Mining categorical sequences from data using a hybrid clustering method

The identification of different dynamics in sequential data has become an every day need in scientific fields such as marketing, bioinformatics, finance, or social sciences. Contrary to cross-sectional or static data, this type of observations (also known as stream data, temporal data, longitudinal data or repeated measures) are more challenging as one has to incorporate data dependency in the clustering process. In this research we focus on clustering categorical sequences. The method proposed here combines model-based and heuristic clustering. In the first step, the categorical sequences are transformed by an extension of the hidden Markov model into a probabilistic space, where a symmetric Kullback–Leibler distance can operate. Then, in the second step, using hierarchical clustering on the matrix of distances, the sequences can be clustered. This paper illustrates the enormous potential of this type of hybrid approach using a synthetic data set as well as the well-known Microsoft dataset with website users search patterns and a survey on job career dynamics.     

Efficient exponential-time algorithms for edit distance between unordered trees

The edit distance problem for rooted unordered trees is known to be NP-hard. Based on this fact, this paper studies exponential-time algorithms for the problem. For a general case, an $O(\min ({1.26}^{n_1+n_2},2^{b_1+b_2}\cdot \text{<mi mathvariant="italic" is="true">poly</mi>}(n_1,n_2)))$ time algorithm is presented, where $n_1$ and $n_2$ are the numbers of nodes and $b_1$ and $b_2$ are the numbers of branching nodes in two input trees. This algorithm is obtained by a combination of dynamic programming, exhaustive search, and maximum weighted bipartite matching. For bounded degree trees over a fixed alphabet, it is shown that the problem can be solved in $O({(1+ϵ)}^{n_1+n_2})$ time for any fixed $ϵ>0$. This result is achieved by avoiding duplicate calculations for identical subsets of small subtrees.     

An energy efficient and QoS aware routing protocol for wireless sensor and actuator networks

Wireless sensor and actuator networks are composed of sensor and actuator nodes interconnected via wireless links. The actuators are responsible for taking prompt decisions and react accordingly to the data gathered by sensor nodes. In order to ensure efficient actions in such networks, we propose a new routing protocol that provides QoS in terms of delay and energy consumption. The network is organized in clusters supervised by CHs (Cluster-Heads), elected according to important metrics, namely the energy capability, the riches of connectivity, which is used to select the CH with high node density, and the accessibility degree regarding all the actuators. The latter metric is the distance in number of hops of sensor nodes relative to the actuator nodes. This metric enhances more the network reliability by reducing the communication delay when alerting the actuator nodes, and hence, reducing the energy consumption. To reach efficiently the actuator nodes, we design a delay and energy sensitive routing protocol based on-demand routing approach. Our protocol incurs less delay and is energy efficient. We perform an evaluation of our approach through simulations. The obtained results show out performance of our approach while providing effective gain in terms of communication delay and energy consumption.     

高性能数字电视QAM均衡器VLSI结构设计

本文设计了一种适用于高清晰数字电视(HDTV)接收芯片的全数字正交幅度调制器(QAM)的均衡器。该均衡器由前馈滤波器、误差判别电路和系数更新电路以及后馈滤波器构成。该均衡器采用了常模算法(CMA)和判决导引最小均方算法(DD-LMS)相结合的算法。重点给出了均衡器的VLSI实现、两种算法间切换的依据、步长的选择以及抽头系数的确定。同时在电路上采用了逻辑简化、重编码、电路时分复用等简化和优化方法来实现性能、面积和功耗的折衷。     

深亚微米条件下的多层VLSI／MCM层分配的遗传算法

本文提出了一个深亚微米条件下的多层ＶＬＳＭＣＭ有约束分层层分配的遗传算法。该算法分为两步：首先进行超层分配，使各线网满足Ｃｒｏｓｓｔａｌｋ约束，且超层数目最少；然后进行各超层的通孔最少化二分层。与目前的层分配算法相比，该遗传算法具有目标全面，全局优化能力强等特点，是一种可应用于深亚微米条件下的ＩＣ ＣＡＤ的有效分层方法。     

FM调频偏移高精度测量方法

在FM调制信号调频频偏参数测量过程中,测量准确度与载波偏移误差、解调带宽、采样速率等密切相关.本文提出的FM调频偏移高精度测量方法,首先采用相位差分法从获得的IQ信号数据中提取瞬时频率波形,然后利用戈泽尔算法计算出瞬时频率波形中存在的直流偏移,并进行去直流处理,最后采用加窗SINC正弦内插方法对瞬时频率波形进行插值,从内插的波形中获得调频偏移.实际应用结果表明,采用该方法能够满足调频频偏测量准确度优于1%的测量要求.     

基于软件无线电的突发模式8DPSK的全数字化实现

提出了一种软件无线电中频收发机的结构和突发模式 8DPSK解调的实现框架 ,在 8DPSK解调时采用了前向初始参数估计算法 ,并证实了该算法的可行性。     

基于MIV-BP型网络实验的房地产项目风险识别研究

为了更准确更客观地识别房地产项目中的风险,为房地产项目投资决策提供科学依据和参考,有效地规避风险,本研究在BP神经网络 (Back-Propagation Neural Network)建模的基础上,采取MIV(Mean Impact Value)算法对BP神经网络模型进行变量筛选的网络优化和改良,从而形成新的优化后的MIV-BP(Mean Impact Value Back-Propagation Neural Network)神经网络,并以此用于评价房地产项目中的风险度以及各因素在风险度中的影响作用大小;同时选取目前相关的房地产项目数据进行仿真实证分析和验证。验证实验结果表明,MIV-BP型神经网络对于房地产项目风险度识别具有良好的适应性和准确性,实验结果客观,达到专家评价的要求,并在风险因素作用度分析上具有良好的应用价值。     

Real-time recursive estimation of statistical parameters

Recursive algorithms for the computation of standard deviation and average deviation are derived and their applications in data acquisition are discussed. The relative speeds and accuracies of the two algorithms are compared for synthetic data. The performance of recursive estimation under shot and proportional noise limitations is also described. As an example of the utility of these algorithms, absorbance data with constant confidence intervals are collected regardless of incident and transmitted intensities. The desired precision is specified prior to data acquisition and used to control signal-averaging of the data in real time.