A new approach is presented to fine tune a fuzzy controller based on very limited information on the nonlinear plant to be controlled. Without any off-line pre-training, the algorithm achieves very high control perfor...
详细信息
A new approach is presented to fine tune a fuzzy controller based on very limited information on the nonlinear plant to be controlled. Without any off-line pre-training, the algorithm achieves very high control performance through a two-stage algorithm. In the first stage, coarse tuning of the fuzzy rules (both rule consequents and membership functions of the premises) is accomplished in real time using a small amount of information about the plant and an overall analysis of the main operating regions. In stage two, fine tuning of the fuzzy rules is achieved based on the controller output error using a gradient-based method.
We measure the experimental error that arises from the use of non-validated simulators in computerarchitecture research, with the goal of increasing the rigor of simulation-based studies. We describe the methodology ...
详细信息
ISBN:
(纸本)0769511627
We measure the experimental error that arises from the use of non-validated simulators in computerarchitecture research, with the goal of increasing the rigor of simulation-based studies. We describe the methodology that we used to validate a microprocessor simulator against a Compaq DS-10L workstation, which contains an Alpha 21264 processor. Our evaluation suite consists of a set of 21 microbenchmarks that stress different aspects of the 21264 microarchitecture. Using the microbenchmark suite as the set of workloads, we describe how we reduced our simulator error to an arithmetic mean of 2%, and include details about the specific aspects of the pipeline that required extra care to reduce the error. We show how these low-level optimizations reduce average error from 40% to less than 20% on macrobenchmarks drawn from the SPEC2000 suite. Finally, we examine the degree to which performance optimizations are stable across different simulators, showing that researchers would draw different conclusions, in some cases, if using validated simulators.
In this paper we propose a genetic algorithm (GA) that is capable of simultaneously optimizing the structure of the system and tuning the parameters that define the fuzzy system. For this purpose, we use the concept o...
详细信息
In this paper we propose a genetic algorithm (GA) that is capable of simultaneously optimizing the structure of the system and tuning the parameters that define the fuzzy system. For this purpose, we use the concept of multiple-deme GAs, in which several populations with different structures (number of input variables) evolve and compete with each other. In each of these populations, the element also has different numbers of membership functions in the input spaces and different numbers of rules. Instead of the normal coding system used to represent a fuzzy system, in which all the parameters are represented in vector form, we performed coding by means of multidimensional matrices, in which the elements are real-valued numbers, rather than the traditional binary or gray coding.
This paper proposes a framework for constructing and training a radial basis function (RBF) neural network, which is an example of fuzzy system. For this purpose, a sequential learning algorithm is presented to adapt ...
详细信息
This paper proposes a framework for constructing and training a radial basis function (RBF) neural network, which is an example of fuzzy system. For this purpose, a sequential learning algorithm is presented to adapt the structure of the network, in which it is possible to create a new hidden unit (rule) and also to detect and remove inactive units. The structure of the gaussian functions (membership functions) is modified using a pseudo-Gaussian function (PG) in which two sealing parameters /spl sigma/ are introduced, which eliminates the symmetry restriction and provides the neurons in the hidden layer with greater flexibility with respect to function approximation. Other important characteristics of the proposed neural system is that instead of using a single parameter for the output weights, these are functions of the input variables which leads to a significant reduction in the number of hidden units compared with the classical RBF network Finally, we examine the result of applying the proposed algorithm to time series prediction.
We present the hardware/software architecture of an application specific processor for real-time bit stream decoding of MPEG-4 object based profiles. Inclusion of complete motion vector and shape decoding enables effi...
详细信息
ISBN:
(纸本)078037147X
We present the hardware/software architecture of an application specific processor for real-time bit stream decoding of MPEG-4 object based profiles. Inclusion of complete motion vector and shape decoding enables efficient pipelining between the bit stream processor and the remaining parts of the MPEG decoder. Through special extensions to the MIPS instruction and register set, the performance was increased by about 90% while only augmenting the gate count from 25,000 to 40,000.
Here, a simple and new algorithm for blind separation of sources based on geometrical concepts is proposed. This algorithm deals with the instantaneous mixtures and it does not require the estimation of high order sta...
详细信息
Here, a simple and new algorithm for blind separation of sources based on geometrical concepts is proposed. This algorithm deals with the instantaneous mixtures and it does not require the estimation of high order statistics (HOS).
作者：
Shaikh, Fayaz A.McClellan, StanCenter of Telecommunications
Education and Research Department of Electrical and Computer Engineering University of Alabama at Birmingham 1150 Tenth Avenue South BEC 253 BirminghamAL35294-4461 United States IP Architecture Group
Technology and Systems Engineering Compaq Telecommunications 1255 West Fifteenth Street Suite 8000 PlanoTX75075 United States
The primary challenges in the deployment of Voice over Internet Protocol technology include Quality of Service (QoS) guarantees in the form of stringent bounds on end-to-end delay, jitter and loss, as well as rigorous...
详细信息
The design of high-performance, high-precision, real-time digital signal processing (DSP) systems, such as those associated with wavelet signal processing, is a challenging problem. This paper reports on the innovativ...
详细信息
The design of high-performance, high-precision, real-time digital signal processing (DSP) systems, such as those associated with wavelet signal processing, is a challenging problem. This paper reports on the innovative use of the residue number system (RNS) for implementing high-end wavelet filter banks. The disclosed system uses an enhanced index-transformation defined over Galois fields to efficiently support different wavelet filter instantiations without adding any extra cost or additional lookup tables (LUT). An exhaustive comparison against existing two's complement (2C) designs for different custom IC technologies was carried out. These structures have been demonstrated to be well suited for field programmable logic (FPL) assimilation as well as for CBIC (cell-based integrated circuit) technologies.
In this paper we consider the successful hybridation of a two modern computational schemes, Clustering and Neural Networks, for the Predictive Classification of the future value of insect infestation levels for Integr...
详细信息
暂无评论