and extend it to spatial envelope of 2-D acoustic field by introducing geometric transformation. In the end, the results of applying the spatial envelope to the holography are envisaged and verified.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.11
• /
• pp.1380-1386
• /
• 2016

In this paper, we propose and evaluate OFDM(Orthogonal Frequency Division Multiplexing)-SSB(Singe Side Band) for improving spectral efficiency. The proposed system is based on OFDM system using DCT(Discrete Cosine Transform) and DHT(Discrete Hilbert Transform) for SSB modulation, this proposed system transmit a BPSK modulation signal through a single sideband and BPSK modulation signal through the other single sideband. In other words, we design and evaluate the proposed system which transmit two BPSK signal through each other single sideband. Also each BPSK BER performance is similar with theoretical BPSK BER performance. the proposed system transmits BPSK signal using half of spectrum of conventional OFDM system or transmit two BPSK signal in same time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.4
• /
• pp.649-659
• /
• 1994

The channel capacity of fadingchannel becomes less than that of Gaussian noise channel. Moreover, the channel capacity of system with modulated signal is expected to be much less than that of system with nonmodulated signal. It is very difficult to evaluate the capacity in fading channel with modulated signal. The channel capacity for M-ary PSK signal has already been derived in Gaussian noise channel but it is very complicate form composed of complex integral formula. In this paper, we have derived the simple form of channel capacity with M-ary PSK signal by using the average symbol error rate. And we also evaluate the capacity in Rician and m-distribution fading channels as well as Gaussian noise channel. We have verified that the channel capacity obtained are almost the same as in other literature. As a results, we show that the channel capacity of M-art PSK signal increases to converge to the number of bits per symbol (log M) as signal to-noise power ratio increases.