We consider optimal filter-and-forward (FF) beam-forming (BF) in multi-user peer-to-peer (MUP2P) relay networks with frequency selective fading in which the signals received at a destination are subject to severe inte...
详细信息
ISBN:
(纸本)9781467355766
We consider optimal filter-and-forward (FF) beam-forming (BF) in multi-user peer-to-peer (MUP2P) relay networks with frequency selective fading in which the signals received at a destination are subject to severe inter-symbol interference (ISI) and multi-user interference (MUI). In the FF protocol, the amount of interference not only depends on the choice of the filter weights at the relays but also on the choice of a decoding delay at the destinations. State-of-the-art methods assume the network to be perfectly synchronized and select the same fixed decoding delay for all destinations. In a practical MUP2P scenario, however, users experience different path delays not only with respect to different relays but also with respect to other users. Hence, unlike in a conventional cellular network, perfect synchronization is impossible in practice and a common decoding delay for all users is not a good'choice. Therefore, we propose to jointly optimize the filter weights at the relays and the individual decoding delays at the destinations to effectively mitigate ISI and MUI. The proposed method outperforms the existing approaches most significantly in asynchronous networks.
Graph-based semi-supervised learning (SSL) methods play an increasingly important role in practical machine learning systems, particularly in agnostic settings when no parametric information or other prior knowledge i...
详细信息
Graph-based semi-supervised learning (SSL) methods play an increasingly important role in practical machine learning systems, particularly in agnostic settings when no parametric information or other prior knowledge is available about the data distribution. Given the constructed graph represented by a weight matrix, transductive inference is used to propagate known labels to predict the values of all unlabeled vertices. Designing a robust label diffusion algorithm for such graphs is a widely studied problem and various methods have recently been suggested. Many of these can be formalized as regularized function estimation through the minimization of a quadratic cost. However, most existing label diffusion methods minimize a univariate cost with the classification function as the only variable of interest. Since the observed labels seed the diffusion process, such univariate frameworks are extremely sensitive to the initial label choice and any label noise. To alleviate the dependency on the initial observed labels, this article proposes a bivariate formulation for graph-based SSL, where both the binary label information and a continuous classification function are arguments of the optimization. This bivariate formulation is shown to be equivalent to a linearly constrained Max-Cut problem. Finally an efficient solution via greedy gradient Max-Cut (GGMC) is derived which gradually assigns unlabeled vertices to each class with minimum connectivity. Once convergence guarantees are established, this greedy Max-Cut based SSL is applied on both artificial and standard benchmark data sets where it obtains superior classification accuracy compared to existing state-of-the-art SSL methods. Moreover, GGMC shows robustness with respect to the graph construction method and maintains high accuracy over extensive experiments with various edge linking and weighting schemes.
In this article we consider a difficult combinatorial optimization problem arising from the operation of a system for testing electronic circuit boards (ECB). This problem was proposed to us by a company that makes a ...
详细信息
In this article we consider a difficult combinatorial optimization problem arising from the operation of a system for testing electronic circuit boards (ECB). This problem was proposed to us by a company that makes a system for testing ECBs and is looking for an efficient way of planning the tests on any given ECB. Because of its difficulty, we first split the problem into a covering subproblem and a sequencing subproblem. We also give a global formulation of the test planning problem. Then we present and discuss results pertaining to the covering and sequencing subproblems. These results demonstrate that their solution yields testing plans that are much better than those currently used by the company. Finally we conclude our article by outlining avenues for future research.
In this paper, we propose a modified evolutionary programming with dynamic domain for solving nonlinear IP/MIP problems with linear constraints, without involving penalty function or any transformation for the problem...
详细信息
In this paper, we propose a modified evolutionary programming with dynamic domain for solving nonlinear IP/MIP problems with linear constraints, without involving penalty function or any transformation for the problem to a linear model or others. The numerical results show that the new algorithm gives a satisfactory performance in which it works of high speed and accuracy in IP/MIP problems.
暂无评论