• ETRI Journal
• /
• 제29권3호
• /
• pp.300-304
• /
• 2007

This paper proposes a new data-driven method for high-pass approaches, which suppresses slow-varying noise components. Conventional high-pass approaches are based on the idea of decorrelating the feature vector sequence, and are trying for adaptability to various conditions. The proposed method is based on temporal local decorrelation using the information-maximization theory for each utterance. This is performed on an utterance-by-utterance basis, which provides an adaptive channel normalization filter for each condition. The performance of the proposed method is evaluated by isolated-word recognition experiments with channel distortion. Experimental results show that the proposed method yields outstanding improvement for channel-distorted speech recognition.

• Current Optics and Photonics
• /
• 제6권6호
• /
• pp.608-618
• /
• 2022

By analyzing the signal-to-noise ratio (SNR) theory of the conventional true color night vision system, we found that the output image SNR is limited by the wavelength range of the system response λ₁ and λ₂. Therefore, we built a double-channel four-band true color night vision system to expand the system response to improve the output image SNR. In the meantime, we proposed an image fusion method based on principal component analysis (PCA) and nonsubsampled shearlet transform (NSST) to obtain the true color night vision images. Through experiments, a method based on edge extraction of the targets and spatial dimension decorrelation was proposed to calculate the SNR of the obtained images and we calculated the correlation coefficient (CC) between the edge graphs of obtained and reference images. The results showed that the SNR of the images of four scenes obtained by our system were 125.0%, 145.8%, 86.0% and 51.8% higher, respectively, than that of the conventional tri-band system and CC was also higher, which demonstrated that our system can get true color images with better quality.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권5호
• /
• pp.1316-1332
• /
• 2012

Formation flying is an important technology that enables high cost-effective organization of outer space aircrafts. The ad-hoc wireless network based on direct-sequence ultra-wideband (DS-UWB) techniques is seen as an effective means of establishing wireless communication links between aircrafts. In this paper, based on the theory of matched filter and error bits correction, a hybrid detection algorithm is proposed for realizing multiuser detection (MUD) when the DS-UWB technique is used in the ad-hoc wireless network. The matched filter is used to generate a candidate code set which may contain several error bits. The error bits are then recognized and corrected by an novel error-bit corrector, which consists of two steps: code mapping and clustering. In the former step, based on the modified optimum MUD decision function, a novel mapping function is presented that maps the output candidate codes into a feature space for differentiating the right and wrong codes. In the latter step, the codes are clustered into the right and wrong sets by using the K-means clustering approach. Additionally, in order to prevent some right codes being wrongly classified, a sign judgment method is proposed that reduces the bit error rate (BER) of the system. Compared with the traditional detection approaches, e.g., matched filter, minimum mean square error (MMSE) and decorrelation receiver (DEC), the proposed algorithm can considerably improve the BER performance of the system because of its high probability of recognizing wrong codes. Simulation results show that the proposed algorithm can almost achieve the BER performance of the optimum MUD (OMD). Furthermore, compared with OMD, the proposed algorithm has lower computational complexity, and its BER performance is less sensitive to the number of users.

• 한국통신학회논문지
• /
• 제31권3C
• /
• pp.286-292
• /
• 2006

직렬 합성(composition)과 병렬 합성(XOR)은 암호 스킴의 안전성을 높이기 위해 널리 사용되고 있는 방법이다. 랜덤 순열을 직렬 합성하는 회수가 많아질수록 보다 안전한 랜덤 순열이 되고, 병렬 합성하는 회수가 많아질수록 보다 안전한 랜덤 함수가 된다. 이 두가지 방법을 결합해서, 본고는 다음과 같은 일반화된 형태의 랜덤 함수를 정의한다. $SUM^s - CMP^c = ({\pi}_{sc} ... {\pi}_{(s-1)c+1}){\oplus}...{\oplus}({\pi}_c...{\pi}_1)$. 여기서, ${\pi}_1...{\pi}_{sc}$는 랜덤 순열이다. 랜덤 순열의 총 개수가 고정되어 있을 때, 직렬 합성과 병렬 합성을 각각 얼마만큼 하느냐에 따라 위 함수의 안전성은 달라질 것이다. 임의의 두 암호 스킴의 안전성을 엄밀히 비교하기 위해서는 각각의 정확한 안전성 값을 대상으로 해야 한다. 그러나, 일반적으로 정확한 값이 알려진 경우는 거의 없다. 특히, 매개변수(위 함수의 경우, s, c)의 값이 작을 경우는 밀계(tight bound)가 알려져 있는 경우가 종종 있으나, 일반적인 매개변수에 대해서는 정확한 값이나 밀계가 알려진 경우가 거의 없다. 그래서, 실제 상황에서는 두 암호 스킴의 안전성 비교는, 각각의 불안전성(insecurity)의 상계(upper bound)를 비교함으로써 이루어진다. 안전성을 중요시하는 상황에서는 더 낮은 상계를 갖는 암호 스킴을 선호하게 된다. $SUM^s - CMP^c$의 불안전성은 기존의 여러 결과들을 조합해서 계산할 수 있다. 따라서, 특정$(s_1,c_1),(s_2.c_2)$에 대한 두 함수의 안전성은 각각의 불안전성의 상계값을 계산함으로써 비교될 수 있다. 본고는 일반적인 (s, c)에 대한 $SUM^s - CMP^c$의 불안전성의 상계값의 변화를 알아보고자 한다. 그리고, 보다 낮은 상계값을 얻기 위한 직렬/병렬 합성의 최적의 개수가 무엇인지 조사한다.