The above paper mistakenly employs the incorrect in- equalities of and in Appendix B, and as a result, it leads to incorrect derivation procedures of the lower and the upper-bounds on the average mutual information.

Physical layer multicasting in wireless networks has been proposed to efficiently send an identical message to multiple users simultaneously. In this paper, we consider multiple antenna multicasting where the transmitter is equipped with Mtantennas and data is transmitted to K single-antenna users. When the downlink channels are assumed to follow an uncorrelated Rayleigh distribution, the multicasting capacity scaling for the large K asymptote is known. On the other hand, the effect of channel spatial correlation on the capacity performance has not been well addressed. Therefore, we investigate the effect of correlation using the channel correlation information at the transmitter. Using extreme value theory, it is shown that signaling using uniformly allocated transmit powers on the spatial channel correlation matrix's eigenvectors with non-zero eigenvalues approaches the multicasting capacity for the large K asymptote. Compared to the performance of uncorrelated fading channels with a constraint on the trace of the spatial correlation matrix, it is shown that the channel correlation degrades the performance. Specifically, if the correlation matrix is full rank the asymptotic outage (average) capacity ratio, which is defined as the ratio of the outage (average) capacities of the correlated fading channel and the uncorrelated fading channel as K goes to infinity, is equivalent to the geometric mean of the eigenvalues of the transmit channel correlation matrix of the correlated channel. On the other hand, if the correlation matrix is not full rank, the capacity ratios become zero. To assess the value of knowledge of the channel correlation information at the transmitter, we compare the asymptotic performances with and without knowledge of the channel correlation information. Specifically, the asymptotic performance improvement due to this correlation knowledge becomes (Mt/M)1/Mfor the large K asymptote, where M is the rank of the transmit channel correlation matrix. In addition, we discuss the issues of nonidentical transmit channel correlations among the users and correlation between users' channels.

Abstract: We used several machine learning techniques to predict the susceptibility to chronic hepatitis from single nucleotide polymorphism (SNP) data. These are integrated with several feature selection algorithms to identify a set of SNPs relevant to the disease. In addition, we apply a backtracking technique to a couple of feature selection algorithms, forward selection and backward elimination, and show that it is beneficial to find the better solutions by experiment. The experimental results show that the decision rule is able to distinguish chronic hepatitis from normal with a maximum accuracy of 73.20%, whereas the accuracy of the support vector machine is 67.53% and that of the decision tree is 72.68%. It is also shown that the decision tree and decision rule are potential tools to predict susceptibility to chronic hepatitis from SNP data.