采用流水化伪随机编码算法的相变存储器寿命延长方法

相变存储器(phase change memory, PCM)是一种颇具前景的新型存储器件,具有非易失性、静态功耗低和存储密度高的优点.然而,该类器件的低写入寿命是其在实用化中亟待克服的关键问题之一.一般来说,通过每次写入时仅写入相异位的策略,可以减少产生的平均写入量,从而延长PCM的写入寿命.然而,应用这一差异式的写入策略通常又会以降低读写速度为代价.为此,提出了一种兼具高效和快速特点的写入量减少方法FEBRE(a fast and efficient bit-flipping reduction technique to extend PCM lifetime).该方法在差分写入阶段前,设计并使用了一种快速的一对多映射,将待写入的数据并行映射为多个编码向量,从而增加了从其中找到一个与已有数据最近的向量的可能性.此外,还提出了一种流水化的伪随机编码算法,用以加速一对多映射中的编码过程,从而降低写入开销.实验表明,与目前领先的PRES(pseudo-random encoding scheme)方法相比,FEBRE方法在写入操作中,平均减少了5%以上的写入量,提升了2倍以上的编码速度;在读取操作中,减少了45%以上的解码操作次数.     

提高定点精度的FPGA信号处理算法

为了满足速度、功耗等诸多限制的要求,数字信号处理算法常使用FPGA实现。而实现时由于硬件特点,通常将浮点运算转换成定点运算,但定点转换设计流程复杂、周期长,且存在数据范围和精度之间的矛盾。利用浮点数的优点,本文改进了基于FPGA的定点数的基本运算规则,有效解决了上述矛盾。本文详细论述了实现移位、加/减、乘、除基本运算模块的方法和步骤,最后以FIR数字滤波器为设计实例。仿真结果表明:改进的定点数算法比定点运算误差小、精度高、数据范围宽,能有效地防止溢出。     

Ink-Jet Cartridge Lifetime

Extensive examination of thermal ink-jet cartridges has revealed that their maximum lifetime is limited by two major mechanisms: The first is the deterioration of the heaters caused by the ink-jet propelling bubble combined with the high temperature of operation of the heaters. The second is deterioration caused by direct current leakage from the heater through the ink chamber creating unwanted electrolytic byproducts. To eliminate both of these effects, it is proposed to eliminate the electrical heaters used to generate the propelling bubble and replacing them by electrodes passing a high-frequency current through the printing ink to form a propelling bubble. To evaluate this system, it is necessary to experimentally determine the electrical characteristics of the ink to determine the efficiency of heating by the applied current, in the range from 100 KHz to 200 MHz. It was found that standard ink-jet inks, now in use, can be efficiently coupled to the drive transistors within this range without altering ink composition     

Lifetime analysis for attributes

Summary Reduction of attribute storage is a vital requirement for attribute evaluators. We present a new method for the analysis of lifetime of attribute instances. It is assumed that attribute evaluation is performed by a visit-oriented evaluator. Its evaluation sequence for any input can be described by a contex-free grammar derived from the visit-sequences. Conditions on that CFG decide for each attribute whether all its instances can be stored in a single global variable. Furthermore one can decide whether several different attributes can be mapped to a single global variable. Similarly conditions for stack implementation are given. All decisions can be made efficiently at evaluator generation time. Hence the method is well suited for compiler generation.     

无线传感器网络生存期优化体系研究

提出了系统设计全生命周期的网络生存期优化体系模型,该模型利用先验统计信息将不同的应用与优化算法统一起来,并以评估结果作为反馈调节网络使之达到网络生存期最大限度的优化;构建了一种优化体系的具体评估模型,利用网络能耗分布量化具体应用服务需求,使用网络能量分布度量网络服务能力,并通过量化结果计算网络能量浪费的上限;给出了一种网络密度调节方法,用以发现有利于生存期优化的网络密度分布规律,并对该方法的收敛性以及解的有效性进行了理论证明.模拟实验结果表明该评估模型和密度调节方法可以有效控制网络能耗分布,发现有利于延长网络生存期的密度规律,从而验证了网络生存期优化体系的有效性.     

Increasing interpretations

The article at hand introduces a refinement of interpretation-based termination criteria for term rewrite systems in the dependency pair setting. Traditional methods share the property that—in order to be successful—all rewrite rules must (weakly) decrease with respect to some measure. One novelty of our approach is that we allow some rules to increase the interpreted value. These rules are found by simultaneously searching for adequate polynomial interpretations while considering the information of the dependency graph. We prove that our method extends the termination proving power of linear interpretations. Furthermore, this generalization perfectly fits the dependency pair framework which is implemented in virtually every termination prover dealing with term rewrite systems. We present two dependency pair processors for increasing interpretations. The novelty of the second one is that it can be used to eliminate single edges from the dependency graph.     

DSP＋FPGA系统中FPGA的动态重构设计

通过DSP外部高速EMIF总线,在Slave Serial（从串）配置模式下实现了FPGA芯片的在线动态重构,从而实现了在DSP＋FPGA系统中在不同情况下调入不同逻辑完成不同算法,使系统具有很好的灵活性和可扩充性。     

延长EEPROM使用寿命的方法

EEPROM (Electrically Erasable Programmable Read Only Memory )即电子擦除式只读存储器,它是一种非挥发性存储器,与擦除式只读存储器(EPROM)类似,电源消失后,储存的数据依然存在,要消除储存在其中的内容,不是用紫外线照射方式,而是以电子信号直接消除即可。正是由于EEPROM具有以上特点,该器件可广泛应用于对数据存储安全性及可靠性要求高的应用场合,如门禁考勤系统,测量和医疗仪表,非接触式智能卡,税控收款机,预付费电度表或复费率电度表、水表、煤气表以及家电遥控器等应用场合。该类型存储器在可靠数据存储领域会获得越来越…     

An MPGA-like FPGA

The CP20K series of field-programmable gate array (FPGA) chips is described. Consisting of an array of individual CMOS transistors and small logic elements linked via antifuse-programmable interconnections to form large functional blocks, the CP20K closely mimics the flexibility and capacity of a mask-programmable gate array (MPGA). It is shown that the chip's fine-grained architecture makes it easy to use logic synthesis tools because area and performance can be traded off exactly as in MPGA logic synthesis     

FPGA软件设计

介绍VHDL语言的特点及其在可编程逻辑器件设计中的广泛应用,同时讲解其设计的过程。     

Lifetime reliability: toward an architectural solution

Developing and maintaining industrywide standards for lifetime reliability is a critical task for all microprocessor manufacturers. Although technology scaling continues to provide significant performance benefits, increasingly smaller feature sizes and increasing power densities are accelerating the onset of wearout-based failures, thus shortening processor life. Microarchitects have traditionally treated processor lifetime reliability as a manufacturing problem, best left to device and process engineers. In current processors, manufacturers enforce lifetime reliability, or qualify it, during device design, circuit layout, manufacture, and chip test. This reliability qualification, which is application-oblivious, is based on estimates of worst case temperature and processor utilization. However, most applications will run at lower temperature and utilization, resulting in higher reliability and longer processor lifetimes than required. As a result, current reliability qualification methodologies are overly conservative, unnecessarily increasing cost or decreasing performance. Sustaining this approach will likely be infeasible in future scaled systems.     

无线传感器网络寿命预测模型

通过对网络能耗模型的分析,综合网络能耗方程和终止条件,建立了一种定量预测传感器网络寿命模型.建立了单个节点稳态能耗速率时间序列ARMA模型,通过最小二乘法预测能耗趋势,提出了应用于实际系统的寿命预测计算方法.利用该方法,对应用LEACH协议和HEED协议的传感器网络进行了寿命预测仿真.实验结果表明,模型和计算方法达到了预期的寿命预测效果.     

Customer Lifetime Value Models for Decision Support

We present and discuss the important business problem of estimating the effect of marketing activities on the Lifetime Value of a customer in the Telecommunications industry. We discuss the components of this problem, in particular customer value and length of service (or tenure) modeling, and present a novel segment-based approach, motivated by the segment-level view marketing analysts usually employ. We describe in detail how we build on this approach to estimate the effects of retention campaigns on Lifetime Value, and also discuss its application in other situations. Our solution has been successfully implemented by the Business Insight (BI) Professional Services.     

线形异构传感器的网络生命周期

无线传感器网络的能耗决定了网络的生命周期,如何有效部署传感器节点来延长网络的生命周期是一个重要的研究课题。针对由高级节点和普通节点组成的线形异构传感器网络,给出了最大化网络生命周期模型。通过分析节点的能量消耗,求解出了两种节点的分配比例,得出了最大化网络生命周期的节点部署方案。     

Toward auto-scheduling compilers

In this paper we propose a general framework for compiling, scheduling, and executing parallel programs on parallel computers. We discuss important aspects of program partitioning, scheduling, and execution, and consider possible realistic alternatives for each issue. Subsequently we propose a possible implementation of an auto-scheduling compiler and give simple examples to illustrate the principles. Our approach to the entire problem is to utilize program information available to the compiler while, at the same time, allowing for run-time corrections and flexibility.This work was supported in part by the National Science Foundation under Grant NSF MIP-8410110, the U.S. Department of Energy under Grant DE-FG02-85ER25001, an IBM donation and a grant from AT&T.     

Toward object-based heuristics

Recovering the 3-D shape of an object from its 2-D image contour is an important problem in computer vision. In this correspondence, the author motivates and develops two object-based heuristics. The structured nature of objects is the motivation for the nonaccidental alignment criterion: parallel coordinate axes within the object's bounding contour correspond to object-centered coordinate axes. The regularity and symmetry inherent in many man-made objects is the motivation for the orthogonal basis constraint. An oblique set of coordinate axes in the image is presumed to be the projection of an orthogonal set of 3-D coordinate axes in the scene. These object-based heuristics are used to recover shape in both real and synthetic images     

Toward sociable robots

This paper explores the topic of social robots—the class of robots that people anthropomorphize in order to interact with them. From the diverse and growing number of applications for such robots, a few distinct modes of interaction are beginning to emerge. We distinguish four such classes: socially evocative, social interface, socially receptive, and sociable. For the remainder of the paper, we explore a few key features of sociable robots that distinguish them from the others. We use the vocal turn-taking behavior of our robot, Kismet, as a case study to highlight these points.     

Toward Language-dependent Applications

Is adaptation of English NLP applications the right way to gomultilingual? Should one prefer ``language-independent'' systems with aview to applying them to a large number of different languages? Experience from the processing of Portuguese in several differentareas (part-of-speech tagging, corpus tools, lexical decomposition,machine translation, etc.) suggests that neither of these offers a satisfactory solution. This paper argues for a thorough study of the way individual languageswork in order to develop applications suited for the language inquestion, i.e., ``language-dependent'' systems.