Schedulability issues for EDZL scheduling on real-time multiprocessor systems

EDZL (Earliest Deadline first until Zero Laxity) is an efficient and practical scheduling algorithm on multiprocessor systems. It has a comparable number of context switch to EDF (Earliest Deadline First) and its schedulable utilization seems to be higher than that of EDF. Previously, there was a conjecture that the utilization bound of EDZL is 3m/4=0.75m for m processors. In this paper, we disprove this conjecture and show that the utilization bound of EDZL is no greater than m(1−1/e)≈0.6321m, where e≈2.718 is the Euler's number.     

基于Linux的多核实时任务调度算法改进

嵌入式实时系统通常被实现为多任务系统,以满足多个外部输入的响应时间的最后期限约束。Linux内核中已经实现了基于EDF(Earliest Deadline First)调度算法的DL调度器,使得实时任务能在截止期限内运行完成。但对于多核处理器,由于实时任务在EDF算法下会出现Dhall效应,论文对 Linux内核中实时任务调度算法进行了改进。在EDF算法的基础上,实现LLF(Least Laxity First)调度算法并对其加以改进,通过降低任务上下文切换频率以及减少松弛度的计算来减小调度过程中的颠簸现象。实验证明该方法既避免了Dhall效应,又减少了任务上下文切换带来的系统开销,并使得任务能在截止期限内完成调度,取得了较好的调度性能。     

视频服务器存储子系统的I/O优化

视频服务器存储子系统的I／O性能决定了视频服务器的总体性能。本文提出的一种新的实时磁盘调度算法（LLF－Window)能有效地服务各种编码格式的视频流，并对传统的SCSI总线不均衡调度机制进行了改造，实验结果表明，新的磁盘调度算法和改造后的SCSI总线调度机制有效地改进了视频服务器存储子系统的I／O性能，保证了视频流的连续播放。     

Real-Time Traffic Scheduling and Routing in Packet-Switched Networks Using a Least-Laxity-First Strategy

In this paper, we address the problem of transmission scheduling and buffer management at a switch in a high-speed packet-switched network. Future computer networks are expected to carry bursty real-time multimedia traffic with specific time-delay requirements. We propose a technique to schedule competing packets at a switch in a packet-switched network with two objectives—(i) maximizing the number of packets that would reach their destination before their deadline has been reached (ii) minimizing the number of packets dropped in transit in the network. The method is broadly based on the popular least-laxity heuristic in real-time process scheduling and is combined with buffer management at the switch.     

多媒体服务器混合磁盘调度策略

基于多媒体服务器的性能要求，提出了一种自适应的混合磁盘调度策略DRT-window．它既能满足实时请求对实时性的要求，根据实时请求的截止期动态选择窗口大小；又能在其松弛度内尽努力(best-effort)地服务非实时请求，从而减少非实时请求的响应时间。DRT-window采用了两级层次调度方案：第一层为不同类型的请求采用各自适合的调度策略；第二层为混合请求调度嚣，混合调度第一层中的不同类型的请求。通过性能比较和理论证明，表明此混合磁盘调度策略能在保证实时请求无抖动执行的同时，尽量地减少非实时请求的响应时间。     

Bayesian EWMA control chart with measurement error under different loss functions

This paper presents the Bayesian EWMA control chart under two different loss functions (i) squared error loss function (SELF) and (ii) linex loss function (LLF) in the presence of measurement error (ME). We take posterior and posterior predictive distribution under the conjugate prior. We used a linear covariate model in the existence of ME to evaluate the control chart. We also studied the effects of multiple measurements and linear increasing variance methods in the existence of the ME. The average run length and standard deviation of run length are used to measure the performance of the Bayesian EWMA control chart with ME. We conducted the Monte Carlo simulation study to evaluate the performance of Bayesian EWMA control chart with ME. A real-life data example is also presented to demonstrate the application of the control chart.     

基于LLF和RBD检测的红外和可见光图像融合

提出一种基于LLF和RBD检测的红外和可见光图像融合方法。运用局部拉普拉斯滤波对红外图像平滑处理和对可见光增强处理,以充分利用红外图像的目标信息和可见光图像的细节信息。在此基础上,采用增强背景检测的RBD显著性检测算法处理红外图像,以很好地检测出目标。此外,为了增强目标信息,减弱背景干扰,对RBD检测的结果进行S曲线变换。然后,对红外和可见光图像应用NSST分解得到高频分量与低频分量。最后,使用S曲线变换后获得的显著图对低频分量进行加权融合,采用绝对值取大的规则对高频分量进行融合。实验结果表明,该方法能够得到红外目标突出,细节增强的融合图像。     

基于CAN总线的调度算法的仿真性能评价

实时调度算法是实时系统中的关键技术，本文主要对常用的动态优先级调度算法EDF和LLF进行了深入分析，进一步在基于CAN总线协议的环境中建立这两种算法的仿真模型，并对这两种算法应用于CAN总线系统的效果进行了仿真分析。     

OFDM多载波系统中子载波调制识别新方法

针对OFDM （Orthogonal Frequency Division Multiplexing）信号在非合作通信系统中,具有多种子载波调制类型且有些子载波调制类型难以被调制识别的问题,提出了一种对其子载波分类识别的新方法.此方法结合并改进了星座图聚类投影法和对数似然函数（Logarithmic Likelihood Function,LLF）算法,先对不同子载波调制信号进行星座图聚类投影从而识别出常规子载波调制类型,再进一步通过对数似然函数对常规子载波调制类型和偏移正交（Offset QAM,OQAM）调制类型进行分类识别,并在此基础上推导出子载波组的对数似然函数使其计算结果值更容易被判决门限分类.理论推导和计算机仿真结果表明这种方法能在信噪比高于15 dB的情况下完全识别子载波的调制方式.