• Journal of the Korean Geographical Society
• /
• v.44 no.3
• /
• pp.301-322
• /
• 2009

This paper is an attempt to explore semiotic representations of subway space as the urban machine space of local mobility in terms of space, time and place. For this, the second section of the paper reviews the contours of the urban space of mobility in terms of 'machine space', 'non-place' and 'cognitive map'. The third section interprets the sings of 'spatial' and 'temporal' representations of subway space in Daegu, and suggests the implications of the semiotic representations. It is uncovered that various sign-scapes which coexist in the subway space in coordinated or contradictory ways product the space into multiple and complex techno-social spaces. That is, the spatio-temporal representations of the subway space form the space of 'non-place' on the one hand and the space of 'place' on the other hand, and involve the spatialization of 'memory' on the one hand and the spatialization of 'forgetting' on the other hand. Thus, the subway space should be regarded to be not only the space of 'mobility' which people move in and through, but also the space of 'identity' which has effects on the ways for them to see the machine space and its urban space.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.1A
• /
• pp.58-65
• /
• 2002

This paper deals with a delay problem in the endpoint detection of double-talk detection algorithm using correlation coefficient in the acoustic echo canceller. In case that past power is much bigger than current power like at the end of double-talking period, the power, estimated using forgetting factor, decreases slowly to cause a delay problem in the endpoint detection. In this paper, two methods are proposed to solve this problem. One is that the current power is periodically replaced by a new average power and the other is that the past power in recursive equation is periodically removed or replaced by other values. The simulation results show that proposed methods outperform conventional method in the endpoint of double-talking periods without increasing the computational burden much more.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.4
• /
• pp.320-325
• /
• 2010

This paper is about the method and procedure to register the candidate sequentially in the face recognition system using the PCA(Principal Components Analysis). We use the method to update the principal components sequentially with the SKL algorithm which is improved R-SVD algorithm. This algorithm enable us to solve the re-training problem of the increase the candidates number sequentially in the face recognition using the PCA. Also this algorithm can use in robust tracking system with the bright change based to the principal components. This paper proposes the procedure in the face recognition system which sequentially updates the principal components using the SKL algorithm. Then we compared the face recognition performance with the batch procedure for calculating the principal components using the standard KL algorithm and confirms the effects of the forgetting factor in the SKL algorithm experimentally.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.1
• /
• pp.29-36
• /
• 2019

Unmanned Autonomous Vehicle's driving road in the national defense includes not only paved roads, but also unpaved roads which have rough and unexpected changes. This Unmanned Autonomous Vehicles monitor and recon rugged or remote areas, and defend own position, they frequently encounter environments roads of various and unpredictable. Thus, they need additional learning to drive in this environment, we propose a Additional Learning Nearest Neighbor (ALNN) which is modified from Approximate Nearest Neighbor to allow for quick learning while avoiding the 'Forgetting' problem. In addition, since the Execution speed of the ALNN algorithm decreases as the learning data accumulates, we also propose a solution to this problem using GPU parallel processing based on OpenGL Shader Language. The ALNN based on GPU algorithm can be used in the field of national defense and other similar fields, which require frequent and quick application of additional learning in real-time without affecting the existing learning data.

• Structural Engineering and Mechanics
• /
• v.37 no.4
• /
• pp.351-365
• /
• 2011

This study proposes an inverse estimation method for the input forces of a fixed beam structural system. The estimator includes the fuzzy Kalman Filter (FKF) technology and the fuzzy weighted recursive least square method (FWRLSM). In the estimation method, the effective estimator are accelerated and weighted by the fuzzy accelerating and weighting factors proposed based on the fuzzy logic inference system. By directly synthesizing the robust filter technology with the estimator, this study presents an efficient robust forgetting zone, which is capable of providing a reasonable trade-off between the tracking capability and the flexibility against noises. The period input of the fixed beam structure system can be effectively estimated by using this method to promote the reliability of the dynamic performance analysis. The simulation results are compared by alternating between the constant and adaptive and fuzzy weighting factors. The results demonstrate that the application of the presented method to the fixed beam structure system is successful.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.355-358
• /
• 2000

기존의 Rayleigh fading 환경에서 칼만 필터를 사용한 다중 사용자 검출에서는 비동기적으로 들어온 다른 사용자 신호의 상태가 바뀜에 따라서 수렴하던 신호에 영향을 미친다. 이러한 특성은 결국 전체 시스템에 영향을 미쳐서 검출 성능을 떨어뜨린다. 본 논문에서는 에러가 발생하였을 때의 경우를 고려하기 위해 VFF(Variable forgetting factor)를 도입하였고 이를 이용해 추정된 신호의 가중 합을 기반으로 시스템을 구성한 결과 성능이 향상되었음을 보이고자 한다.

• Proceedings of the IEEK Conference
• /
• 2003.07d
• /
• pp.1283-1286
• /
• 2003

Human can update the memory with new information not forgetting acquired information in the memory. ART(Adaptive Resonance Theory) does not need to change all information. The methodology of ART is followed. The ART updates the memory with the new information that is unknown if it is similar with the memorized information. On the other hand, if it is unknown information the ART adds it to the memory not updating the memory with the new one. This paper shows that ART is able to classify sensory information of a certain object. When ART receives new information of the object as an input, it searches for the nearest thing among the acquired information in the memory. If it is revealed that new information of the object has similarity with the acquired object, the model is updated to reflect new information to the memory. When new object does not have similarity with the acquired object, the model register the object into new memory

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.4
• /
• pp.11-16
• /
• 1996

This paper proposes a deadzone estimator for robust system identification. In order to cope with the drift phenomena occurred in where system inputs are not sufficiently excited in adaptive control, we introduce a novel and flexible bound condition against a fixed constant. It is derived from a forgetting factor and a rational value of the traces of the covariance matrices between step k and k-1. The key feature of this is that it does not require a priori for the bound. Also, the calculation of it is more simple than the one of literatures. The simulation results are examined for showing the practical performance of this algorithm.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.5
• /
• pp.102-108
• /
• 1996

Adaptive resonance theory (ART2) characterized by its built-in mechanism of handling the stability-plasticity switching and by the adaptive learning without forgetting informations learned in the past, is based on an unsupervised template matching. We propose an improved tow-stage learning algorithm for aRT2: the original unsupervised learning followed by a new supervised learning. Each of the output nodes, after the unsupervised learning, is labeled according to the category informations to reinforce the template pattern associated with the target output node belonging to the same category some dominant classes from exhausting a finite number of template patterns in ART2 inefficiently. Experimental results on a set of 2545 FLIR images show that the ART2 trained by the two-stage learning algorithm yields better accuracy than the original ART2, regardless of th esize of the network and the methods of evaluating the accuracy. This improvement shows the effectiveness of the two-stage learning process.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.318-321
• /
• 2003

In case, sensor system performs where it is difficult to access physically and it is in the poor environment, it is limited to communicate by using wire and installing power module in sensor system. In this paper, it suggests how information is obtained from telemetry sensor by means of inductive coupling without battery. Comparing with the telemetry sensor system of inductive coupling by the power supply, this system estimates the capacitance of sensor with high precision in using RLSE, not the process of modulation and demodulation. In order to activate this system, inductive model is used and in case of time variant parameter, telemetry sensor system which has got high rate in accuracy is implemented by using the forgetting factor.