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.

The idea of cognitive radio (CR) has embodied concretely in hierarchical cellular systems by deploying an underlying microcellular system to reuse the underutilized spectrum licensed by an macrocellular system. The fundamental challenges for successfully realizing such hierarchical systems are to manage the intercell interference between the macrocell and microcell and to pursue the goal of maximizing the spectrum recycling efficiency. In this paper, we jointly consider antenna beamforming, power allocation, and multiuser scheduling for the microcellular system to opportunistically utilize the uplink spectrum of the macrocell and to concurrently serve multiple secondary users in the downlink. With the objective of maximizing the sum rate, the three-dimensional joint design problem is often formulated as a mixed integer nonlinear programming (MINLP) which is an NP-hard problem and very complicated to solve. We resort to semidefinite relaxation (SDR) techniques to convert to the cumbersome optimization problem into a convex problem by introducing an interference-related auxiliary variable. An iterative algorithm based on semidefinite programming is proposed to achieve the optimal solution. The zero-forcing (ZF) beamforming and the singular value decomposition (SVD)-based beamforming with the best scheduling are simulated for performance comparisons, and our simulation shows that the proposed scheme is much superior to the ZF scheme and quite close to the SVD scheme with a slight performance gap of 1 bps/Hz.

In this study, a novel spatial information based support vector machine for hyperspectral image classification, named spatial-contextual semi-supervised support vector machine (SC3SVM), is proposed. This approach modifies the SVM algorithm by using the spectral information and spatial-contextual information. The concept of SC3SVM is to utilize other information, obtain from the pixels of a neighborhood system in the spatial domain, to modify the effective of each patterns. Experimental results show a sound performance of classification on the famous hyperspectral images, Indian Pine site. Especially, the overall classification accuracy of whole hyperspectral image (Indian Pine site with 16 classes) is up to 96.4%, the kappa accuracy is up to 95.9%.

Each pixel of the hyperspectral image is composed of hundreds of individual bands. Usually, these pixels are considered as high dimensional vectors. NWFE is a very robust and superior feature extraction method in this aspect of view of image pixel. On the other hand, since adjacent bands in a pixel are usually highly correlated, each pixel can also be viewed as a time series or signal. Therefore, the classification of hyperspectral data becomes the problem of distinguishing between different time series. As the consequence, time series discrimination methods, such as SLEX related time series methods, can then be applied in the classification of hyperspectral image. In this paper, a selection ensemble of NWFE and SLEX is proposed for classifying multi-group hyperspectral image. The performance of the proposed scheme is compared to SLEX and NWFE both by simulation data set and real hyperspectral image dataset, Washington DC Mall. These results show that the proposed scheme has higher testing data classification accuracy than others.