學術出版;;Publications

Permanent URI for this communityhdl:11536/1

收集所有交通大學研究人員與學生個人或團體所發表之各項學術著作，涵蓋期刊論文、教師專書、學位論文等項。;;Collections of Publications of NCTU Researchers and Students Like Journal Papers and Thesis.

Browse

🏛️國立陽明交通大學機構典藏

Search Results

Now showing 1 - 10 of 10

REINFORCEMENT STRUCTURE PARAMETER LEARNING FOR NEURAL-NETWORK-BASED FUZZY-LOGIC CONTROL-SYSTEMS
(1994-02-01)
LIN, CT
;
LEE, CSG
;
電控工程研究所
;
Institute of Electrical and Control Engineering
This paper proposes a reinforcement neural-network-based fuzzy logic control system (RNN-FLCS) for solving various reinforcement learning problems. The proposed RNN-FLCS is constructed by integrating two neural-network-based fuzzy logic controllers (NN-FLC's), each of which is a connectionist model with a feedforward multilayered, network developed for the realization of a fuzzy logic controller. One NN-FLC performs as a fuzzy predictor, and the other as a fuzzy controller. Using the temporal difference prediction method, the fuzzy predictor can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the fuzzy controller. The fuzzy controller performs a stochastic exploratory algorithm to adapt itself according to the internal reinforcement signal. During the learning process, both structure learning and parameter learning are performed simultaneously in the two NN-FLC's using the fuzzy similarity measure. The proposed RNN-FLCS can construct a fuzzy logic control and decision-making system automatically and dynamically through a reward/penalty signal (i.e., a ''good'' or ''bad'' signal) or through very simple fuzzy information feedback such as ''high,'' ''too high,'' ''low,'' and ''too low.'' The proposed RNN-FLCS is best applied to the learning environment, where obtaining exact training data is expensive. The proposed RNN-FLCS also preserves the advantages of the original NN-FLC, such as the ability to find proper network structure and parameters simultaneously and dynamically and to avoid the rule-matching time of the inference engine in the traditional fuzzy logic systems. Computer simulations were conducted to illustrate the performance and applicability of the proposed RNN-FLCS.
LOW-TEMPERATURE FORMATION OF PALLADIUM SILICIDED SHALLOW P(+)N JUNCTIONS USING IMPLANT THROUGH METAL TECHNOLOGY
(1994-06-01)
LIN, CT
;
CHOU, PF
;
CHENG, HC
;
奈米中心
;
Nano Facility Center
Excellent silicided shallow pin junctions have been successfully achieved by the implantation of BF: ions into thin Pd films on a Si substrate to a dose of 5 x 10(15) cm(-2) and subsequent low-temperature (as low as 500 degrees C) furnace annealing. The formed junctions have been characterized for the respective implantation conditions. In this experiment, the implant energy plays the key role in obtaining a low leakage diode. Reverse current density of about 1 nA/cm(2) and the ideality factor of about 1.03 can be attained by the implantation of BF2+ ions at 100 keV and subsequent annealing at 600 degrees C. The junction depth is about 0.08 mu m, measured by the spread resistance method. As compared with the results of unimplanted specimens, the implantation of BF2+ ions into a thin Pd layer can stabilize the Pd silicide film and prevent it from forming islands during high-temperature annealing. High-temperature stability of palladium silicides and the leakage current mechanism are also discussed in this report.
A SILICIDATION-INDUCED PROCESS CONSIDERATION FOR FORMING SCALE-DOWN SILICIDED JUNCTION
(1994-09-01)
CHENG, HC
;
JUANG, MH
;
LIN, CT
;
HUANG, LM
;
電子工程學系及電子研究所
A process consideration for forming silicided shallow junctions, arising from silicidation process, has been discussed. The CoSi2 shallow p+n junctions formed by various schemes are characterized. The scheme that implants BF2+ ions into thin Co films on Si substrates and subsequent silicidation yields good junctions, but the problems about the dopant drive-in and knock-on of metal deeply degrade this scheme. In the regime that implants the dopant into Si and then Co deposition, however, a large perimeter leakage of 0.1 nA/cm is caused. Generation current, associated with a defect-enhanced diffusion of Co in Si during silicidation, dominates the leakage. A high-temperature pre-activation prior to Co deposition reduces the perimeter leakage to 0.038 nA/cm, but which deepens the junctions.
SILICIDE-CAUSED ANOMALOUS REVERSE CURRENT-VOLTAGE CHARACTERISTICS OF COSI2 SHALLOW P(+)N JUNCTIONS
(1995-01-01)
JUANG, MH
;
LIN, CT
;
CHENG, HC
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
Silicided shallow p (+) n junctions formed by BF2+ implantation into thin Co films on Si substrates to a low dosage (5 x 10(14) cm(-2)) and subsequent rapid thermal annealing (RTA) or conventional furnace annealing (CFA) are used to show the impact of silicides on junction characteristics. CFA results in a lower leakage than RTA at a low bias as 5 V at high temperatures attributable to longer annealing time. All the diodes made by RTA exhibit a hard-breakdown behavior. For CFA 700 degrees C annealing, however, an anomalously poor reverse I-V behavior indicative of athermal emission is found at high bias. In addition, the 800 degrees C-formed diodes and the CFA-treated 1x10(16) cm(-2) implanted samples show good reverse characteristics even at high bias. As a result, annealing conditions should be properly chosen to reduce the impact of silicides on shallow junctions.
EFFECTS OF COBALT SILICIDATION ON THE ELECTRICAL CHARACTERISTICS OF SHALLOW P(+)N JUNCTIONS FORMED BY BF2+ IMPLANTATION INTO THIN POLYCRYSTALLINE SI FILMS
(1995-03-01)
LIN, CT
;
CHAO, CH
;
JUANG, MH
;
JAN, ST
;
CHOU, PF
Excellent shallow p(+)n junctions have been formed by implanting BF2+ ions into thin polycrystalline Si films and subsequent annealing. The samples implanted at 50 keV to a dose 5 x 10(15) cm(-2) show a leakage of 1 nA/cm(2) and a junction depth of about 0.05 mu m after a 800 degrees C annealing. To reduce the series resistances of the junctions, silicidation with different col;alt thickness was used to drive the as-implanted dopants in the polysilicon films into the resultant junctions of the silicon substrates. For the low energy implantation at 50 keV at all dosages, silicidation can result in poor electrical characteristics due to the confinement of the dopants by the silicidation process. On the other hand, the electrical characteristics can be retained when a higher implantation energy of 100 keV with a dosage higher than 5 x 10(15) cm 2 was used. In addition, the samples implanted at 125 keV show poor electrical characteristics for the nonsilicided junctions but good characteristics after the silicidation. It is attributed to the enhanced defect annihilation by the formed silicide. Furthermore, silicided implanted through poly-Si junctions with excellent characteristics can be fabricated after a low temperature (600 degrees C) annealing if the samples are implanted at 100 keV with a dosage higher than 5 X 10(15) cm(-2).
A NEURAL FUZZY CONTROL-SYSTEM WITH STRUCTURE AND PARAMETER LEARNING
(1995-03-20)
LIN, CT
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
;
Institute of Electrical and Control Engineering
A general connectionist model, called neural fuzzy control network (NFCN), is proposed for the realization of a fuzzy logic control system. The proposed NFCN is a feedforward multilayered network which integrates the basic elements and functions of a traditional fuzzy logic controller into a connectionist structure which has distributed learning abilities. The NFCN can be constructed from supervised training examples by machine learning techniques, and the connectionist structure can be trained to develop fuzzy logic rules and find membership functions. Associated with the NFCN is a two-phase hybrid learning algorithm which utilizes unsupervised learning schemes for structure learning and the backpropagation learning scheme for parameter learning. By combining both unsupervised and supervised learning schemes, the learning speed converges much faster than the original backpropagation algorithm. The two-phase hybrid learning algorithm requires exact supervised training data for learning. In some real-time applications, exact training data may be expensive or even impossible to obtain. To solve this problem, a reinforcement neural fuzzy control network (RNFCN) is further proposed. The RNFCN is constructed by integrating two NFCNs, one functioning as a fuzzy predictor and the other as a fuzzy controller. By combining a proposed on-line supervised structure-parameter learning technique, the temporal difference prediction method, and the stochastic exploratory algorithm, a reinforcement learning algorithm is proposed, which can construct a RNFCN automatically and dynamically through a reward-penalty signal (i.e., ''good' or ''bad'' signal), Two examples are presented to illustrate the performance and applicability of the proposed models and learning algorithms.
A MULTIVALUED BOLTZMANN MACHINE
(1995-04-01)
LIN, CT
;
LEE, CSG
;
電控工程研究所
;
Institute of Electrical and Control Engineering
The idea of Hopfield network is based on the Ising spin glass model in which each spin has only two possible states: up and down. By introducing stochastic factors into this network and performing a simulated annealing process on it, it becomes a Boltzmann machine which can escape from local minimum states to achieve the global minimum. This paper generalizes the above ideas to multi-value case based on the XY spin glass-model in which each spin can be in any direction in a plane, Simply using the gradient descent method and the analog Hopfield network; two different analog connectionist structures and their corresponding evolving rules are first designed to transform the XY spin glass model to distributed computational models. These two analog computational models are single-layered connectionist structures and multi-layered Hopfield analog networks. The latter network eases the node (neuron) computational requirement of the former at the expense of more neurons and connections. With the proposed evolving rules, the proposed models evolve according to a predefined Hamiltonian (energy function) which will decrease until it reaches a (perhaps local) minimum. Since these two structures can easily get stuck in local minima, a multi-valued Boltzmann machine is proposed which adopts the discrete planar spin glass model for the local minimum problem. Each neuron in the multi-valued Boltzmann machine can only take n discrete directions (states). The stochastic simulated annealing method is introduced to the evolving rules of the multi-valued Boltzmann machine to solve the local minimum problem. The multi-valued Boltzmann machine can be applied to the mobile robot navigation problem;problem by defining proper artificial magnetic field on the traverse terrain. This new artificial magnetic field approach for the mobile robot navigation problem has shown to have several advantages over existing graph search and potential field techniques.
FUZZY ADAPTIVE LEARNING CONTROL NETWORK WITH ONLINE NEURAL LEARNING
(1995-04-14)
LIN, CT
;
LIN, CJ
;
LEE, CSG
;
電控工程研究所
;
Institute of Electrical and Control Engineering
This paper addresses the structure and the associated on-line learning algorithms of a feedforward multilayered connectionist network for realizing the basic elements and functions of a traditional fuzzy logic controller. The proposed Fuzzy Adaptive Learning COntrol Network (FALCON) can be contrasted with the traditional fuzzy logic control systems in their network structure and learning ability. The connectionist structure of the proposed FALCON can be constructed from training examples by neural learning techniques to find proper fuzzy partitions, membership functions, and fuzzy logic rules. Two complementary on-line structure/parameter learning algorithms, FALCON-FSM and FALCON-ART, are proposed for constructing the FALCON dynamically. The FALCON-FSM combines the backpropagation learning scheme for parameter learning and a fuzzy similarity measure for structure learning. The FALCON-FSM can find proper fuzzy logic rules, membership functions, and the size of output partitions simultaneously. In the FALCON-FSM algorithm, the input and output spaces are partitioned into ''grids''. The grid-typed space partitioning certainly makes both the fuzzy logic controller software emulation and fuzzy chip implementation convenient. However, as the number of input/output variables increases, the number of partitioned grids will grow combinatorially. To avoid the problem of combinatorial growth of partitioned grids in some complex systems, the FALCON-ART algorithm is developed, which can partition the input and output spaces in a more flexible way based on the distribution of the training data. The FALCON-ART combines the backpropagation learning scheme for parameter learning and a fuzzy ART algorithm for structure learning. The FALCON-ART can on-line partition the input and output spaces, tune membership functions and find proper fuzzy logic rules dynamically. Computer simulations were conducted to illustrate the performance and applicability of both FALCON-FSM and FALCON-ART learning algorithms.
A NEURAL FUZZY SYSTEM WITH LINGUISTIC TEACHING SIGNALS
(1995-05-01)
LIN, CT
;
LU, YC
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
A neural fuzzy system learning with linguistic teaching signals is proposed in this paper. This system is able to process and learn numerical information as well as linguistic information. It can be used either as an adaptive fuzzy expert system or as an adaptive fuzzy controller. At first, we propose a five-layered neural network for the connectionist realization of a fuzzy inference system. The connectionist structure can house fuzzy logic rules and membership functions for fuzzy inference. We use alpha-level sets of fuzzy numbers to represent linguistic information, The inputs, outputs, and weights of the proposed network can be fuzzy numbers of any shape. Furthermore, they can be hybrid of fuzzy numbers and numerical numbers through the use of fuzzy singletons. Based on interval arithmetics, two kinds of learning schemes are developed for the proposed system: fuzzy supervised learning and fuzzy reinforcement learning. They extend the normal supervised and reinforcement learning techniques to the learning problems where only linguistic teaching signals are available. The fuzzy supervised learning scheme can train the proposed system with desired fuzzy input-output pairs which are fuzzy numbers instead of the normal numerical values, With fuzzy supervised learning, the proposed system can be used for rule base concentration to reduce the number of rules in a fuzzy rule base. In the fuzzy reinforcement learning problem that we consider, the reinforcement signal from the environment is linguistic information (fuzzy critic signal) such as ''good,'' very good,'' or ''bad,'' instead of the normal numerical critic values such as ''0'' (success) or ''-1'' (failure). With the fuzzy critic signal from the environment, the proposed system can learn proper fuzzy control rules and membership functions. We discuss two kinds of fuzzy reinforcement learning problems: single-step prediction problems and multistep prediction problems. Simulation results are presented to illustrate the performance and applicability of the proposed system.
LOW-TEMPERATURE FORMATION OF SHALLOW P(+)N JUNCTIONS BY BF2+ IMPLANTATION INTO THIN PD2SI FILMS ON SI SUBSTRATES
(1995-05-01)
LIN, CT
;
MA, KP
;
CHOU, PF
;
CHENG, HC
;
奈米中心
Excellent silicided shallow p(+)n junctions have been successfully achieved by the implantation of BF2+ ions into thin Pd2Si films on Si substrates to a dose of 5 x 10(15) cm(-2) and subsequent low temperature (even at 550 degrees C) furnace annealing. The formed junctions have been characterized for respective implantation conditions. In this experiment, the implant energy is the key role in obtaining a low leakage diode. Reverse current density of about 3 nA/cm(2) and an ideality factor of about 1.05 can be attained by the implantation of BF2+ ions at 80 keV and subsequent annealing at 550 degrees C. The junction depth is about 0.09 mu m, measured by the spread resistance method. As compared with the results of unimplanted specimens, the implantation of BF2+ ions into a thin Pd2Si layer can stabilize the silicide film and prevent it from forming islands during high temperature annealing.

學術出版;;Publications

Browse

Filters

Settings

Sort By

Results per page

🏛️國立陽明交通大學機構典藏

Search Results