Search

Designing a fuzzy neural network can be considered as solving an optimization problem. A pruning algorithm for solving the optimization problem is presented based on particle swarm optimization with division of work. It can optimize the fuzzy rules automatically, and determine the structure of the fuzzy neural network based on the effective rules obtained through fuzzy rules searching. Numerical simulations show the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

As a newly emerging technology with some inherent advantages over other wireless networks, wireless mesh networks (WMNs) have received much attention recently. In this paper, we present a cross-layer protocol design for WMNs, which combines the MIMO technique in the physical layer and the opportunistic medium access in the MAC layer into an integrated entity and jointly considers their interactions. In particular, we propose a protocol named opportunistic medium access control in MIMO WMNs (OMAC-MWMN). In an infrastructure/backbone WMN, mesh routers are grouped into clusters, in each of which a cluster head has multiple pending links with its neighbors. This traffic-driven clustering provides a great opportunity for a cluster head to locally coordinate the multiuser medium access. In each iteration of the scheduling, a cluster head opportunistically chooses some compatible neighbors among multiple candidates by utilizing the benefits of multiuser diversity. Then the cluster head simultaneously communicates with multiple selected compatible neighbors through multiuser spatial multiplexing. We formalize the problem of finding a scheduler for a cluster head to select compatible neighbors as finding a compatible pair with the maximum SINR (signal-to-interference and noise ratio) product. Our simulation results show that the proposed protocol can significantly improve system capacity with minimum extra overhead. [ABSTRACT FROM AUTHOR]

A new fuzzy clustering algorithm is proposed. By using kernel methods, this paper maps the data in the original space into a high-dimensional feature space in which a fuzzy dissimilarity matrix is constructed. It not only accurately reflects the difference of attributes among classes, but also maps the difference among samples in the high-dimensional feature space into the two-dimensional plane. Using the particularity of strong global search ability and quickly converging speed of Differential Evolution (DE) algorithms, it optimizes the coordinates of the samples distributed randomly on a plane. The clustering for random distributing shapes of samples is realized. It not only overcomes the dependence of clustering validity on the space distribution of samples, but also improves the flexibility of the clustering and the visualization of high-dimensional samples. Numerical experiments show the effectiveness of the proposed algorithm. Keywords: Fuzzy clustering, kernel methods, differential evolution. [ABSTRACT FROM AUTHOR]