Browsing by Author "Liu, Yu-Lung"

Now showing 1 - 2 of 2

An Automatic Method for Selecting the Parameter of the Normalized Kernel Function to Support Vector Machines
(2012-01-01)
Li, Cheng-Hsuan
;
Ho, Hsin-Hua
;
Liu, Yu-Lung
;
Lin, Chin-Teng
;
Kuo, Bor-Chen
Soft-margin support vector machine (SVM) is one of the most powerful techniques for supervised classification. However, the performances of SVMs are based on choosing the proper kernel functions or proper parameters of a kernel function. It is extremely time consuming by applying the k-fold cross-validation (CV) to choose the almost best parameter. Nevertheless, the searching range and fineness of the grid method should be determined in advance. In this paper, an automatic method for selecting the parameter of the normalized kernel function is proposed. In the experimental results, it costs very little time than k-fold cross-validation for selecting the parameter by our proposed method. Moreover, the corresponding soft-margin SVMs can obtain more accurate or at least equal performance than the soft-margin SVMs by applying k-fold cross-validation to determine the parameters.
Joint Antenna Beamforming, Multiuser Scheduling, and Power Allocation for Hierarchical Cellular Systems
(2015-05-01)
Ku, Meng-Lin
;
Wang, Li-Chun
;
Liu, Yu-Lung
;
電機資訊學士班
;
Undergraduate Honors Program of Electrical Engineering and Computer Science
Cognitive radio (CR) is concretely embodied in hierarchical cellular systems by deploying an underlying microcellular system to reuse the underutilized spectrum of a macrocellular system. One of the key challenges to the success of hierarchical cellular systems is to manage the intercell interference between the macrocell and the microcell and to maximize the spectrum efficiency. In this paper, antenna beamforming, power allocation, and multiuser scheduling are jointly designed to opportunistically utilize the macrocell\'s uplink spectrum in serving multiple secondary microcellular users concurrently. The joint design of antenna beamforming, power allocation, and scheduling with the objective of maximizing the sum rate is indeed a mixed-integer nonlinear programming NP-hard problem. The proposed simpler iterative subgradient projection and semidefinite programming approach can obtain better performance than the conventional zero-forcing beamforming. Furthermore, unlike the optimal singular value decomposition (SVD) beamforming that requires all users to have channel knowledge at the receiver for cooperation, the proposed joint design methodology requires no channel knowledge at the receiver and can outperform the SVD beamforming without user scheduling. When considering both implementation complexity and performance enhancement issues, the proposed joint power allocation, multiuser scheduling, and antenna beamforming technique can help provide important insights into the design of interference management techniques for hierarchical CR systems.