arXiv subjects

Xinyi Xu

Publications and source records attributed to Xinyi Xu.

60 records · Page 4Linked to original sources

arXiv2015

Is the Low-Complexity Mobile-Relay-Aided FFR-DAS Capable of Outperforming the High-Complexity CoMP?

Coordinated multi-point transmission/reception aided collocated antenna system (CoMP-CAS) and mobile relay assisted fractional frequency reuse distributed antenna system (MR-FFR-DAS) constitute a pair of virtual-MIMO based technical options for achieving high spectral efficiency in interference-limited cellular networks. In practice both techniques have their respective pros and cons, which are studied in this paper by evaluating the achievable cell-edge performance on the uplink of multicell systems. We show that assuming the same antenna configuration in both networks, the maximum available cooperative spatial diversity inherent in the MR-FFR-DAS is lower than that of the CoMP-CAS. However, when the cell-edge MSs have a low transmission power, the lower-complexity MR-FFR-DAS relying on the simple single-cell processing may outperform the CoMP-CAS by using the proposed soft-combining based probabilistic data association (SC-PDA) receiver, despite the fact that the latter scheme is more complex and incurs a higher cooperation overhead. Furthermore, the benefits of the SC-PDA receiver may be enhanced by properly selecting the MRs' positions. Additionally, we show that the performance of the cell-edge MSs roaming near the angular direction halfway between two adjacent RAs (i.e. the "worst-case direction") of the MR-FFR-DAS may be more significantly improved than that of the cell-edge MSs of other directions by using multiuser power control, which also improves the fairness amongst cell-edge MSs. Our simulation results show that given a moderate MS transmit power, the proposed MR-FFR-DAS architecture employing the SC-PDA receiver is capable of achieving significantly better bit-error rate (BER) and effective throughput across the entire cell-edge area, including even the "worst-case direction" and the cell-edge boundary, than the CoMP-CAS architecture.

cs.IT↗

arXiv2015

3D Power-map for Smart Grids---An Integration of High-dimensional Analysis and Visualization

Data with features of volume, velocity, variety, and veracity are challenging traditional tools to extract useful analysis for decision-making. By integrating high-dimensional analysis with visualization, this paper develops a 3D power-map animation as an effective solution to the challenge. An architecture design, with detailed data processing procedure, is proposed to realize the integration. Two of the most important components in the architecture are presented: the Single-Ring Law for random matrices as solid mathematic foundation, and the proposed statistical index MSR as high-dimensional data for visualization. The whole procedure is easy in logic, fast in speed, objective and even robust against bad data. Moreover, it is an unsupervised machine learning mechanism directly oriented to the raw data rather than logics or models based on simplifications and assumptions. A case study validates the effectiveness and performance of the developed 3D power-map in analysis extraction.

stat.ME↗

arXiv2012

From Minimax Shrinkage Estimation to Minimax Shrinkage Prediction

In a remarkable series of papers beginning in 1956, Charles Stein set the stage for the future development of minimax shrinkage estimators of a multivariate normal mean under quadratic loss. More recently, parallel developments have seen the emergence of minimax shrinkage estimators of multivariate normal predictive densities under Kullback--Leibler risk. We here describe these parallels emphasizing the focus on Bayes procedures and the derivation of the superharmonic conditions for minimaxity as well as further developments of new minimax shrinkage predictive density estimators including multiple shrinkage estimators, empirical Bayes estimators, normal linear model regression estimators and nonparametric regression estimators.

stat.ME↗

arXiv2010

Asymptotic minimax risk of predictive density estimation for non-parametric regression

We consider the problem of estimating the predictive density of future observations from a non-parametric regression model. The density estimators are evaluated under Kullback--Leibler divergence and our focus is on establishing the exact asymptotics of minimax risk in the case of Gaussian errors. We derive the convergence rate and constant for minimax risk among Bayesian predictive densities under Gaussian priors and we show that this minimax risk is asymptotically equivalent to that among all density estimators.

math.ST↗

arXiv2008

Admissible predictive density estimation

Let $X|μ\sim N_p(μ,v_xI)$ and $Y|μ\sim N_p(μ,v_yI)$ be independent $p$-dimensional multivariate normal vectors with common unknown mean $μ$. Based on observing $X=x$, we consider the problem of estimating the true predictive density $p(y|μ)$ of $Y$ under expected Kullback--Leibler loss. Our focus here is the characterization of admissible procedures for this problem. We show that the class of all generalized Bayes rules is a complete class, and that the easily interpretable conditions of Brown and Hwang [Statistical Decision Theory and Related Topics (1982) III 205--230] are sufficient for a formal Bayes rule to be admissible.

math.ST↗

arXiv2006

Improved minimax predictive densities under Kullback--Leibler loss

Let $X| μ\sim N_p(μ,v_xI)$ and $Y| μ\sim N_p(μ,v_yI)$ be independent p-dimensional multivariate normal vectors with common unknown mean $μ$. Based on only observing $X=x$, we consider the problem of obtaining a predictive density $\hat{p}(y| x)$ for $Y$ that is close to $p(y| μ)$ as measured by expected Kullback--Leibler loss. A natural procedure for this problem is the (formal) Bayes predictive density $\hat{p}_{\mathrm{U}}(y| x)$ under the uniform prior $π_{\mathrm{U}}(μ)\equiv 1$, which is best invariant and minimax. We show that any Bayes predictive density will be minimax if it is obtained by a prior yielding a marginal that is superharmonic or whose square root is superharmonic. This yields wide classes of minimax procedures that dominate $\hat{p}_{\mathrm{U}}(y| x)$, including Bayes predictive densities under superharmonic priors. Fundamental similarities and differences with the parallel theory of estimating a multivariate normal mean under quadratic loss are described.

math.ST↗