• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.605-608
• /
• 2002

A text-dependent speaker recognition system using a robust matching process is studied. The feature histogram of LPC cepstral coefficients for matching is used. The matching process uses mixture network with penalty scores. Using probability and shape comparison of two feature histograms, similarity values are obtained. The experiment results will be shown to show the effectiveness of the proposed algorithm.

• Journal of the Semiconductor & Display Technology
• /
• v.6 no.2 s.19
• /
• pp.29-33
• /
• 2007

Low frequency (400 kHz) Ferrite ICP has been proposed for large area processing. Because the coupling coefficient is close to 1, the transformer matching can be adapted to Ferrite ICP. The transformer matching system is simple. In this paper the new matching system by controlling the turns of transformer using relays is proposed for impedance matching. We developed a simple matching system and characterized it when the turns were changed. It was observed that the 2-channel relay is available for transformer matching.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.6
• /
• pp.110-116
• /
• 2014

Pattern matching algorithms are widely used in computer security systems such as computer networks, ubiquitous networks, sensor networks, and so on. However, the advances in information technology causes grow on the amount of data and increase on the computation complexity of pattern matching processes. Therefore, there is a strong demand for a novel high performance pattern matching algorithms. In light of this fact, this paper newly proposes a suffix tree based pattern matching algorithm. The suffix tree is constructed based on the suffix values of all patterns. Then, the shift nodes which informs how many characters can be skipped without matching operations are added to the suffix tree in order to boost matching performance. The proposed algorithm reduces memory usage on the suffix tree and the amount of matching operations by the shift nodes. From the performance evaluation, our algorithm achieved 24% performance gain compared with the traditional algorithm named as Wu-Manber.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.8B
• /
• pp.736-744
• /
• 2006

For the handhold devices, minimizing repetitive CPU operations such as multiplications is a major factor for their performances. In this paper, we propose efficient algorithms for finding similar sequences from streaming time-series data such as stock prices, network traffic data, and sensor network data. First, we formally define the problem of similar subsequence matching from streaming time-series data, which is called the stream sequence matching in this paper. Second, based on the window construction mechanism adopted by the previous subsequence matching algorithms, we present an efficient window-based approach that minimizes CPU operations required for stream sequence matching. Third, we propose a notion of window MBR and present two stream sequence matching algorithms based on the notion. Fourth, we formally prove correctness of the proposed algorithms. Finally, through a series of analyses and experiments, we show that our algorithms significantly outperform the naive algorithm. We believe that our window-based algorithms are excellent choices for embedded stream sequence matching in handhold devices.

• Journal of IKEEE
• /
• v.26 no.4
• /
• pp.686-692
• /
• 2022

In this paper, the matching error rates of modified area-based, energy-based algorithms, and learning-based structures were compared for stereo image matching. Census transform (CT) based on region and life propagation (BP) algorithm based on energy were selected, respectively.Existing algorithms have been improved and implemented in an embedded processor environment so that they can be used for stereo image matching in mobile systems. Even in the case of the learning base to be compared, a neural network structure that utilizes small-scale parameters was adopted. To compare the error rates of the three matching methods, Middlebury's Tsukuba was selected as a test image and subdivided into non-occlusion, discontinuous, and disparity error rates for accurate comparison. As a result of the experiment, the error rate of modified CT matching improved by about 11% when compared with the existing algorithm. BP matching was about 87% better than conventional CT in the error rate. Compared to the learning base using neural networks, BP matching was about 31% superior.

• The KIPS Transactions:PartB
• /
• v.13B no.3 s.106
• /
• pp.323-328
• /
• 2006

Proteins in a cell appear as spots in a two dimensional gel image which is used in protein analysis. The spots from the same protein are in near position when comparing two gel images. Finding out the different proteins between a normal tissue and a cancer one is important information in drug development. Automatic matching of gel images is difficult because they are made from biological experimental processes. This matching problem is known to be NP-hard. Neural networks are usually used to solve such NP-hard problems. Hopfield neural network is selected since it is appropriate to solve the gel matching. An energy function with location and distance parameters is defined. The two spots which make the energy function minimum are matching spots and they came from the same protein. The energy function is designed to reflect the topology of spots by examining not only the given spot but also neighborhood spots.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.7 no.4
• /
• pp.602-607
• /
• 2006

In this paper, the distributed network synthesis, which is useful to the design of wireless power amplifiers, is proposed, and a RF power amplifier is designed using the technique. The transfer function of distributed matching circuits is derived by Chebyshev approximation, and network element values for a specified topology are given as a function of minimum insertion losses and ripples. As an example, after a power transistor is modeled by load-pull data, the synthesis for distributed matching networks is applied to a power amplifier design, which has the electrical performance of 17dB gain and less IM3 than -43dBc at the 20W output power between 800 to 900MHz frequency range. Experimental results from a fabricated amplifier are shown to approach the design performance in the operating frequency range. The design of impedance matching networks by the transfer function synthesis is a useful method for the design of RF power amplifiers.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.7
• /
• pp.807-815
• /
• 2019

In this paper, we design a power amplifier to support quad-band in wireless communication devices using RF CMOS 180-nm process. The proposed power amplifier consists of low-band 0.9, 1.8, and 2.4 GHz and high-band 5 GHz. We proposed a structure that can support each input matching network without using a switch. For maximum linear output power, the output matching network was designed for impedance conversion to the power matching point. The fabricated quad-band power amplifier was verified using modulation signals. The long-term evolution(LTE) 10 MHz modulated signal was used for 0.9 and 1.8 GHz, and the measured output power is 23.55 and 24.23 dBm, respectively. The LTE 20 MHz modulated signal was used for 1.8 GHz, and the measured output power is 22.24 dBm. The wireless local area network(WLAN) 802.11n modulated signal was used for 2.4 GHz and 5.0 GHz. We obtain maximum linear output power of 20.58 dBm at 2.4 GHz and 17.7 dBm at 5.0 GHz.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.6
• /
• pp.800-808
• /
• 2008

A fingerprint verification system based on a set of invariant moment features and a nonlinear Back Propagation Neural Network(BPNN) verifier is proposed. An image-based method with invariant moment features for fingerprint verification is used to overcome the demerits of traditional minutiae-based methods and other image-based methods. The proposed system contains two stages: an off-line stage for template processing and an on-line stage for testing with input fingerprints. The system preprocesses fingerprints and reliably detects a unique reference point to determine a Region-of-Interest(ROI). A total of four sets of seven invariant moment features are extracted from four partitioned sub-images of an ROI. Matching between the feature vectors of a test fingerprint and those of a template fingerprint in the database is evaluated by a nonlinear BPNN and its performance is compared with other methods in terms of absolute distance as a similarity measure. The experimental results show that the proposed method with BPNN matching has a higher matching accuracy, while the method with absolute distance has a faster matching speed. Comparison results with other famous methods also show that the proposed method outperforms them in verification accuracy.

• Journal of information and communication convergence engineering
• /
• v.9 no.4
• /
• pp.447-451
• /
• 2011

Recently, various in-vehicle networks have been developed respectively in order to accomplish their own purposes such as CAN and MOST. Especially, the MOST network is usually adapted to provide entertainment service. The car navigation system is also widely used for guiding driving paths to driver. The position for the navigation system is usually acquired by GPS technology. However, the GPS technique has two serious problems. The first is unavailability in urban canyons. The second is inherent positional error rate. The problems have been studied in many literatures. However, the second still leads to incorrect locational information in some area, especially parallel roads. This paper proposes a performance tuning method of image matching algorithm for the car navigation system. The method utilizes images obtained from in-vehicle MOST network and a real-time image matching algorithm which determines the direction of moving vehicle in parallel section of road. In order to accuracy improvement of image matching algorithm, three conditions are applied. The experimental tests show that the proposed system increases the accuracy.