Alpha-cut implemented fuzzy clustering algorithms and switching regressions.

In the fuzzy c-means (FCM) clustering algorithm, almost none of the data points have a membership value of 1. Moreover, noise and outliers may cause difficulties in obtaining appropriate clustering results from the FCM algorithm. The embedding of FCM into switching regressions, called the fuzzy c-regressions (FCRs), still has the same drawbacks as FCM. In this paper, we propose the alpha-cut implemented fuzzy clustering algorithms, referred to as FCMalpha, which allow the data points being able to completely belong to one cluster. The proposed FCMalpha algorithms can form a cluster core for each cluster, where data points inside a cluster core will have a membership value of 1 so that it can resolve the drawbacks of FCM. On the other hand, the fuzziness index m plays different roles for FCM and FCMalpha. We find that the clustering results obtained by FCMalpha are more robust to noise and outliers than FCM when a larger m is used. Moreover, the cluster cores generated by FCMalpha are workable for various data shape clusters, so that FCMalpha is very suitable for embedding into switching regressions. The embedding of FCMalpha into switching regressions is called FCRalpha. The proposed FCRalpha provides better results than FCR for environments with noise or outliers. Numerical examples show the robustness and the superiority of our proposed methods.     

Alternative c-means clustering algorithms

In this paper we propose a new metric to replace the Euclidean norm in c-means clustering procedures. On the basis of the robust statistic and the influence function, we claim that the proposed new metric is more robust than the Euclidean norm. We then create two new clustering methods called the alternative hard c-means (AHCM) and alternative fuzzy c-means (AFCM) clustering algorithms. These alternative types of c-means clustering have more robustness than c-means clustering. Numerical results show that AHCM has better performance than HCM and AFCM is better than FCM. We recommend AFCM for use in cluster analysis. Recently, this AFCM algorithm has successfully been used in segmenting the magnetic resonance image of Ophthalmology to differentiate the abnormal tissues from the normal tissues.     

Optimal lot sizing with maintenance actions and imperfect production processes

Porteus (1986) explored an economic order quantity model with imperfect production processes that the approximate lot size is derived. Basically, he dealt with the lot size problem is rather meaningful. However, for mathematical simplicity, he adopted a truncated Taylor series expansion to present the approximate expected total cost function that results in overvalue of expected total cost. In this paper, we extend Porteus (1986) to present the optimal lot size model for defective items with a constant probability when the system is out-of-control and taking the maintenance cost into account. We show that there exists a unique optimal lot size such that the expected total cost is minimised. In addition, the bounds of optimal lot size are provided to develop the solution procedure. Finally, numerical examples are given to illustrate the theoretical results and compare optimal solutions obtained by using our approach and Porteus's approach. Numerical results show that our approach is better.     

Implementation of first-order and second-order microwave differentiators

Simple and accurate formulations are employed to represent discrete-time infinite impulse response processes of both first- and second-order differentiators in the Z-domain. These formulations, in conjunction with the representations of transmission-line elements in the Z-domain, lead to transmission-line configurations that are eligible for wide-band microwave differentiators. Both the first- and second-order differentiators in microstrip circuits are implemented to verify this method. The experimental results are in good agreement with simulation values.     

A novel two-degree-of-freedom controller design for a permanent magnet linear synchronous motor control system

This paper proposes a novel two-degree-of-freedom optimal controller design for a permanent magnet linear synchronous motor position-control system. The param-eters of the controller are obtained by using a frequency-domain optimization technique. A systematic design of the controller and the detailed implementation of the proposed system are discussed. The closed-loop control system possesses good transient responses and good load disturbance responses. In addition, the system has a good tracking ability. Several experimental results are provided to validate the theoretical analysis.     

A Fair Scheme for Multi-channel Selection in Vehicular Wireless Networks

In recent years, the IEEE 802.11p/1609 wireless access in vehicular environments standards adopt the dedicated short-range communications multi-channel architecture for vehicular wireless networks. To utilize the multi-channel architecture, each vehicle equipped with two sets of transceivers can operate concurrently on three different channels. For example, in cluster-based multi-channel schemes, a cluster head vehicle coordinates and assigns an appropriate channel to its cluster members. However, these schemes are unsuitable for a single channel device performing on only one RF channel at a time which would waste channel resource and increase time to allocate a channel. Another approach, called LEACH-based scheme, selects channels randomly and ensures that each channel is selected once within a round in each vehicle. However, this leads to a situation that different vehicles might select the same channel in short-term duration. In this paper, we propose a multi-channel selection scheme, called minimum duration counter (MDC) scheme, which could apply to a single channel device, while utilizing the multi-channel architecture of an 802.11p/1609 network. In addition, we compare the MDC scheme with pure random (PR) and LEACH-based schemes in terms of fairness index (FI) and utilization to emphasize the fairness and to balance the traffic of multi-channel usage. Furthermore, we analyze the counter overflow probability distribution and propose solutions to the MDC scheme. Numerical results show that our scheme outperforms the PR and LEACH-based schemes in terms of multi-channel usage, traffic balancing, and fairness.     

Anisotropic thermal conductivity of nanoporous silica film

In this paper, thermal conductivity of porous silica film with porosity from 21 to 64% was studied comprehensively. The corresponded dielectric constant is from 2.5 to 1.5. It is observed that the porous silica material has strong anisotropic characteristic. A serial-parallel hybrid model is proposed to explain the correlation between porosity and thermal conductivity in both in-plane and cross-plane components. The pores in the higher porosity silica film tend to distribute horizontally. This distribution of the pores in the dielectric film is the main factor that induces the anisotropic characteristic. The nonuniform distribution of pores also makes the conventional two-dimensional model of 3u/spl grave/ method inappropriate for extracting the in-plane thermal conductivity. A new method based on the hybrid model was proposed to extract the in-plane thermal conductivity successfully. The anisotropic characteristic of the thermal conductivity may be accompanied by the anisotropic dielectric constant, which will greatly complicate the thermal management and resistance-capacitance delay simulation of the circuits and should be avoided. The proposed model would be helpful on evaluation of new porous low dielectric constant materials.     

Mountain c-regressions method

Since Quandt [The estimation of the parameters of a linear regression system obeying two separate regimes, Journal of the American Statistical Association 53 (1958) 873-880] initiated the research on 2-regressions analysis, switching regression had been widely studied and applied in psychology, economics, social science and music perception. In fuzzy clustering, the fuzzy c-means (FCM) is the most commonly used algorithm. Hathaway and Bezdek [Switching regression models and fuzzy clustering, IEEE Transactions on Fuzzy Systems 1 (1993) 195-204] embedded FCM into switching regression where it was called fuzzy c-regressions (FCR). However, the FCR always depends heavily on initial values. In this paper, we propose a mountain c-regressions (MCR) method for solving the initial-value problem. First, we perform data transformation for the switching regression data set, and then implement the modified mountain clustering on the transformed data to extract c cluster centers. These extracted c cluster centers in the transformed space will correspond to c regression models in the original data set. The proposed MCR method can form well-estimated c regression models for switching regression data sets. According to the properties of transformation, the proposed MCR is also robust to noise and outliers. Several examples show the effectiveness and superiority of our proposed method.     

Analysis of parameter selections for fuzzy c-means

The weighting exponent m is called the fuzzifier that can influence the performance of fuzzy c-means (FCM). It is generally suggested that m∈[1.5,2.5]. On the basis of a robust analysis of FCM, a new guideline for selecting the parameter m is proposed. We will show that a large m value will make FCM more robust to noise and outliers. However, considerably large m values that are greater than the theoretical upper bound will make the sample mean a unique optimizer. A simple and efficient method to avoid this unexpected case in fuzzy clustering is to assign a cluster core to each cluster. We will also discuss some clustering algorithms that extend FCM to contain the cluster cores in fuzzy clusters. For a large theoretical upper bound case, we suggest the implementation of the FCM with a suitable large m value. Otherwise, we suggest implementing the clustering methods with cluster cores. When the data set contains noise and outliers, the fuzzifier m=4 is recommended for both FCM and cluster-core-based methods in a large theoretical upper bound case.     

Alternative learning vector quantization

In this paper, we discuss the influence of feature vectors contributions at each learning time t on a sequential-type competitive learning algorithm. We then give a learning rate annealing schedule to improve the unsupervised learning vector quantization (ULVQ) algorithm which uses the winner-take-all competitive learning principle in the self-organizing map (SOM). We also discuss the noisy and outlying problems of a sequential competitive learning algorithm and then propose an alternative learning formula to make the sequential competitive learning robust to noise and outliers. Combining the proposed learning rate annealing schedule and alternative learning formula, we propose an alternative learning vector quantization (ALVQ) algorithm. Some discussion and experimental results from comparing ALVQ with ULVQ show the superiority of the proposed method.