MOSFET abr,金属氧化物半导体场效应管 PROM abbr.可编程只读存储器 传播 RAM abr.随机存储器 肖特基(半导体器件) 尖脉冲 sceptibility 触发(器) 没管 触发 基极-发射极结 carmrpettive wt 与…不相上下 驱动能力 脉冲持续期 触发器 negative logic 负逻辑 noise immunity 抗干扰性 正逻辑 power consump 移位寄存器 truth table 真值表 1. NANd gate,与非门,下文出现的AND,QR和NOR分别是与门或门或 非门 EXCLUSIVE OR和 EXCLUSIVE NOR分别为异或门和异或非 I both A and B are open or tied to a(+)voltage, a current flows from R through the base-emitter junction of the transistor, which is
hmed“hard-on" or put into the so- called saturation mode.如果A和 B都是开路的或连正电压电流从R1通过三极管的基极发射极结调整 三极管(工作点)接近或进入饱和状态 3. The analog word of full of relative numbers, tradootis, and app mations, all of which depend heavily on the basic semiconductor prop- erties.模拟领城充满了相对数、折衷和近似,所有这些都很大程度地取 决于半导体的基本特性。 4. When digital circuits are operated near their maximum speeds they ap- proach a failre mode that is largely analog in nature and all the trou- bles and uncettainites of the analog cirauit are 当数字电路以接 近它们最大的速度工作时,它们也接近了一个失效的模式这个模式在本 质上很大程度是模拟的,并且所有的模拟电路的麻烦和不确定性又随之 而来 5. The speed limitations of a digital circuit show up in four differe forms: propagation delay, setup time, rise time, and fall time. W 电路速度的限制以四种不同的形式表现出来:传播延迟建立时间、上升 时间和下降时间。 6. When a similar set of digital devices are operating in parallel, their propagation delays, because of the tolerances, is not necessarily the sme.当一组相近的数字器件并行工作时,由于器件的公差传播延迟可 以不必相同。 7. The rise and fall times can be somewhat controlled through good layout to reduce capacitance and inductance, by limiting the number cs that are driven, and by the occasional use of a pull-up resistor in the output circuits.通过优秀的设计,用限制墨动的级数以及偶尔在输 出电路中应用负载电阻的方法来减少电容和电惑并用此可以控制上升 和下降时间。 8.TI(Tm),M0S和ECL分别为晶体管-晶体管亚辑、金属氧化物半导 体逻舞和射极精合逻辑 9. Fairchild为仙童公司本文还涉及一些其它半导体制造公司
1.3 Digital Signal and Discrete-Time Systems 1.3.1 Signals amd Systems 1.3.1.1 Signals Signals are scalar-valued functions of one or more independent vari ables. Often for convenience, when the signals are one-dimensional, the independent variable is referred to as time, The independent variable may be continuous or discrete. Signals that are continuous in both ampli bde and time(often referred to as continuous -time or analog signals)are e most commonly encountered in sigmal processing contexts. Discrete time signals are typically associated with sampling of continuous-time nals. In a digital implementation of a sigmal processing system, quantiza tion of signal amplitude is also required. Although not precisely comet in every context, discrete-time signal proceasing is often referred to as digital Discrete-time signals, also referred to as sequence, are denoted by nctions whose arguments are integers. For example, *(n)represents sequence that is defined for all integer values of n and undefined for non- vau tion x or to the value of the function x at a specifie value of n. The dis tinction between these two will be obvious from the context Some sequences and classes of sequences play a particularly impor tant oe in discrete- time signal processing. These are summarized below The unit sample sequence, denoted by &(n), is defined as 0 8(n)= (1) The sequence 8 (n)plays a role similar to an og system amalysis The unit step sequence, denoted by u(n),is
(3) play a role in discrete-time signal processing similar to the role played by exponential fumctions in continuous-time signal processing. Specifically they are eigenfunctions of discrete-time linear systens and for that reason form the basis for transform analysis techniques, When lal=l,s(n) takes the form of a complex exponential sequence typically expressed in separated by integer multiples of 2x in w(frequency)are identical se- (5) This fact forms the core of many of the important differences between the representation of discrete-time signals and systems and that of continu ous-time signala and systems A general sinusoidal sequence can be expressed es x(n)=Acos(∞n+φ) where A is the amplitude, a the fro andφ the phase.hn rast with contimuous-time sinusoids, a discrete-time sinusoidal signal is not necessarily periodic and if it is, the period is2πlω po anly when2u/∞u is an integer. In both oontinuous time and discrete time, the importance o sinusoidal signals lies in the facts that a broad class of signal can be represented as a linear combination of sinusoidal signals and that the re- apamse of limear time-invariant systems to a sinusoidal signal is sinusoidal with the same frequency and with a change in only the amplitude and 1.3.1.2 Systems In general, a system maps an input signal *(n)to an output signal y(n)through a system transfomation TI F. This definition of a system is very broad. Without some restrictions, the characterization of a system tules
system to certain set of inputs does not allow us to detemine the output of the system to other sets of inputs. Two types of restrictions that greatly time invariance, altematively refered to as shift invariance, aly earity and simplify the characterization and analysis of a system are Ii ly, many systems can often be approximated in practice by a linear and tiame-Invanant system The linearity of a aystem is defined through the principle of superpo Tαx:(n)+b2(n)}=1(n)+b2(n)(7 where Tl*, (n)=yi(n), rl=2(n)l= ya(n),and a and b are any scalar constants Time invariance of a system is defned as x一0 (8) here y(n)=rl=(n)l and no is any integer. Linearity and time in- e, a system me the other property, both or neither For a linear and time-invariant(LTT)system, i) the system response y(n)i廖nby x(h)A(n-k)x(n)操h(n) whene s(n)is the input and A(n)is the response d the system when the input is&(n). Eq (9)is the convolution sum. As with continuous time convolution, the convolution operator in Eq ( 9) is oon mutative and associative and distributes over addition (n)*y(n)=y(n)*x(n) [x(n)*y(n)]*珈(n)=x(n)*[y(n)*(n)](1 x(n)*[y(n)+(n)]=[x(n)*y(n)+x(n)*w(n) (12) In continuous-time systems, convolution is primarily an analyti