• Journal of Institute of Control, Robotics and Systems
• /
• v.9 no.11
• /
• pp.917-924
• /
• 2003

Recurrent neural networks have been successfully applied to communications channel equalization. Major disadvantages of gradient-based learning algorithms commonly employed to train recurrent neural networks are slow convergence rates and long training sequences required for satisfactory performance. In a high-speed communications system, fast convergence speed and short training symbols are essential. We propose decision feedback equalizers using a recurrent neural network trained with Kalman filtering algorithms. The main features of the proposed recurrent neural equalizers, utilizing extended Kalman filter (EKF) and unscented Kalman filter (UKF), are fast convergence rates and good performance using relatively short training symbols. Experimental results for two time-varying channels are presented to evaluate the performance of the proposed approaches over a conventional recurrent neural equalizer.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.06a
• /
• pp.81-83
• /
• 2014

This paper provides the performance comparison of selected post filters for depth map in 3D video system. In the performance comparison, the dilation filter which is currently adopted in 3D-ATM reference S/W, the bilateral filter, and the depth-oriented depth boundary reconstruction filter. In the paper, we first introduce those filters in details, and show the experimental results as post filter in 3D video system.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.8
• /
• pp.711-715
• /
• 2001

In this paper, an effective filter structure for filtering of target glint in tracking radar systems is used to improve the performance of CLOS(Command to Line-Of-Sight) guidance. The filter decouples range and angel channels to that it has a sound mathematical basis as well as computation efficiency as applied to the IMM algorithm. The effective filter structure in conjunction with CLOS guidance is tested by a series of simulation runs and it is shown to have superior performance compared with the other filter structures.

• MALSORI
• /
• no.53
• /
• pp.93-102
• /
• 2005

The performance of a speech recognition system is generally degraded in telephone environment because of distortions caused by background noise and various channel characteristics. In this paper, data-driven temporal filters are investigated to improve the performance of a specific recognition task such as telephone speech. Three different temporal filtering methods are presented with recognition results for Korean connected-digit telephone speech. Filter coefficients are derived from the cepstral domain feature vectors using the principal component analysis. According to experimental results, the proposed temporal filtering method has shown slightly better performance than the previous ones.

• Journal of electromagnetic engineering and science
• /
• v.11 no.2
• /
• pp.91-96
• /
• 2011

In this paper, we propose a microstrip bandpass filter using stepped-impedance coupled-line hairpin resonators. The stepped-impedance coupled-line hairpin resonator has extended harmonic suppression in comparison with a conventional hairpin resonator due to transmission zero and the movement of harmonic frequencies resulting from the stepped-impedance characteristic. A high-pass type impedance/admittance inverter is employed in order to improve the lower frequency skirt characteristics of the passband. A 4-pole bandpass filter is designed and fabricated at 1.8 GHz. The measured results show the excellent attenuation performance at the stopband which is greater than 30 dB up to 10 GHz.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.9
• /
• pp.2348-2360
• /
• 2023

As science and technology evolve, object detection of moving objects has been widely used in the context of machine learning and artificial intelligence. Traditional moving object detection algorithms, however, are characterized by relatively poor real-time performance and low accuracy in detecting moving objects. To tackle this issue, this manuscript proposes a modified Kalman filter algorithm, which aims to expand the equations of the system with the Taylor series first, ignoring the higher order terms of the second order and above, when the nonlinear system is close to the linear form, then it uses standard Kalman filter algorithms to measure the situation of the system. which can not only detect moving objects accurately but also has better real-time performance and can be employed to predict the trajectory of moving objects. Meanwhile, the accuracy and real-time performance of the algorithm were experimentally verified.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.5
• /
• pp.367-377
• /
• 2019

As an alternative navigation system for the non-GNSS (Global Navigation Satellite System) environment, a new type of DBRN (DataBase Referenced Navigation) which applies both gravity gradient and terrain, and combines filter-based algorithm with profile matching was suggested. To improve the stability of the performance compared to the previous study, both centralized and decentralized EKF (Extended Kalman Filter) were constructed based on gravity gradient and terrain data, and one of filters was selected in a timely manner. Then, the final position of a moving vehicle was determined by combining a position from the filter with the one from a profile matching. In the simulation test, it was found that the overall performance was improved to the 19.957m by combining centralized and decentralized EKF compared to the centralized EKF that of 20.779m. Especially, the divergence of centralized EKF in two trajectories located in the plain area disappeared. In addition, the average horizontal error decreased to the 16.704m by re-determining the final position using both filter-based and profile matching solutions. Of course, not all trajectories generated improved performance but there is not a large difference in terms of their horizontal errors. Among nine trajectories, eights show smaller than 20m and only one has 21.654m error. Thus, it would be concluded that the endemic problem of performance inconsistency in the single geophysical DB or algorithm-based DBRN was resolved because the combination of geophysical data and algorithms determined the position with a consistent level of error.

• Journal of Satellite, Information and Communications
• /
• v.12 no.3
• /
• pp.63-68
• /
• 2017

SAR(Synthetic Aperture Radar) uses electromagnetic signals to acquire ground information and has been used for wide coverage reconnaissance missions regardless of weather conditions. However SAR is known to be vulnerable to interference signals by other communication devices or radar instruments and may suffer from undesirable performance degradations and image quality. In this paper, a modified Eigen-subspace based filter is proposed that can be easily applied to SAR images affected by interference signals. The method of constructing Eigen-subspace based filter is briefly described and various simulations are performed to show the performance of the interference mitigation process. The suppression filter is applied to a ALOS PALSAR raw data affected by interfering signals in order to verify its superiority over the Notch filter.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.1
• /
• pp.176-183
• /
• 2010

This paper presents some practical design issues that should be carefully considered when a notch filter is included in a linear feedback controller. A notch filter is generally used to compensate the effects of resonant modes that may result in poor performance. It is very common that the practical engineers prefer to add such a notch filter after having previously designed a feedback controller without the filter. It is known that the resulting performance by this approach is not seriously different from when a feedback controller is designed for a plant previously compensated by a notch filter. However, we will point out that there are some cases where both approaches have different performances. In order to show this, a low-order controller design using the partial model matching method has been applied to a linear time invariant (L Tn model. The results suggest that there is a tendency to achieve much better time responses in terms of reducing the overshoot and shortening the settling time, and in the frequency domain characteristics such as the sensitivity function and the stability margins when the design of a feedback controller after including a notch filter is carried out.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.3
• /
• pp.43-49
• /
• 2014

This paper presents a theoretical study of transmission performance for an optical duobinary transmitter employed a Mach-Zehnder modulator and a electrical low pass filter at 40Gbps optical communication links. It depends on the bandwidth of the low pass filter in the transmitter, the optical filter and the filter in the receiver. Also, each filter affected to the various parts of the optical power spectrum. By optimizing the bandwidth of each filter, we could control the side robes and the ripples and improve the dispersion tolerance of the transmission system.