Earticle

The genetic algorithm (GA), one of the artificial intelligence (AI), is developed based on Darwin's theory of evolution, i.e., the mating of randomly selected objects. If more optimal solution is generated, then it is better to repeat the process of setting the optimum value. In this paper, the method of background music using the genetic algorithm is exploited when the computer game is executed each time. As a result, it has created several music that can be used in the actual game, and it could be confirmed that the other music that is created is different music when performed each time.

본 논문에서는 자동차 환경의 인포테인먼트 시스템을 위한 음악 검색 알고리즘을 제안한다. 제안된 방법은 음악신호의 로그 스펙트럼기반의 정점을 이용하여 오디오지문을 추출하고, 추출된 음악 핑거프린트에 해시값을 적용하여 클라우드 서버에 저장한다. 클라우드 서버에서는 사용자의 쿼리 음악과 클라우드 서버의 해시 테이블에 저장되어 있는 오디오 지문을 비교함으로써 가장 유사한 음악이 검색된다. 제안된 음악 검색 알고리즘의 성능평가를 위해, 주행 중인 자동차 내부에서 녹음한 잡음에 노출된 다양한 쿼리 음악의 길이에 따른 검색 결과의 정확도를 측정하였고, 해시 테이블의 저장 곡수에 따른 검색 소요 시간을 측정하였다.

In this paper, we propose a music search algorithm for automotive infotainment system. The proposed method extracts fingerprints using the high peaks based on log-spectrum of the music signal, and the extracted music fingerprints store in cloud server applying a hash value. In the cloud server, the most similar music is retrieved by comparing the user's query music with the fingerprints stored in hash table of cloud server. To evaluate the performance of the proposed music search algorithm, we measure an accuracy of the retrieved results according to various length of the query music and measure a retrieval time according to the number of stored music database in hash table.

맞춤형 교육을 위해 학습자에 맞는 학습 경로를 탐색하는 것은 필수적이다. 유전 알고리즘은 해공간이 매우 커서 결정적 방법으로 해를 구하기 어려울 때 타당한 시간 내에 최적해를 찾게 해준다. 본 연구는 유전 알고리즘을 이용하여 200 개 코드를 가진 악보 27개를 대상으로 학습자 부담을 최소화하고 단계별 학습량을 균등하게 분산함으로써 학습 효과를 최대 화 할 수 있도록 학습 경로를 최적화하였다. 학습 컨텐츠가 27개만 되어도 학습 경로의 순열 크기는 10 28을 넘지만, 본 연구 에서 구현한 도구로 평균 20분 이내에 최적해를 구할 수 있었다. 실험 결과는 유전 알고리즘이 다양한 목적의 맞춤형 교육을 위한 복잡한 학습 경로 설계에 효과적임을 보여주었다. 제안한 방법은 다른 교육 도메인에도 활용할 수 있을 것으로 기대된다.

For customized education, it is essential to search the learning path for the learner. The genetic algorithm makes it possible to find optimal solutions within a practical time when they are difficult to be obtained with deterministic approaches because of the problem’s very large search space. In this research, based on genetic algorithm, the learning paths to learn 200 chords in 27 music sheets were optimized to maximize the learning effect by balancing and minimizing learner’s burden and learning size for each step in the learning paths. Although the permutation size of the possible learning path for 27 learning contents is more than 10 28, the optimal solution could be obtained within 20 minutes in average by an implemented tool in this research. Experimental results showed that genetic algorithm can be effectively used to design complex learning path for customized education with various purposes. The proposed method is expected to be applied in other educational domains as well.

MUSIC algorithm has often used to solve the direction of arrival estimation problem to take advantage of the benefits of beamspace operations, such as reduced computation complexity, reduced sensitivity to system errors, good resolution, reduced bias in the estimation. In this paper, a variant of the beamspace MUSIC algorithm is developed, taking advantage of unitary transformations. In general, the beamspace MUSIC algorithm improves direction of arrival estimation by using reduced order characteristic polynomial rooting though Gaussian column reduction. The unitary beamspace MUSIC algorithm reduces the computational complexity of MUSIC algorithm by using real valued forward and backward, while maintaining the original precision. We use forward backward spatial smoothing technique as the preprocessor of beamspace MUSIC algorithm to recover the reduced rank of the covariance matrix due to the coherence of source signals. Through computer simulations show that the combination of spatial smoothing a beamspace MUSIC can achieve good resolution performance.

In this paper we propose an improved MUSIC (Multiple Signal Classification) algorithm applicable for direction of arrival (DOA) estimation of coherent signals in the presence of one-dimensional uniform linear array (ULA), which is based on Toeplitz matrix theory and Fourth-order-cumulants (Foc). In the signal model, a new Toeplitz construction method combining SVD (singular value decomposition) with mean calculation is explored to reconstruct the covariance matrix of array output. The DOA estimation problem can be addressed when the covariance matrix is full-rank. Foc theory is used to eliminate the Gaussian noise in the signals, after that the space of the array matrix is changed, which determines the final signal subspace and noise subspace. According to the subspace, we can adopt the conventional MUSIC to estimate the DOAs of coherent signals. Simulation results show that this algorithm provides a significant performance in comparison with other de-correlation algorithms. It has a better resolution under the condition of small angle interval. In addition, a lower root-mean-square error (RMSE) is obtained at low signal-to-noise (SNR) situation.

Subspace partition is a common method in normal MUSIC algorithm that divides the signal covariance matrix into signal subspace and noise subspace by eigenvalue decomposition. By this method, the effect of environmental noise is curbed. However, when the signal angle interval becomes small and the signal-noise ratio reduces, some certain limitations in multiple signal estimation such as loss and confusion will be presented, which means the normal method of estimation is unable to distinguish those signals we need actually. A modified MUSIC algorithm is proposed in this paper to solve the problem. A modified part in the spatial spectrum called weighting function is introduced. Some weighted operation are given to the steering vectors when the spatial spectrum is formed, making the most of subspaces and there eigenvalues. Some simulations followed are taken to discuss the performace of the modified method. Through the analysis we can see that, under the condition of a small signal angle interval and a low signal-noise ratio, the improved algorithm could achieve satisfactory result for the DOA estimation.

Acoustic Emission (AE) source localization is a useful implement to diagnose the incipient faults in rotating machine. This paper proposes an improved Near-field Multiple Signal Classification method using four-order cumulants (NFC-MUSIC) to accurately locate the faults position. In order to overcome dispersion effect and revised velocity, the feature sub narrow band is extracted by Wavelet Packet reconstitution (WPR) and Modal Plate Wave Theory (MPWT). For multi-source decorrelation and increasing localized resolution, the four-order cumulants of observed signal can be selected for localization. The experiment results indicate that the improved method can accurately locate multi-rubbing faults. It is an efficient way to assist incipient rubbing faults diagnosis.

In this paper, we studied to direction of arrival (DoA) estimation to use DoA and optimum weight algorithms in coherent interference channels. The DoA algorithm have been considerable attention in signal processing with coherent signals and a limited number of snapshots in a noise and an interference environment. This paper is a proposed method for the desired signal estimation using MUSIC algorithm and adaptive beamforming to compare classical subspace techniques. Also, the proposed method is combined the updated weight value with LCMV beamforming algorithm in adaptive antenna array system for direction of arrival estimation of desired signal. The proposed algorithm can be used with combination to MUSIC algorithm, linearly constrained minimum variance beamforming (LCMV) and the weight value method to accurately desired signal estimation. Through simulation, we compare the proposed method with classical direction of in order to desired signals estimation. We show that the propose method has achieved good resolution performance better that classical direction arrival estimation algorithm. The simulation results show the effectiveness of the proposed method.

Many people enjoy digital music (e.g., MP3 songs), usually with random play mode, or their own favorable play list that they have composed. However, such play modes do not consider and support listener preferences of feeling or mood changing with time. Usually listeners have dynamic, not static, demands on music based on their arbitrary situation or mood (e.g., when studying, exercising, being sorrowful, being happy, etc.), so an adaptive algorithm to meet the momentary demand is required. This paper proposes an adaptive algorithm to recommend favorable songs successively, and enable people to seamlessly keep listening to favorable songs, without the action of skipping disliked ones. The algorithm monitors if a listener likes or dislikes a song currently being played. Once the algorithm detects that a listener likes the song, the algorithm recommends the next song that is most similar to the current song. Otherwise, the algorithm recommends quite a different style of a song as the next one, by recognizing that the listener now has a different demand. In our experiment, the proposed algorithm showed better performance, in terms of reducing the action of frequently skipping songs, than random play mode, with statistical significance.

This study aims to develop a personalized music digital therapeutic based on MBTI personality types and apply it to depression treatment. In the data collection stage, participants' MBTI personality types and music preferences were surveyed to build a database, which was then preprocessed as input data for the KNN model. The KNN model calculates the distance between personality types using Euclidean distance and recommends music suitable for the user's MBTI type based on the nearest K neighbors' data. The developed system was tested with new participants, and the system and algorithm were improved based on user feedback. In the final validation stage, the system's effectiveness in alleviating depression was evaluated. The results showed that the MBTI personality type-based music recommendation system provides a personalized music therapy experience, positively impacting emotional stability and stress reduction. This study suggests the potential of nonpharmacological treatments and demonstrates that a personalized treatment experience can offer more effective and safer methods for treating depression.

본 연구는 공간상에서 전파를 이용하여 원하는 목표물의 도래 방향을 추정 한다. 도래방향 추정은 수신 배열 안테나들로 입사하는 신호들 중에서 원하는 목표물의 위치를 찾는 것이다. 본 연구에서는 도래방향 추정의 고 분해능 MUSIC알고리즘과 비용함수를 사용하여 목표물에 대한 도래방향을 추정하였고, 최적의 가중치를 계산하였다. 모의실험을 통하여 목표물 도래 방향 추정에서 기존 ESPRIT 알고리즘과 제안 알고리즘의 성능을 비교 분석 하였다. 목표물 도래 방향 추정에서 제안한 알고리즘이 기존의 ESPRIT 알고리즘보다 도래 방향 추정 능력이 향상되었다.

This paper estimates the direction of arrival of desired a target using propagation wave in spatial. Direction of arrival estimation is to find desired target position among received signal to receiver array antennas. In this paper, we estimated direction of arrival for target, by using cost function and high resolution MUSIC algorithm, in order to direction of arrival estimation, and calculated optimum weight vector. Through simulation, in regard to the estimation of the arrival direction of a target, the performances of the existing ESPRIT algorithm and the proposed algorithm were comparatively analyzed. In the estimation time of the arrival direction of a target object, the proposed algorithm showed an improvement of approximately as compared to the existing ESPRIT algorithm.

노인인구 증가는 치매를 비롯한 각종 노인성 질환을 앓는 환자의 급증을 의미한다. 본 연구는 인지 기능 향상이나 저하 방지를 위해 개발된 신경학적 음악치료 프로그램의 효과 및 그 프로그램 알고리 즘을 살펴보기 위한 예비 연구였다. 이를 위해 경도인지장애 상태의 노인 19명을 대상으로 그 일부 는 1회기 그리고 다른 일부는 8회기 프로그램에 참여시켰다. 프로그램 참여 효과를 측정하기 위해 참여자들에게 한국판 간이정신상태검사(K-MMSE)를 프로그램 참여 직전, 종료 후 30분 지점, 종료 후 1주일, 그리고 종료 후 2주일이 지난 시점에서 네 차례 시행하였다. 단 1회기만 참여했던 자들의 자 료를 분석한 결과, K-MMSE 점수는 프로그램 시행 직전에 비해 프로그램 종료 직후에만 유의미하게 상승했다. 이에 비해 8회기 프로그램에 참여한 집단에서는 K-MMSE 점수가 프로그램 시행 직전에 비 해 프로그램 종료 직후 그리고 종료 후 1주일까지 유의미하게 상승이 유지되었으나, 프로그램 종료 후 2주일에서는 프로그램 시행 직전 수준으로 돌아갔다. 이와 같은 결과는 경도인지장애 노인의 인 지능력 향상을 위해 개발했던 본 프로그램의 효과가 비교적 짧은 기간(즉, 1주일)만 유지되었음을 보 여주었다. 그러나 그와 같은 결과가 종속 측정치의 적합성 문제에 기인할 수 있었음을 논하였으며, 이러한 문제점 보완과 함께 프로그램 회기와 같은 알고리즘을 살피는 후속 연구를 제안했다.

The increase in the old population means a rapid increase in the number of patients suffering from various geriatric diseases, including dementia. This study was a pilot study to examine the effects of a neurological music therapy program developed to improve or prevent cognitive function decline and an algorithm of the program. For this purpose, 19 community-dwelling older adults with mild cognitive impairment were recruited, some of whom participated in a one-session program while the others in an eight-session program. To measure the effect of program participation, the Korean version of the Mini-Mental State Examination (K-MMSE) was administered to participants four times: immediately before participating in the program, 30 minutes after completion of program participation, one week after the completion, and two weeks after the completion. As a result of analyzing the data of those who participated in only one-session, it was found that K-MMSE scores significantly increased only immediately after the end of the program compared to immediately before the program was implemented. However, in the group that participated in the eight-session program, it was found that the K-MMSE score maintained a significant increase immediately after the end of the program and one week after the end of the program compared to just before the program was implemented, but returned to the level just before the program was implemented at two weeks after the end of the program. The findings showed that the effects of this program, which was developed to improve the cognitive abilities of older people with mild cognitive impairment, were maintained for only a relatively short period of one week. However, it was discussed that the findings might have been due to problems with the suitability of dependent measures, and thus follow-up studies were proposed to make up the problems and look at algorithms such as program sessions.

본 논문에서는 ATW(Automatic Tracking Window)를 사용하여 입력신호를 처리한 후에 MUSIC을 사용하여 주파수를 추정하도록 알고리즘을 수정하므로써 MUSIC알고리즘의 Threshold효과를 개선할 수 있음을 보인다[1]. ATW 전처리는 일종의 대역 여파기 효과를 가지나 일반 대역 여파기와 다른 점은 사용자가 입력신호의 중심 주파수를 알지 못해도 된다는 장점을 갖는다.

In this paper, we propose an efficient two-dimensional geo-location estimation system for mobile social security robots, where the location is obtained by direction of arrival (DOA) and time of arrival (TOA) of the radio signal. The proposed system requires only one reference signal while the conventional systems generally demand more than three reference signals. For estimating TOA and DOA information simultaneously, we employ a two-dimensional multiple signal classification (2-D MUSIC) algorithm. In addition, the performance analysis of proposed system is provided in terms of location accuracy and computational complexity by comparing it with two-dimensional Matrix Pencil (2-D MP) algorithm. The simulation results show that the proposed geo-location estimation system achieves the positioning accuracy within 3 meters in 2 kilo-meters coverage which is usable for mobile robots.

The diagnosis of motor failures using an on-line method has been the aim of many researchers and studies. Several spectral analysis techniques have been developed and are used to facilitate on-line diagnosis methods in industry. This paper discusses the first application of a motor flux spectral analysis to the identification of broken rotor bar (BRB) faults in induction motors using a multiple signal classification (MUSIC) technique as an on-line diagnosis method. The proposed method measures the leakage flux in the radial direction using a radial flux sensor which is designed as a search coil and is installed between stator slots. The MUSIC technique, which requires fewer number of data samples and has a higher detection accuracy than the traditional fast Fourier transform (FFT) method, then calculates the motor load condition and extracts any abnormal signals related to motor failures in order to identify BRB faults. Experimental results clearly demonstrate that the proposed method is a promising candidate for an on-line diagnosis method to detect motor failures.

The pitch tracking of music has been researched for several decades. Several possible improvements are available for creating a good t-distribution, using the instantaneous robust algorithm for pitch tracking framework to perfectly detect pitch. This article shows how to detect the pitch of music utilizing an improved detection method which applies a statistical method; this approach uses a pitch track, or a sequence of frequency bin numbers. This sequence is used to create an index that offers useful features for comparing similar songs. The pitch frequency spectrum is extracted using a modified instantaneous robust algorithm for pitch tracking (IRAPT) as a base combined with the statistical method. The pitch detection algorithm was implemented, and the percentage of performance matching in Thai classical music was assessed in order to test the accuracy of the algorithm. We used the longest common subsequence to compare the similarities in pitch sequence alignments in the music. The experimental results of this research show that the accuracy of retrieval of Thai classical music using the t-distribution of instantaneous robust algorithm for pitch tracking (t-IRAPT) is 99.01%, and is in the top five ranking, with the shortest query sample being five seconds long.

본 논문에서는 고해상도 도래방향 추정기법인 MUSIC(Multiple Signal Classification) 알고리즘의 설계에 대해서 연구하였다. MUSIC 알고리즘은 고유벡터와 방향벡터의 요소가 복소수이기 때문에 하드웨어 구현을 위해서는 입력상관행렬을 확장하거나 유니터리(Unitary) 개념을 적용해야 한다. 이에 따라 MUSIC 알고리즘의 방향벡터와 잡음고유벡터가 서로 직교한다는 성질을 이용하여, 소자 간격과 도래방향을 고려한 기지의 방향벡터와 신호에 의한 잡음고유벡터의 실수연산을 통해 도래방향을 구하였다. 본 논문에서는 MUSIC 알고리즘을 안테나 소자가 2개, 소자 간격이 0.5A인 경우에 대해서 하드웨어 구현이 가능하도록 Verilog HDL(Verilog Hardware Description Language)을 이용하여 설계하고 결과를 확인하였다.

In this paper, design of MUSIC algorithm, which is one of high resolution DOA (direction of arrival) estimation techniques was studied. Generally the complex-valued correlation matrix of MUSIC algorithm is transformed to unitary matrix or matrix expansion for the real hardware implementation. Using the orthogonality between the noise subspace eigenvectors and the steering vectors corresponding to signal component, we estimate DOA with the real-valued computation between steering vectors and noise subspace eigenvectors. The DOA algorithm was designed with VHDL models with considerations of 2 elements and 1 incident wave and its simulation results are derived.