• 대한임베디드공학회논문지
• /
• 제15권1호
• /
• pp.1-8
• /
• 2020

A derivation method for a quantized gradient for machine learning on an embedded system is proposed, in this paper. The proposed differentiation method induces the quantized gradient vector to an objective function and provides that the validation of the directional derivation. Moreover, mathematical analysis shows that the sequence yielded by the learning equation based on the proposed quantization converges to the optimal point of the quantized objective function when the quantized parameter is sufficiently large. The simulation result shows that the optimization solver based on the proposed quantized method represents sufficient performance in comparison to the conventional method based on the floating-point system.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1988년도 한국자동제어학술회의논문집(국제학술편); 한국전력공사연수원, 서울; 21-22 Oct. 1988
• /
• pp.772-776
• /
• 1988

In quantized control systems, the control values can take only given discrete (e.g. integer) values. In case of dealing with the control problem on the discrete-time, final-stage fixed, quantized control systems with multidimensional performance functions, the first thing, new definition on noninferior solutions in these systems is necessary because of their discreteness in state variables, and the efficient search for those solutions at final-stage is unavoidable for seeking their discrete-time optimal controls to these systems. In this paper, to the quantized control problem given by the formulation of Halkin-typed linear control systems with two performance functions, a new definition on noninferior solutions of this system control problem and a heuristic effective search on these noninferior solutions are stated. By use of these concepts, two definitions on noninferior solutions and the algorithm consisted of 8 steps and attained by geometric approaches are given. And a numerical example using the present algorithm is shown.

• Journal of Computing Science and Engineering
• /
• 제4권3호
• /
• pp.207-224
• /
• 2010

This paper presents a compression scheme for mesh geometry, which is suitable for mobile graphics. The main focus is to enable real-time decoding of compressed vertex positions while providing reasonable compression ratios. Our scheme is based on local quantization of vertex positions with mesh partitioning. To prevent visual seams along the partitioning boundaries, we constrain the locally quantized cells of all mesh partitions to have the same size and aligned local axes. We propose a mesh partitioning algorithm to minimize the size of locally quantized cells, which relates to the distortion of a restored mesh. Vertex coordinates are stored in main memory and transmitted to graphics hardware for rendering in the quantized form, saving memory space and system bus bandwidth. Decoding operation is combined with model geometry transformation, and the only overhead to restore vertex positions is one matrix multiplication for each mesh partition. In our experiments, a 32-bit floating point vertex coordinate is quantized into an 8-bit integer, which is the smallest data size supported in a mobile graphics library. With this setting, the distortions of the restored meshes are comparable to 11-bit global quantization of vertex coordinates. We also apply the proposed approach to compression of vertex attributes, such as vertex normals and texture coordinates, and show that gains similar to vertex geometry can be obtained through local quantization with mesh partitioning.

• 방송공학회논문지
• /
• 제1권1호
• /
• pp.29-34
• /
• 1996

차등적 오류 방지는 전송되는 정보에 따라 채널에서 오류가 발생했을 때 미치는 영향이 다른 것을 이용하여 채널 오류에 민감한 정보를 다른 정보보다 강력하게 채널 오류로부터 보호하는 방법이다. 그러나 채널에서 오류가 없다는 가정하에 LBG 알고리듬으로 설계한 벡터 양자화 신호에 차등적 오류 방지를 그대로 적용할 수 없다. 따라서 본 논문에서는 이러한 벡터 양자화 신호의 전송에서 차등적 오류 방지를 적용할 수 있도록 부호 벡터에 체계적으로 이진 인덱스를 할당하는 기법을 제안하였다. 또한 제안한 기법을 벡터 양자화 한 1차 Gauss-Marcov 신호의 전송에 중요한 정보의 비율을 50%로 하여 적용하였다.

• Journal of Communications and Networks
• /
• 제15권2호
• /
• pp.164-172
• /
• 2013

We consider the problem of secure and robust clustering for quantized target tracking in wireless sensor networks (WSN) where the observed system is assumed to evolve according to a probabilistic state space model. We propose a new method for jointly activating the best group of candidate sensors that participate in data aggregation, detecting the malicious sensors and estimating the target position. Firstly, we select the appropriate group in order to balance the energy dissipation and to provide the required data of the target in the WSN. This selection is also based on the transmission power between a sensor node and a cluster head. Secondly, we detect the malicious sensor nodes based on the information relevance of their measurements. Then, we estimate the target position using quantized variational filtering (QVF) algorithm. The selection of the candidate sensors group is based on multi-criteria function, which is computed by using the predicted target position provided by the QVF algorithm, while the malicious sensor nodes detection is based on Kullback-Leibler distance between the current target position distribution and the predicted sensor observation. The performance of the proposed method is validated by simulation results in target tracking for WSN.

• Journal of information and communication convergence engineering
• /
• 제15권1호
• /
• pp.7-13
• /
• 2017

We consider a distributed estimation where many nodes remotely placed at known locations collect the measurements of the parameter of interest, quantize these measurements, and transmit the quantized data to a fusion node; this fusion node performs the parameter estimation. Noting that quantizers at nodes should operate in a non-regular framework where multiple codewords or quantization partitions can be mapped from a single measurement to improve the system performance, we propose a low-weight estimation algorithm that finds the most feasible combination of codewords. This combination is found by computing the weighted sum of the possible combinations whose weights are obtained by counting their occurrence in a learning process. Otherwise, tremendous complexity will be inevitable due to multiple codewords or partitions interpreted from non-regular quantized data. We conduct extensive experiments to demonstrate that the proposed algorithm provides a statistically significant performance gain with low complexity as compared to typical estimation techniques.

• 전기전자학회논문지
• /
• 제23권2호
• /
• pp.710-715
• /
• 2019

We consider parameter estimation in distributed systems in which measurements at local nodes are quantized in a non-regular manner, where multiple codewords are mapped into a single local measurement. For the system with non-regular quantization, to ensure a perfect independent encoding at local nodes, a local measurement can be encoded into a set of a great number of codewords which are transmitted to a fusion node where estimation is conducted with enormous computational cost due to the large cardinality of the sets. In this paper, we propose an efficient estimation technique that can handle the non-regular quantized data by efficiently finding the feasible combination of codewords without searching all of the possible combinations. We conduct experiments to show that the proposed estimation performs well with respect to previous novel techniques with a reasonable complexity.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 제38회 하계학술대회
• /
• pp.7-8
• /
• 2007

This paper presents an vulnerability index for hidden failure of protective relays in transmission system. The bad influence can be quantized by an vulnerability index. When there is mis-operation, no-operation or mis-setting, power flow can be quantized by an index. According to the index, relays can be resetting. So the wide area blackout can be prevented.

• 센서학회지
• /
• 제24권6호
• /
• pp.368-372
• /
• 2015

In this paper, we proposed a post-processing technique for improving classification performance of electronic nose (E-Nose) system which may be occurred drift signals from sensor array. An adaptive radial basis function network using stochastic gradient (SG) and singular value decomposition (SVD) is applied to process signals from sensor array. Due to drift from sensor's aging and poisoning problems, the final classification results may be showed bias and fluctuations. The predicted classification results with drift are quantized to determine which identification level each class is on. To mitigate sharp fluctuations moving-averaging (MA) technique is applied to quantized identification results. Finally, quantization and some edge correction process are used to decide levels of the fluctuation-smoothed identification results. The proposed technique has been indicated that E-Nose system was shown correct odor identification results even if drift occurred in sensor array. It has been confirmed throughout the experimental works. The enhancements have produced a very robust odor identification capability which can compensate for decision errors induced from drift effects with sensor array in electronic nose system.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 추계학술대회논문집
• /
• pp.279-284
• /
• 2005

Knowledge based fault diagnosis has a limitation in determining the cause and scheme for the fault, because it detects faults from signal pattern only Therefore, model-based fault diagnosis is requested to determine the fault by analyzing output of the equipment from its dynamic model. This research shows a method how to devise the automaton of system as a model for normal and faulty condition through the reduction of handling data by quantization of vibration signals and the example which is concerning to the bearing of ATM. The developed model based fault diagnosis was applied to detect the faulty bearing of ATM, which results.