• Journal of Broadcast Engineering
• /
• v.14 no.5
• /
• pp.543-550
• /
• 2009

The most widely used block convolution method is the overlap save algorithm (OSA), where a block of M data to be convolved with a filter is concatenated with the previous block and 2M-point FFT and multiplications are performed for this overlapped block. By discarding half of the results, we obtain linear convolution results from the circular convolution. This paper proposes a new transform which reduces the block size to only M for the block convolution. The proposed transform can be implemented as the M multiplications followed by M-point FFT Hence, existing efficient FFT libraries and hardware can be exploited for the implementation of proposed method. Since the required transform size is half that of the conventional method, the overall computational complexity is reduced. Also the reduced transform size results in the reduction of data access time and cash miss-hit ratio, and thus the overall CPU time is reduced. Experiments show that the proposed method requires less computation time than the conventional OSA.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.6
• /
• pp.1231-1243
• /
• 1997

Digital satellite communication channels have nonlinearities with memory due to saturation characteristics of traveling wave tube amplifier in the satellite and transmitter/receiver linear filters. In this paper, we propose a network structure and a learning algorithm for complex pi-sigma network (CPSK) and exploit CPSN in the problem of equalization of nonlinear satellite channels. The proposed CPSN is a complex-valued extension of real-valued pi-sigma network that is a higher-order feedforward network with fast learning while greatly reducing network complexity by utilizing efficient form of polynomials for many input variables. The performance of the proposed CPSN is demonstrated by computer simulations on the equalization of complex-valued QPSK input symbols distorted by a nonlinear channel modeled as a Volterra series and additive noise. The results indicate that the CPSN shows good equalization performance, fast convergence, and less computations as compared to conventional higher-order models such as Volterra filters.

• Korean Journal of Remote Sensing
• /
• v.33 no.4
• /
• pp.423-436
• /
• 2017

This paper presents a deep learning-based road segmentation framework from very high-resolution orthophotos. The proposed method uses Deep Convolutional Autoencoders for end-to-end mapping of orthophotos to road segmentations. In addition, a set of post-processing steps were applied to make the model outputs GIS-ready data that could be useful for various applications. The optimization of the model's parameters is explained which was conducted via grid search method. The model was trained and implemented in Keras, a high-level deep learning framework run on top of Tensorflow. The results show that the proposed model with the best-obtained hyperparameters could segment road objects from orthophotos at an average accuracy of 88.5%. The results of optimization revealed that the best optimization algorithm and activation function for the studied task are Stochastic Gradient Descent (SGD) and Exponential Linear Unit (ELU), respectively. In addition, the best numbers of convolutional filters were found to be 8 for the first and second layers and 128 for the third and fourth layers of the proposed network architecture. Moreover, the analysis on the time complexity of the model showed that the model could be trained in 4 hours and 50 minutes on 1024 high-resolution images of size $106{\times}106pixels$, and segment road objects from similar size and resolution images in around 14 minutes. The results show that the deep learning models such as Convolutional Autoencoders could be a best alternative to traditional machine learning models for road segmentation from aerial photographs.

• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.1511_1512
• /
• 2009

This paper presents a polarization switching antenna integrated with silicon RF MEMS SPDT switches in the form of a package. A low-loss quartz substrate made of SoQ (silicon-on-quartz) bonding is used as a dielectric material of the patch antenna, as well as a packaging lid substrate of RF MEMS switches. The packaging/antenna substrate is bonded with the bottom substrate including feeding lines and RF MEMS switches by BCB adhesive bonding, and RF energy is transmitted from signal lines to antenna by slot coupling. Through this approach, fabrication complexity and degradation of RF performances of the antenna due to the parasitic effects, which are all caused from the packaging methods, can be reduced. This structure is expected to be used as a platform for reconfigurable antennas with RF MEMS tunable components. A linear polarization switching antenna operating at 19 GHz is manufactured based on the proposed method, and the fabrication process is carefully described. The s-parameters of the fabricated antenna at each state are measured to evaluate the antenna performance.

• Journal of Biomedical Engineering Research
• /
• v.28 no.3
• /
• pp.429-440
• /
• 2007

A coded excitation has been studied to improve the performance for ultrasound imaging in term of SNR, imaging frame rate, contrast to tissue ratio, and so forth. However, it requires a complicated arbitrary waveform transmitter for each active channel that is typically composed of a multi-bit Digital-to-Analog Converter (DAC) and a linear power amplifier (LPA). Not only does the LPA increase the cost and size of a transmitter block, but it consumes much power, increasing the system complexity further and causing a heating-up problem. This paper proposes an optimized 1.5bit fourth order sigma-delta modulation technique applicable to design an efficient arbitrary waveform generator with greatly reduced power dissipation and hardware. The proposed SDM can provide a required SQNR with a low over-sampling ratio of 4. To this end, the loop coefficients are optimized to minimize the quantization noise power in signal band while maintaining system stability. In addition, the decision level for the 1.5 bit quantizer is optimized for a given input waveform, which results in the SQNR improvement of more than 5dB. Computer simulation results show that the SQNR of a FM(frequency modulated) signal generated by using the proposed method is about 26dB, and the peak side-lobe level (PSL) of its compressed waveform on receive is -48dB.

• Journal of Computing Science and Engineering
• /
• v.5 no.2
• /
• pp.111-120
• /
• 2011

Data mining algorithms should exploit new hardware technologies to accelerate computations. Such goal is difficult to achieve in database management system (DBMS) due to its complex internal subsystems and because data mining numeric computations of large data sets are difficult to optimize. This paper explores taking advantage of existing multithreaded capabilities of multicore CPUs as well as caching in RAM memory to efficiently compute summaries of a large data set, a fundamental data mining problem. We introduce parallel algorithms working on multiple threads, which overcome the row aggregation processing bottleneck of accessing secondary storage, while maintaining linear time complexity with respect to data set size. Our proposal is based on a combination of table scans and parallel multithreaded processing among multiple cores in the CPU. We introduce several database-style and hardware-level optimizations: caching row blocks of the input table, managing available RAM memory, interleaving I/O and CPU processing, as well as tuning the number of working threads. We experimentally benchmark our algorithms with large data sets on a DBMS running on a computer with a multicore CPU. We show that our algorithms outperform existing DBMS mechanisms in computing aggregations of multidimensional data summaries, especially as dimensionality grows. Furthermore, we show that local memory allocation (RAM block size) does not have a significant impact when the thread management algorithm distributes the workload among a fixed number of threads. Our proposal is unique in the sense that we do not modify or require access to the DBMS source code, but instead, we extend the DBMS with analytic functionality by developing User-Defined Functions.

• Journal of Korea Multimedia Society
• /
• v.6 no.3
• /
• pp.419-428
• /
• 2003

Several methods have been studied to prune the parasitic branches that cause unfortunately from thinning a shape to get its skeleton. We found that the symmetric path finding method was most efficient because it followed the boundary pixels of the shape just once. In this paper, its extended method is proposed to apply to removing the noise branches that protrude out of the boundary of a segmented or extracted shape in a given image. The proposed method can remove a noise branch with one-pixel width and also remove the noise branch that includes a round shape called a noise blob. The method uses a 4-8-directional boundary-following technique to determine symmetric paths and finds noise branches with noise blobs by detecting quasi-symmetric paths. Its time complexity is a linear function of the number of boundary pixels. Interactively selectable parameters are used to define various types of noise branches flexibly, which are the branch - size parameter and the blob-size parameter. Experimental results for a practical shape and various artificial shapes showed that the proposed method was very useful for simplifying the shapes.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.3
• /
• pp.63-69
• /
• 2012

Given art gallery P with n vertices, the maximum (sufficient) number of mobile guards is${\lfloor}n/4{\rfloor}$ for simple polygon and${\lfloor}(3n+4)/16{\rfloor}$ for simple orthogonal polygon. However, there is no polynomial time algorithm for minimum number of mobile guards. This paper suggests polynomial time algorithm for the minimum number of mobile guards. Firstly, we obtain the visibility graph which is connected all edges if two vertices can be visible each other. Secondly, we select vertex u with ${\Delta}(G)$ and v with ${\Delta}(G)$ in $N_G(u)$ and delete visible edges from u,v and incident edges. Thirdly, we select $w_i$ in partial graphs and select edges that is the position of mobile guards. This algorithm applies various art galley problems with simple polygons and orthogonal polygons art gallery. As a results, the running time of proposed algorithm is linear time complexity and can be obtain the minimum number of mobile guards.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.4
• /
• pp.708-717
• /
• 2017

Tracking of sound-location is widely used in various area such as intelligent CCTV, video conference and voice commander. In this paper we introduce the real-time sound-location tracking method for smart lecture system using TDoA(Time Difference of Arrival) with orthogonal microphone array on the ceiling. Through discussion on some models of TDoA detection, cross correlation method using linear microphone array is proposed. Orthogonal array with 5 microphone could detect omni direction of sound-location. For real-time detection we adopt the threshold of received energy for eliminating no-voice interval, signed cross correlation for reducing computational complexity. The detected azimuth angles are processed using median filter for lowering the angle deviation. The proposed system is implemented with high performance MCU of TMS320F379D and MEMs microphone module and shows the accuracy of 0.5 and 6.5 in degree for white noise and lectured voice, respectively.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.3
• /
• pp.533-545
• /
• 2017

Quantile regression models provide a variety of useful statistical information by estimating the conditional quantile function of the response variable. However, the traditional linear quantile regression model can lead to the distorted and incorrect results when analysing real data having a nonlinear relationship between the explanatory variables and the response variables. Furthermore, as the complexity of the data increases, it is required to analyse multiple response variables simultaneously with more sophisticated interpretations. For such reasons, we propose a multivariate quantile regression tree model. In this paper, a new split variable selection algorithm is suggested for a multivariate regression tree model. This algorithm can select the split variable more accurately than the previous method without significant selection bias. We investigate the performance of our proposed method with both simulation and real data studies.