UNIT 7 DSP Processor Fundamentals
1
New words and Technical Terms
outperform delay discrete time difference digital signal processing (DSP) development time register transfer level (RTL) general purpose microprocessor (GPP) cycles per instruction (CPI)
2
flexibility operand coefficient
latency filter algorithm
Finite impulse response (IIR) infinite impulse response (FIR) analog to digital converter (ADC) digital to analog converter (DAC)
3
Text
In the literature, the definition of a digital signal processor takes many forms. In a strict sense, a DSP is any microprocessor that processes digitally represented signals. A DSP filter for example, takes one or more discrete inputs, xi[n], and produces one corresponding output, y[n] for n = …, -1,0, 1, 2, …, and i = 1, …, N, where n is the nth input or output at time n, i is the ith coefficient and N is the length of the filter.
4
In effect, the DSP implements the discrete-time system. As its name implies, it is assumed that there must be some form of preprocessing if the signals are in the continuous time domain, and this is easily accomplished by an analog to digital converter (ADC) [1]. [1].正如它名字所暗示的,连续时域信号必须要 正如它名字所暗示的, 正如它名字所暗示的 经过某种形式的预处理才能被它所处理, 经过某种形式的预处理才能被它所处理,当然这 很容易由模数转换器来完成。 很容易由模数转换器来完成。
5
句中its指代 引导的是状语, 句中 指代DSP,as引导的是状语,it is 指代 , 引导的是状语 assumed that there must be some form of preprocessing if the signals are in the continuous time domain中that之后是宾语从 中 之后是宾语从 引导的是条件状语从句。 句,if引导的是条件状语从句。 引导的是条件状语从句 In general, DSP functions are mathematical operations on real-time signals and are repetitive an numerically intensive.
6
Samples from real-time signals can number in the millions and hence a large memory bandwidth is needed. processors are created with an architecture unlike those of conventional microprocessors [2]. [2].正是由于这一特性,在发明DSP处理器时采 正是由于这一特性,在发明 正是由于这一特性 处理器时采 用了一种与传统的微处理器不同的结构体系。 用了一种与传统的微处理器不同的结构体系。
7
It is
because of this very nature that DSP
此句中的It 是强调句型, 此句中的 is…that是强调句型,这种句型可以 是强调句型 强调句子的主语,宾语,状语等, 强调句子的主语,宾语,状语等,被强调的部分 直接放在be动词之后,本句是强调状语部分, 直接放在 动词之后,本句是强调状语部分, 动词之后 即because of this very nature; unlike those ; of conventional microprocessors则是用于说 则是用于说 明architecture的。 的 Most DSP algorithms are not complicated and only require multiply and accumulate calculations.
8
Most, if not all, DSP processors have circuitry built and hard wired to execute these calculations as fast as possible. The signal processing algorithms and functions define a suitable architecture for implementation. We use a simple example of an FIR filter as a basis for the building blocks of the DSP architecture.
9
One algorithm used to create an FIR filter uses a direct form on tapped delay line structure with M+1 taps. The M+1 most recent input samples are saved as “filter states”.
10
According to Equation (
  1),
y(n) = ∑ci x(n i)
i=0
M
(2-
  1)
the products of each filter state x(n-i) and its corresponding coefficient ci are accumulated or added to produce the current output sample y(n). We can also use the signal flow graph as shown in Fig. 2-1 to represent this algorithm [3].
11
[3].我们也可以用图 中的信号流程图来介绍 我们也可以用图2-1中的信号流程图来介绍 我们也可以用图 这个算法。 这个算法。 However it is not clear as to the sequence of the computations since it looks like all the operations can be carried out at the same time. Thus, a more accurate picture has to be formed by using microoperations at the register transfer level (RTL),
12
sequenced temporarily from left to right as seen in Fig. 2-
  2.
x(n)
z-1
z-1
z-1
c0
c1
...
c(M-
  1)
cM y(n)
Fig. 2-
  1. Tapped delay line structure of a FIR filter.
13
x(n) y(n) D1
Input and Output
D1
D1
Date M emory
T1
T1
T1
T1
T1 Temporary M emory
...
c0 c1 c (M)
time
Arithmetic Operations
Coefficient M emory
Fig. 2-
  2. Register transfer level representation of a FIR filter. 14
The delayed inputs are stored in the data memory, D1 and the coefficients, c0, c1, c(M) are located in the coefficient memory. The contents of both memories are fetched and multiplied together. The result is then added to the temporary memory, T1 is where the results of the previous taps are stored. This cycle is repeated with a different coefficient until completion, producing the final result as y(n). 15
We can make certain assumptions for a fundamental general purpose DSP architecture. From our understanding of DSP algorithms, we see that most computations are multiply and add operations. Looking at the example from the previous section, we will require multiple memory units for storage of different data as well as memory for the arithmetic operation sequences.
16
Registers can serve as temporary storage locations and buses will be needed to connect these units together. At this point, the reader may be tempted to ask how this design is different from a general purpose microprocessor (GPP). If we react the issues central to a DSP function, most DSP calculations are repetitive, require a large memory bandwidth and numeric precision, all executed in real time. 17
One might also argue that modern GPPs have clock speeds and cycles per instruction (CPI) that outperform DSP processors but GPPs have operations and program flexibility that are unnecessary for DSP [4]. [4].也许有人认为现代的通用处理器比 也许有人认为现代的通用处理器比DSP有更 也许有人认为现代的通用处理器比 有更 高的处理速度和指令周期, 高的处理速度和指令周期,但是通用处理器有许 不必要的运算和编程的灵活性。 多DSP不必要的运算和编程的灵活性。 不必要的运算和编程的灵活性
18
DSPs must execute their tasks efficiently while keeping cost, power consumption, memory usage and development time low, especially in the age of mobile computing. Since many signal processing applications process millions of samples of data for every second of operation, the minimum sample period is usually more important than the computationa latency of the processor. 19
We define the sample period as the time between each sequential sample of the input data. The time difference between the input data and the result of its computation is known as the computational latency. Once the initial sample is calculated with certain latency, the subsequent results will however, be produced at the sample period rate.
20
As the number of calculations increases, the relatively larger latency of the processor will be negligible compared to the sample rate.
Exercises: Ⅰ,Ⅱ
21
数学公式的表示与读法(Ⅰ) 数学公式的表示与读法(
基本运算符号
A+B = C A-B = C A×B = C × A/B=C A:B
  0.3 % A plus B equals C A minus B equals C A multiplied by B equals C A divided by B equals C The ratio of A to B Zero point three percent x squared
22
A : B = C : D A is to B of A to B as C is to D
X
2
X3 X4
X
1 2
x cubed x to the fourth power x to the one-half power
19
e =
  1.6×10
x
3 n
e equals one point multiplied by ten to minus nineteenth power The square root of x The cube of x The nth root of x A equals B A is identical with B; A is equivalent to B
23
x x
A=B A≡B
A≠B A≈B A>B A<B A≥B A << B
A is not equal to B A be approximately equal to B A is greater than B A is less than B A is greater than or equal to B A is far less than B
24
简单函数
f (x)
Function of x
The function of x equals a f (x) = ax2 + bx + c times the square of x plus b times x plus c |a|=b Max f(x) Min f(x) The absolute value of a equals that of b The maximum value of f(x) The minimum value of f(x) A sub n approaches / tends to infinity. The limit of b
25
an →∞
lim b
1 limSn = 3 n→∞
( A+ B)C
x +A 2 2 (x + B)
5
The limit of Sn as n gets arbitrarily large is one third. The quantity A plus B times C x to the fifth power plus A over (divided by) the quantity x squared plus B, to the two-thirds power Log of x Log of x to the base two
26
log x
log 2 x
 

相关内容

《电子信息工程专业英语导论》课件PART 2 unit7

   UNIT 7 DSP Processor Fundamentals 1 New words and Technical Terms outperform delay discrete time difference digital signal processing (DSP) development time register transfer level (RTL) general purpose microprocessor (GPP) cycles per instruction ...

《电子信息工程专业英语导论》课件PART 2 unit5

   PART 2 Modern Electronics Devices 1 UNIT 5 The basic components of computer system 2 New words and Technical Terms memory processor performance location signal instruction bus semiconductor flip-flop chip transistor binary central processing unit ...

电子信息工程专业英语

   电子信息工程专业英语-教师用书 Part 1 第一课 关于电子技术 一、课文习题参考答案 Ⅰ. (1) alternating current circuits (2) semiconductor diodes (3) passive component(4) the combinatory logic electric circuit (5) rectification(6) Laplace transform (7) inductor(8) Fourier series and Four ...

电子信息工程专业英语

   电子信息工程专业英语-教师用书 Part 1 第一课 关于电子技术 一、课文习题参考答案 Ⅰ. (1) alternating current circuits (2) semiconductor diodes (3) passive component(4) the combinatory logic electric circuit (5) rectification(6) Laplace transform (7) inductor(8) Fourier series and Four ...

电子信息工程专业英语课件Technical English 10

   Technical English For Information and Communication Engineering 2010-11-6 Unit Ten Third Generation Wireless Networks 概述 第一代移动通信系统从80年代中期开始, 第一代移动通信系统从80年代中期开始,以模拟传输和采 年代中期开始 用频分复用(FDMA)这样的简单复用技术为特征。 用频分复用(FDMA)这样的简单复用技术为特征。 第二代系统使用数字技术,出现于90年代初,通过 ...

电子信息工程专业英语课件Technical English 07

   Technical English For Information and Communication Engineering 2010-11-6 Unit Seven Computerized Tomographic Imaging 概述 计算机层析成像的基本原理 Hounsfield的发明及其意义 Hounsfield的发明及其意义 X射线CT和衍射CT 射线CT和衍射 和衍射CT CT技术的应用领域 CT技术的应用领域 2 tomography transmission reflect ...

电子信息工程专业英语翻译清华出版社English for IT and EE-01

   信息科学与电子工程专业英语 Technical English For Information Science and Electronic Engineering 开场白:好的英语是“读”出来的 学习任何语言都要大量阅读。“熟读唐诗三百首,不会作诗 学习任何语言都要大量阅读。 熟读唐诗三百首, 也会吟” 大凡读得多的人,语言表达能力都比较强。 也会吟” 。大凡读得多的人,语言表达能力都比较强。 学英语也要大量阅读。英语好的人,一般说来都读过相当数 学英语也要大量阅读。英语好的人, 量的书。 ...

电子信息工程专业英语翻译清华出版社English for IT and EE-07

   信息科学与电子工程专业英语 Technical English For Information Science and Electronic Engineering 信息科学与电子工程专业英语 Unit 7 Satellite Communications 信息科学与电子工程专业英语 Part I Applications of Communications satellite New Words abbreviate 缩略,简写成 缩略, elliptical orbit 椭圆轨道 sub ...

unit4专业英语翻译-2

   专业英语的翻译-2 词性的转换 英语和汉语在语言结构和表达上有很多 不同的特点,科技英语句子中的名词化 结构较多,而谓语动词则只有一个;相 对而言,汉语的一个句子中往往包含有 几个动词;此外汉语中没有英语中的冠 词、关系代词、关系副词、分词、不定 式、动名词等语法成分,因而翻译时就 需要做适当的词性转换处理。 词性的转换 1、英语名词化结构及动名词译成汉语的 动词。如: CIMS is fairly recent development. 计算机集成制造是近年来才发展起来的。 In the ...

unit7七年级英语第七单元课件

   Unit7 How much are these pants? Grammar(语法): (语法) 1.How much引导的特殊疑问句 much引导的特殊疑问句 2.基数词(0-31) 基数词( 基数词 ) 3.复习 复习What colour询问颜色 复习 询问颜色 课文): Text(课文): How much is this T-shirt?这件 恤衫多少钱? 这件T恤衫多少钱 恤衫多少钱? It's seven dollars. 7美元 . 美元 How much are thes ...

热门内容

英语_个人履历类英文词汇

   个人履历类英文词汇 个人资料: name 姓名 alias 别名 pen name 笔名 date of birth 出生日期 birth date 出生日期 born 出生于 birth place 出生地点 age 年龄 native place 籍贯 province 省 city 市 autonomous region 自治区 prefecture 专区 county 县 nationality 民族,国籍 citizenship 国籍 duel citizenship 双重国籍 a ...

2010年高考英语试卷详解~~江西

   2010-6-10 2010 年普通高等学校招生全国统一考试 江西卷 英语试题 年普通高等学校招生全国统一考试(江西卷 江西卷)英语试题 第一卷(选择题 满分 115 分) 第一部分 听力(共两节,满分 30 分) 做题前先将答题标在试卷上. 录音内容结束后, 你将有两分钟的时间将试卷上的答案转 涂到答题卡上. 第一节(共 5 小题,每小题 1.5 分,满分 7.5 分) 听下面 5 段对话.每段对话后有一个小题,从题中所给的 A,B,C 三个选项中选出最 佳选项,并标在试卷的相应位置.听完 ...

2005年12月英语三级考试真题

   2005 年 12 月英语三级考试真题 Part I Reading Comprehension (www.pkucet.com) (30%) Directions:There are three passages in this part. Each passage is followed by some questions or unfinished statements. For each of them there are four choices marked A, B, C an ...

2008年 深圳中考英语科试题

   2008 年深圳中考英语科试题 I 听力测试(25 分) 一、选择合适的答语: 1.A.Sorry. 2.A.Yes, it does 3.A.Never mind 4.A.He is working. B. Thanks. B. Not, it doesn’t B. It doesn’t matter B. He is a doctor C. Not at all C. Yes, it is C. It’s my pleasure C. He is at home 5.A.Thank yo ...

影响中学生英语听力的因素与解决策略[1].doc

   影响中学生英语听力的因素与解决策略时间:2010-08-24 00:43:13 来源:2010 年 7 期基础 教育 作者:魏爱晶??白城职业技术学院 吉林 白城 137000 [摘要]《中学英语课程标准》中指出,语言技能包括听、说、读、写四种技能以及这四种技 能的综合运用。其中听是学习英语的基础,放在首位。听力如此重要,我们应该采取什么措 施来提高听力水平, 是摆在英语教师面前的一个新课题。 本文将从影响中学生听力的几个因 素入手,提出了提高中学生听力水平的具体方法。 [关键词]英语听力 ...