Price-based power control algorithm in cognitive radio networks based on monotone optimization

This paper considers a price-based power control problem in the cognitive radio networks(CRNs).The primary user(PU) can admit secondary users(SUs) to access if their interference powers are all under the interference power constraint. In order to access the spectrum, the SUs need to pay for their interference power.The PU first decides the price for each SU to maximize its revenue. Then, each SU controls its transmit power to maximize its revenue based on a non-cooperative game. The interaction between the PU and the SUs is modeled as a Stackelberg game. Using the backward induction, a revenue function of the PU is expressed as a non-convex function of the transmit power of the SUs. To find the optimal price for the PU, we rewrite the revenue maximization problem of the PU as a monotone optimization by variable substitution. Based on the monotone optimization, a novel price-based power control algorithm is proposed. Simulation results show the convergence and the effectiveness of the proposed algorithm compared to the non-uniform pricing algorithm.     

A Distributed Power Allocation Scheme in Green Cognitive Radio Ad Hoc Networks

For the realization of green communications in cognitive radio ad hoc networks(CRAHNs), selfadaptive and efficient power allocation for secondary users(SUs) is essential. With the distributed and timevarying network topology, it needs to consider how to optimize the throughput and power consuming, avoid the interference to primary users(PUs) and other SUs, and pay attention to the convergence and fairness of the algorithm. In this study, this problem is modeled as a constraint optimization problem. Each SU would adjust its power and corresponding strategy with the goal of maximizing its throughput. By studying the interactions between SUs in power allocation and strategy selection, we introduce best-response dynamics game theory and prove the existence of Nash equilibrium(NE) point for performance analysis. We further design a fully distributed algorithm to make the SUs formulate their strategy based on their utility functions, the strategy and number of neighbors in local area. Compared with the water-filling(WF) algorithm, the proposed scheme can significantly increase convergent speed and average throughput, and decrease the power consuming of SUs.     

Reputation-based collaborative spectrum sensing scheme in cognitive radio networks

Collaborative spectrum sensing is proposed to improve the detection performance in cognitive radio (CR) networks. However, most of the current collaborative sensing schemes are vulnerable to the interference of the malicious secondary users (SUs). In this paper we propose a reputation-based collaborative spectrum sensing scheme to improve the security of cooperative sensing by mitigating the impacts of misbehaviors. The fusion center calculates the reputation rating of each SU according to their history reports to weight their sensing results in the proposed scheme. We analyze and evaluate the performance of the proposed scheme and its advantages over previous schemes in expansibility and integrity. Simulation results show that the proposed scheme can minimize the harmful influence from malicious SUs.     

区域公共停车设施选址优化方法分析

为解决城市停车供需矛盾问题,使停车设施分布相对合理,提出最大限度地满足停车设施使用者、投资者、管理者三方效益最优的停车选址优化模型。建立综合考虑使用者的出行效益、投资者投资利润率与管理者区域服务水平的多目标约束模型,采用交叉概率和变异概率的自适应调整公式来优化基本遗传算法并求解该模型,给出模型求解过程。最后,经实例分析,得出规划区域内最优公共停车场建设个数、各停车场的具体位置以及泊位数,验证模型的有效性。     

A Bandit Method Using Probabilistic Matrix Factorization in Recommendation

In recommendation system,sparse data and cold-start user have always been a challenging problem.Using a linear upper confidence bound(UCB) bandit approach as the item selection strategy based on the user historical ratings and user-item context,we model the recommendation problem as a multi-arm bandit(MAB)problem in this paper.Enabling the engine to recommend while it learns,we adopt probabilistic matrix factorization(PMF) in this strategy learning phase after observing the payoff.In particular,we propose a new approach to get the upper bound statistics out of latent feature matrix.In the experiment,we use two public datasets(Netfilx and MovieLens) to evaluate our proposed model.The model shows good results especially on cold-start users.     

基于数据驱动的物流电动汽车充电行为分析

为研究城市配送中物流电动汽车用户的充电行为规律,本文采集70辆物流电动车2014年一年的充放电数据,并采用数据挖掘相关分析方法,建立考虑物流电动汽车的充电电量状态及充电时刻的充电行为模型.研究结果表明:用户一般在SOC为30%~50%时为车辆进行充电,车辆开始充电时的剩余电量服从μ=0.48、σ=0.22的正态分布;车辆开始充电时刻主要集中在14:00-16:00之间.通过数据实验证明本文所建立的充电行为模型具有较高的精确性,同时具有较好的实用性,为车辆的充电调度和用户的出行安排提供科学的决策支持.     

多面函数法在GPS高程拟合中的应用

为了改善GPS大地高向正常高转换的精度,在局部区域内,建立多面函数模型进行高程拟合,可以达到较高的精度。文中利用多面函数模型进行高程拟合,除选取分布均匀的GPS水准联测点外,还对核函数形式的选取做了详细地分析,并与高程拟合中常用的二次曲面拟合法做了对比,进行了精度分析。     

基于AFC数据和RF模型的城轨车站服务功能分类

城轨线网上服务不同类型客流车站的分布能反映城市结构特征,本文基于多年的城轨AFC数据,应用随机森林模型(RF),分析了全网车站服务客流类型的时空演变.对于传统RF在选择训练集时过于依靠主观经验的问题,本文首先利用传统随机森林度量车站之间的相似性,再采用Partitioning Around Medoid(PAM)方法对度量结果进行聚类,结果表明无监督随机森林方法拥有更好的准确性;最后采用无监督随机森林对北京市2014-2017年的车站服务客流属性的时空变化情况进行分析,结果显示,北京市近4年来城市职住结构在大的空间尺度上基本保持不变,但城市内部各功能区正经历缓慢变化,而服务于居住与工作混合类客流的车站所处区域将是今后城市结构演变中热点区域.本文的研究结果可以为城市轨道交通与城市结构互动关系的认识提供新视角.     

Continuous multiplicative attribute graph model

Network modeling is an important approach in many fields in analyzing complex systems. Recently new series of methods have emerged, by using Kronecker product and similar tools to model real systems. One of such approaches is the multiplicative attribute graph (MAG) model, which generates networks based on category attributes of nodes. In this paper we try to extend this model into a continuous one, give an overview of its properties, and discuss some special cases related to real-world networks, as well as the influence of attribute distribution and affinity function respectively.     

出租车超速行为道路因素随机系数空间模型构建

为准确揭示道路因素与路段出租车超速行为特征的关联关系，本文以成都市中心城区主 干道路段为研究单元，通过车载GPS设备获取大范围出租车超速行为样本。在筛选出10类典型城市道路特征的基础上，选择路段超速线密度和不同百分位数超速严重度作为研究对象；为减少空间效应对估计的干扰，本文验证了超速特征的空间相关性，并分别构建基于对数-高斯分布的标准线性模型和空间条件自回归模型、空间误差模型及空间滞后模型这3类空间模型，探究不同空间模型与出租车超速行为特征的拟合情况；同时，考虑道路组间异质性，进一步构建随机系数空间模型。结果表明：出租车路段超速线密度和超速严重度均存在显著空间自相关性，空间模型 对其拟合效果普遍优于传统模型；不同超速行为特征适用不同的空间模型，随机系数空间条件自回归模型对超速线密度拟合效果最优，而不同百分位数超速严重度适用的最佳拟合模型差异较大；路段限速、一块板横断面及非机动车车道这3类因素表现出对出租车超速行为特征的组间影响异质性；因素解释方面，路段限速、一块板横断面、非机动车车道、路段施工区与超速线密度显著相关；路段限速、非机动车车道、上下坡匝道、路段施工区及路段长度与超速严重度显著相关。     

基于生存时间的映射缓存建模与仿真分析

映射缓存在身份与位置分离网络中用于暂时存储通信对端的身份标识与位置标识的映射关系,以减少映射查询时延,提高网络性能.本文对基于生存时间（Time-To-Live,TTL）的映射缓存的命中概率进行了数学建模.该解析模型得到的缓存命中概率,可以作为数据包到达间隔分布的函数进行计算.通过仿真分析对数据包到达间隔的分布进行数据拟合,得到其累积概率分布,并验证了所建模型的正确性.进一步仿真评估了映射缓存的性能.     

共享单车出行空间异质性特征及驱动因素研究

为实现基于轨迹数据挖掘的共享单车出行空间异质性特征及其驱动因素评估，本文应用核密度分析和热点探测，获取采样分析区域并以热力值表征共享单车出行发生量，减少尺度效应的干扰；引入空间统计学的半变异函数模拟共享单车出行发生量的结构性和随机性变化规律，挖掘空间异质性特征，确定邻域效应的尺度范围；利用空间序列的斜率表征变化趋势，同时，结合改进的空间滞后和残差模型，区分土地利用、邻域效应和其他建成环境各自对共享单车出行空间异质性特征的驱动力。以北京市为案例进行分析，结果表明：北京市的共享单车出行存在中等的、 正的空间自相关性，空间异质性特征的最佳拟合模型为指数模型；空间自相关性的衰减半径为 1860 m，大于此距离时邻域效应消失；建成环境对空间异质性特征的相对驱动力最大，邻域效应对其的相对驱动力则处于中间水平，而土地利用对其的相对驱动力最小。     

基于无碰撞区跳频序列的两级FFH/MFSK系统

为提高系统容量,提出了一种基于无碰撞区(NHZ)跳频序列的两级FFH/MFSK系统(称为NHZ/FFH系统).系统中全体用户被分成若干组,每组分配一个互不相同的NHZ跳频序列.由于NHZ跳频序列的汉明互相关在NHZ内等于0,只要用户间相对延迟不超过无碰撞区,就可以消除组间用户的干扰,多址干扰只来源于同一组内的用户.通过采用适于异步系统的多用户检测器,可以进一步提高系统的检测性能,且多用户检测器的规模和复杂度只与组内用户数有关.结合系统模型进行了干扰分析和仿真研究,结果表明,与现有MS/FFH系统相比,在相同的信道条件下,对于相同的误码率级别,NHZ/FFH系统支持的用户数多.     

基于时间阈值的旅客登机模型及动态登机策略

减少登机时间有利于提高旅客体验和航空公司运行效率，故优化基于旅客行为的登机策略尤为重要. 基于旅客登机过程中两个阶段的延时，建立考虑时间阈值的新登机模型；在干扰时间阈值分布的基础上，通过优化算法分配行李，从而使登机时间最小，进一步得到基于不同行李构成的动态旅客登机策略. 通过元胞自动机仿真和现场模拟实验比较动态登机策略的效率. 结果显示：新登机模型能准确反映登机过程，Steffen 次序具有最大的干扰时间阈值和优化空间；本文提出的动态登机策略可以缩短登机时间，使登机效率显著提高.     

基于加权核密度估计的自适应运动前景检测方法

为解决监控视频背景初始化过程中前景干扰的问题,提出了一种基于加权核密度估计(KDE)的自适应运动前景检测方法.该方法对时间域变化稳定的像素值进行加权,并利用核密度估计构建背景模型,避免了背景初始化过程中前景的干扰.基于该背景模型,提出了一种新的阈值设定策略.该策略根据前景空间分布的连续性自适应获得前景阈值,填充前景中的"孔",并更新阈值.实验结果表明:即使场景中存在运动前景,该方法能够在多种场景下获得90%以上的查准率和查全率,其性能优于传统的背景差法.      

Joint user and antenna selection for multiuser MIMO downlink with block diagonalization

User selection is necessary for multiuser multiple-input multiple-output (MIMO) downlink systems with block diagonalization (BD) due to the limited free spatial transmit dimensions. The pure user selection algorithms can be improved by performing receive antenna selection (RAS) to increase sum rate. In this paper, a joint user and antenna selection algorithm, which performs user selection for sum rate maximization in the first stage and then performs antenna selection in the second stage, is proposed. The antenna selection process alternately drops one antenna with the poorest channel quality based on maximum determinant ranking (MDR) from the users selected during the first stage and activates one antenna with the maximum norm of projected channel from the remaining users. Simulation results show that the proposed algorithm significantly outperforms the algorithm only performing user selection as well as the algorithm combining user selection with MDR receive antenna selection in terms of sum rate.     

左截断的Turnbull估计与相应右截断的Kaplan-Meier估计的关系

本文推导只有左截断的Turnbull估计与某种右截断机制下的KaplanMeier估计间的关系。     

连续刚构0#块局部应力分析

连续刚构0#块处于墩梁固结处,该梁段构造复杂,内有大量的预应力束通过。为揭示其在各种不利工况下的空间应力分布情况,需建立相应的空间有限元模型进行分析。文章以一座连续刚构桥为工程背景,运用ANSYS有限元软件对0#块进行局部应力分析,结果表明0#块局部位置应力分布较复杂,应进行钢筋补强。     

有限理性用户风险态度对系统效率的影响

为研究有限理性用户风险态度对随机退化路网均衡和系统效率的影响,构建了考虑路网能力退化的连续风险态度双参考点用户均衡(CPV_UE)模型和系统效率分析模型,证明了解的存在性,并设计了嵌套求解算法.模型从外部客观随机性(随机能力退化路网)和个体主观随机性(双参考点累积前景出行效用)两方面对交通网络均衡状态进行定义、划分与分析. 最后以Nguyen & Dupuis路网为例,分析了连续风险态度对网络均衡和系统效率的影响.研究验证了保守用户在网络均衡时感受到的价值更低的特点,也发现保守用户群体在均衡时流量分布更分散.同时发现,虽然CPV_UE比系统最优均衡行程时间期望大,但其标准差较小,行程时间可靠性更高.     

城市交通状态的空间依赖性和异质性分析

有关城市交通状态空间特征的研究相对匮乏。从空间角度分析城市宏观交通特征,首先,针对城市路网几何形态和交通数据的关系提出适合于城市交通数据分析的空间模型。然后,以从长期的浮动车数据中提取的路段行程速度作为描述交通状态的实验数据,分别采用空间自相关和核密度估计法定量分析路网行程速度的空间依赖性和空间异质性,并分析其在典型时段的空间分布特征。结果显示:城市交通状态的关联性在空间上随着路段间距的增长而衰减;城市交通状态的空间分布特征随时间变化,受路网结构和区域功能的影响在空间中呈现集聚特征。