A network infrastructure for IP mobility support in metropolitan areas

The original design of the Internet and its underlying protocols did not anticipate users to be mobile. With the growing interest in supporting mobile users and mobile computing, a great deal of work is taking place to solve this problem. For a solution to be practical, it has to integrate easily with existing Internet infrastructure and protocols, and offer an adequate migration path toward what might represent the ultimate solution. In that respect, the solution has to be incrementally scalable to handle a large number of mobile users and wide geographical scopes, and well performing so as to support all application requirements including voice and video communications and a wide range of mobility speeds. In this paper, we present a survey of the state-of-the-art and propose a scalable infrastructure to support mobility in Internet protocol networks. In that respect, we exploit local area network (LAN) technologies to create the network infrastructure necessary to offer connectivity to mobile users across any geographical area (building, campus and metropolis). The intrinsic properties of LAN technologies and their underlying protocols, namely flat address space, transparent learning and low complexity renders this solution particularly cost effective for supporting user mobility. In particular, we propose a network topology and a set of protocols that render the infrastructure scalable to a large geographical area and many users.     

Shape optimization for tumor location

In non-invasive thermal diagnostics, accurate correlations between the thermal image at skin surface and interior human physiology are desired. In this work, an estimation methodology to determine unknown geometrical parameters of an embedded tumor is proposed. We define a functional that represents the mismatch between a measured experimental temperature profile, which may be obtained by infrared thermography on the skin surface, and the solution of an appropriate boundary problem. This functional is related to the geometrical parameters through the solution of the boundary problem, in such a way that finding the minimum of this functional form also means finding the unknown geometrical parameters of the embedded tumor. Sensitivity analysis techniques coupled with the adjoint method were considered to compute the shape derivative of the functional. Then, a nonmonotone spectral projected gradient method was implemented to solve the optimization problem of finding the optimal geometric parameters.     

An effective location fingerprint model for wireless indoor localization

A model for predicting precision and accuracy performance of indoor fingerprint based positioning systems is very desirable for system designers as it helps estimate the probability of location selection before actual deployment. Such information can be used to tune the fingerprint database or improve the offline fingerprint collection phase. This paper presents a new analytical model that applies proximity graphs for approximating the probability distribution of error distance given a location fingerprint database using WLANs received signals, and its associated statistics. Simulations are used to validate the analytical model, which is found to produce results close to that from simulations. The model permits an analysis of the internal structure of location fingerprints. We employ the analysis of the fingerprint structure to identify and eliminate inefficient location fingerprints stored in the fingerprint database. Knowledge of where the inefficient fingerprints are can potentially be employed in a better location fingerprint collecting scheme from a grid system in the offline phase.     

An integrated model for facility location and technology acquisition

This paper presents an analytical approach for simultaneous optimization of the facility location, capacity acquisition and technology selection decisions. The proposed approach can be useful when there is considerable interaction between these structural decisions, e.g., in global manufacturing companies. In these cases, a sequential optimization approach is bound to produce sub-optimal facility designs, and hence an integrated model is needed. We present a formal definition of the facility location and technology acquisition problem, and provide a basic model. We describe the analytical properties of the model, which led to the development of a solution procedure. We report on a set of experiments, and conclude that the computational performance of the proposed algorithm is quite satisfactory.     

An enhanced model for evacuation vulnerability assessment in urban areas

Rapid urbanization has led to a massive influx of people into cities. When many people congregate in urban areas, crowd crushing emergencies are likely to occur. If vulnerable areas with potential evacuation problems are detected in advance, crowd crushing emergencies may be minimized or even avoided. Thus, an evacuation vulnerability assessment from a precautionary perspective is fundamental. However, the current evacuation vulnerability assessment models are limited in spreading time estimation and evacuation capacity evaluation. To mitigate these limitations, in this study, we propose an enhanced model to quantitatively assess the evacuation vulnerability in urban areas. Our model enhances the current models in two ways. First, we employ a hexagon gridding scheme to construct a network to meet the prerequisite of evacuation spreading at the equal time intervals. Second, we quantify the grid connectivity on the network by considering the grid capacity to avoid underestimation of the evacuation vulnerability. Using a mobile phone location dataset of Shanghai, we systematically investigate the evacuation vulnerability of urban areas in a fine-grained spatio-temporal scale. Areas that may encounter evacuation problems to various degrees can be identified in advance. This information can support emergency management agencies in monitoring dense crowds and ensure early warnings of potential crowd crushing emergencies.     

An optimization model for the container pre-marshalling problem

In most container yards around the world, containers are stacked high to utilize yard space more efficiently. In these yards, one major factor that affects their operational efficiency is the need to re-shuffle containers when accessing a container that is buried beneath other containers. One way to achieve higher loading efficiency is to pre-marshal the containers in such a way that it fits the loading sequence. In this research, we present a mathematical model for the container pre-marshalling problem. With respect to a given yard layout and a given sequence that containers are loaded onto a ship, the model yields a plan to re-position the export containers within the yard, so that no extra re-handles will be needed during the loading operation. The optimization goal is to minimize the number of container movements during pre-marshalling. The resulting model is an integer programming model composed of a multi-commodity flow problem and a set of side constraints. Several possible variations of the model as well as a solution heuristic are also discussed. Computation results are provided.     

An aggregated optimization model for multi-head SMD placements

In this article we propose an aggregate optimization approach by formulating the multi-head SMD placement optimization problem into a mixed integer program (MIP) with the variables based on batches of components. This MIP is tractable and effective in balancing workload among placement heads, minimizing the number of nozzle exchanges, and improving handling class. The handling class which specifies the traveling speed of the robot arm, to the best of our knowledge, has been for the first time incorporated in an optimization model. While the MIP produces an optimal planning for batches of components, a new sequencing heuristics is developed in order to determine the final sequence of component placements based on the outputs of the MIP. This two-stage approach guarantees a good feasible solution to the multi-head SMD placement optimization problem. The computational performance is examined using real industrial data.     

An LP-based hyperparameter optimization model for language modeling

In order to find hyperparameters for a machine learning model, algorithms such as grid search or random search are used over the space of possible values of the models’ hyperparameters. These search algorithms opt the solution that minimizes a specific cost function. In language models, perplexity is one of the most popular cost functions. In this study, we propose a fractional nonlinear programming model that finds the optimal perplexity value. The special structure of the model allows us to approximate it by a linear programming model that can be solved using the well-known simplex algorithm. To the best of our knowledge, this is the first attempt to use optimization techniques to find perplexity values in the language modeling literature. We apply our model to find hyperparameters of a language model and compare it to the grid search algorithm. Furthermore, we illustrate that it results in lower perplexity values. We perform this experiment on a real-world dataset from SwiftKey to validate our proposed approach.     

不确定条件下应急资源布局的鲁棒双层优化模型

针对非常规突发事件中应急资源布局问题,在受灾点需求不确定和应急救援过程分为多个阶段的情景下,建立了省市两级应急储备仓库定位和物资配置的鲁棒双层规划模型。运用相对鲁棒优化方法,将上述具有不确定性系数的双层规划模型转化为从者无关联的确定性线性双层规划,提出了一种混合遗传算法进行求解,实现了省市两级应急资源布局的协同优化。通过实例验证了模型及算法的可行性和有效性。     

An integrated quantitative and qualitative FMCDM model for location choices

International logistics is a very popular and important issue in the present international supply chain system. In order to reduce the international supply chain operation cost, it is very important for enterprises to invest in the international logistics centers. Although a number of research approaches for solving decision-making problems have been proposed, most of these approaches focused on developing quantitative models for dealing with objective data or qualitative models for dealing with subjective ratings. Few researchers proposed approaches for dealing with both objective data and subjective ratings. Thus, this paper attempts to fill this gap in current literature by establishing an integrated quantitative and qualitative fuzzy multiple criteria decision-making model for dealing with both objective crisp data and subjective fuzzy ratings. Finally, the utilization of the proposed model is demonstrated with a case study on location choices of international distribution centers.     

Endogenous rise and collapse of housing price: An agent-based model of the housing market

On the basis of interviews with local real estate agents, this study develops an agent-based model of housing market to determine the cause of rise and collapse of US housing price during the years immediately preceding the US financial crisis (2007–2009). We study the key factors affecting housing price volatility, such as lenient financing and speculation. The dynamic simulation findings in the study show in concrete terms how lenient lending practices combined with speculation can lead to increased volatility in housing price, including sharp rises immediately followed by collapses. The exploratory work in this study will contribute to the understanding of the causes of housing bubbles and inform policy decisions.     

一种基于BP-GA的室内定位模型

使用BP-GA算法相结合的方法来作为室内定位模型的主要定位算法,在得到初始坐标后利用泰勒级数定位算法优化得到最终待测点坐标。该算法很好地减弱了室内环境对定位精度的影响,并且具有较高的精度。仿真实验验证了该模型的有效性。     

网络安全态势评估指标体系优化模型研究

针对指标选取的主观性带来的评估结果准确率低、实时性较差等问题,提出了基于因子分析法和主成分分析法的网络安全态势评估指标优化模型。该模型可以用一组具有较强独立性的综合变量来描述原有的指标体系,从而减少网络安全评估时的计算量。实验表明,模型在不影响准确率的情况下能够得出较为实时的评估结果。     

An automatic hyperparameter optimization DNN model for precipitation prediction

Applied Intelligence - Deep neural networks (DNN) have gained remarkable success on many rainfall predictions tasks in recent years. However, the performance of DNN highly relies upon the...     

宽带城域网网络安全与优化

随着互联网技术的不断发展,我国互联网事业的发展取得了很大的进步。但是随之而来的网络安全问题也逐渐显现,现在我国的网络安监部门进行数据统计分析,用在网络安全问题上的资金也是很大的数字,主要就是防火墙、防病毒和加密等方面。现在,我国网络安全问题已经越来越受重视,企业也意识到了网络安全的重要。本文主要介绍了城域网络安全的重要性以及城域网络安全优化和城域网核心层网络安全保护措施等,通过简单的介绍使大家了解城域网络安全对于我们日常工作和生活的影响。     

An exploratory study on mobile banking adoption in Indian metropolitan and urban areas: a scenario-based experiment

Security concerns have thwarted the widespread adoption of mobile banking in India. To respond to the concerns of Indian banks and their customers, we present in this paper our exploratory attempts to understand how the levels of security affect perceived risk and control and ultimately, adoption of mobile banking by Indian customers. This study also examines the moderating influence of the type of city on the relationship between security levels and risk/control perceptions associated with mobile banking. Using a scenario-based experiment, we classified security-enhancing approaches into three categories and examined their effectiveness in decreasing Indian customers’ perceived risk, increasing their perceived control, and then in turn, facilitating mobile banking adoption. Our findings reveal the important role of perceived risk and control in influencing customers’ intention to adopt mobile banking. Moreover, perceived risk and control significantly influenced mobile banking adoption by customers in urban areas, but only perceived control significantly influenced mobile banking adoption by metropolitan customers. Additional analyses show that customers’ risk and control perceptions differ according to the level of security; however, these perceptions do not have a significant influence on risk and control.     

An optimization model for collaborative recommendation using a covariance-based regularizer

This paper suggests a convex regularized optimization model to produce recommendations, which is adaptable, fast, and scalable—while remaining very competitive to state-of-the-art methods in terms of accuracy. We introduce a regularizer based on the covariance matrix such that the model minimizes two measures ensuring that the recommendations provided to a user are guided by both the preferences of the other users in the system and the known preferences of the user being processed. It is adaptable since (1) it can be viewed from both user and item perspectives (allowing to choose, depending on the task, the formulation with fewer decision variables) and (2) multiple constraints depending on the context (and not only based on the accuracy, but also on the utility of personalized recommendations) can easily be added, as shown in this paper through two examples. Since our regularizer is based on the covariance matrix, this paper also describes how to improve computational and space complexities by using matrix factorization techniques in the optimization model, leading to a fast and scalable model. To illustrate all these concepts, experiments were conducted on four real datasets of different sizes (i.e., FilmTrust, Ciao, MovieLens, and Netflix) and comparisons with state-of-the-art methods are provided, showing that our context-sensitive approach is very competitive in terms of accuracy.