• Journal of the Korea Society of Computer and Information
• /
• v.25 no.8
• /
• pp.55-63
• /
• 2020

Recently, people have come to enjoy using movies and music applications, such as YouTube, Netflix, and Watcha Play, on their smart devices while riding public transportation. However, they sometimes miss their destination stations because of these activities. Therefore, in this paper, we propose a disembarking notification system for public city buses using smart devices and high frequencies generated via the buses' speakers. The high frequencies are generated by the public buses' station information broadcasters. Smart devices then analyze the high-frequency signals with their inner microphones, and the proposed application displays a disembarking notification to the user when the user's destination or stop-over station's signal is the same. To evaluate this system's performance, we tested 1,000 real-time disembarking notifications, and the test results showed 98.9% accuracy. Moreover, we compared these results to those using only OpenAPI, and our proposed system featured far better outcomes. Thus, this proposed notification system can prove a useful technology for many people who often use public city buses, as it can notify specific users of their destination stations. Furthermore, this system will become innovative technology for global public transportation by informing users of their desired stations using speakers.

• Journal of Korea Multimedia Society
• /
• v.5 no.4
• /
• pp.393-398
• /
• 2002

The conventional bit allocation scheme for MPEG-1 audio encoding searches the subband with minimum MNR(mask-to-noise ratio) repetitively until its operation is completed, which occupies most of its total computational complexity. In this paper, as a computationally efficient approximation of it, we propose a new bit allocation scheme with a simple subband search and compare it with the existing schemes[1][2] in terms of the computational complexity and sound quality. For the performance comparison, we used the pop music signal contained in SQAM(sound quality assess material) CD from EBU. Simulation results show that the computational complexity of the proposed method is about 42% of that of the existing one in [1] and the sound quality difference in terms of MNR between the two schemes is within the 0.2 ㏈, for the case of using the layer II at the bit rate of 128 kbps.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.30 no.2
• /
• pp.169-172
• /
• 2019

Frequency-modulated continuous wave(FMCW) radars with array antennas are widely used because of their light weight and relatively high resolution. A usual approach for the joint range and angle estimation of a target using an array FMCW radar is to create a range-angle matrix with the deramped received signal, and subsequently apply two-dimensional(2D) frequency estimation methods such as 2D fast Fourier transform on the range-angle matrix. However, such frequency estimation approaches cause bias errors since the frequencies in the range-angle matrix are not independent. Therefore, we propose a new maximum likelihood-based algorithm for joint range and angle estimation of targets using array FMCW radar, and demonstrate that the proposed algorithm achieves the Cram?r-Rao bounds, both for range as well as angle estimation.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.3
• /
• pp.234-240
• /
• 2021

In this paper, the mask-based speech enhancement is improved for effective speech recognition in noise environments. In the mask-based speech enhancement, enhanced spectrum is obtained by multiplying the noisy speech spectrum by the mask. The VoiceFilter (VF) model is used as the mask estimation, and the Spectrogram Inpainting (SI) technique is used to remove residual noise of enhanced spectrum. In this paper, we propose a combined loss to further improve speech enhancement. In order to effectively remove the residual noise in the speech, the positive part of the Triplet loss is used with the component loss. For the experiment TIMIT database is re-constructed using NOISEX92 noise and background music samples with various Signal to Noise Ratio (SNR) conditions. Source to Distortion Ratio (SDR), Perceptual Evaluation of Speech Quality (PESQ), and Short-Time Objective Intelligibility (STOI) are used as the metrics of performance evaluation. When the VF was trained with the mean squared error and the SI model was trained with the combined loss, SDR, PESQ, and STOI were improved by 0.5, 0.06, and 0.002 respectively compared to the system trained only with the mean squared error.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.5
• /
• pp.177-187
• /
• 2014

This paper proposes a fast subspace tracking methods, which is called GVFF FAPI, based on FAPI (Fast Approximated Power Iteration) method and GVFF RLS (Gradient-based Variable Forgetting Factor Recursive Lease Squares). Since the conventional FAPI uses a constant forgetting factor for estimating covariance matrix of source signals, it has difficulty in applying to non-stationary environments such as continuously changing DOAs of source signals. To overcome the drawback of conventioanl FAPI method, the GVFF FAPI uses the gradient-based variable forgetting factor derived from an improved means square error (MSE) analysis of RLS. In order to achieve the decreased subspace error in non-stationary environments, the GVFF-FAPI algorithm used an improved forgetting factor updating equation that can produce a fast decreasing forgetting factor when the gradient is positive and a slowly increasing forgetting factor when the gradient is negative. Our numerical simulations show that GVFF-FAPI algorithm offers lower subspace error and RMSE (Root Mean Square Error) of tracked DOAs of source signals than conventional FAPI based MUSIC (MUltiple SIgnal Classification).