英文原文

The simulation and the realization of the digital filter

With the information age and the advent of the digital world, digital signal processing has become one of today's most important disciplines and door technology. Digital signal processing in communications, voice, images, automatic control, radar, military, aerospace, medical and household appliances, and many other fields widely applied. In the digital signal processing applications, the digital filter is important and has been widely applied.

1、 figures Unit on :

Analog and digital filters

In signal processing, the function of a filter is to remove unwanted parts of the signal, such as random noise, or to extract useful parts of the signal, such as the components lying within a certain frequency range.

There are two main kinds of filter, analog and digital. They are quite different in their physical makeup and in how they work. An analog filter uses analog electronic circuits made up from components such as resistors, capacitors and op amps to produce the required filtering effect. Such filter circuits are widely used in such applications as noise reduction, video signal enhancement, graphic equalisers in hi-fi systems, and many other areas. There are well-established standard techniques for designing an analog filter circuit for a given requirement. At all stages, the signal being filtered is an electrical voltage or current which is the direct analogue of the physical quantity (e.g. a sound or video signal or transducer output) involved. A digital filter uses a digital processor to perform numerical calculations on sampled values of the signal. The processor may be a general-purpose computer such as a PC, or a specialised DSP (Digital Signal Processor) chip. The analog input signal must first be sampled and digitised using an ADC (analog to digital converter). The resulting binary numbers, representing successive sampled values of the input signal, are transferred to the processor, which carries out numerical calculations on them. These calculations typically involve multiplying the input values by constants and adding the products together. If necessary, the results of these calculations, which now represent sampled values of the filtered signal, are output through a DAC (digital to analog converter) to convert the signal back to analog form.

Note that in a digital filter, the signal is represented by a sequence of numbers, rather than a voltage or current.

Unit refers to the input signals used to filter hardware or software. If the filter input, output signals are separated, they are bound to respond to the impact of the Unit is separated, such as digital filters filter definition. Digital filter function, which was to import sequences X transformation into export operations through a series Y.

According to figures filter function 24-hour live response characteristics, digital filters can be divided into two, namely, unlimited long live long live the corresponding IIR filter and the limited response to FIR filters. IIR filters have the advantage of the digital filter design can use simulation results, and simulation filter design of a large number of tables may facilitate simple. It is the shortcomings of the nonlinear phase; Linear phase if required, will use the entire network phase-correction. Image processing and transmission of data collection is required with linear phase filters identity. And FIR linear phase digital filter to achieve, but an arbitrary margin characteristics. Impact from the digital filter response of the units can be divided into two broad categories : the impact of the limited response (FIR) filters, and unlimited number of shocks to (IIR) digital filters.

FIR filters can be strictly linear phase, but because the system FIR filter function extremity fixed at the original point, it can only use the higher number of bands to achieve their high selectivity for the same filter design indicators FIR filter called band than a few high-IIR 5-10 times, the cost is higher, Signal delay is also larger. But if the same linear phase, IIR filters must be network-wide calibration phase, the same section also increase the number of filters and network complexity. FIR filters can be used to achieve non-Digui way, not in a limited precision of a shock, and into the homes and quantitative factors of uncertainty arising from the impact of errors than IIR filter small number, and FIR filter can be used FFT algorithms, the computational speed. But unlike IIR filter can filter through the simulation results, there is no ready-made formula FIR filter must use computer-aided design software (such as MATLAB) to calculate. So, a broader application of FIR filters, and IIR filters are not very strict requirements on occasions.

2、 MATLAB introduced

MATLAB is a matrix laboratory (Matrix Laboratory) is intended. In addition to an excellent value calculation capability, it also provides professional symbols terms, word processing, visualization modeling, simulation and real-time control functions. MATLAB as the world's top mathematical software applications, with a strong engineering computing, algorithms research, engineering drawings, applications development, data analysis and dynamic simulation, and other functions, in aerospace, mechanical manufacturing and construction fields playing an increasingly important role. And the C language function rich, the use of flexibility, high-efficiency goals procedures. High language both advantages as well as low level language features. Therefore, C language is the most widely used programming language. Although MATLAB is a complete, fully functional programming environment, but in some cases, data and procedures with the external environment of the world is very necessary and useful. Filter design using Matlab, could be adjusted with the design requirements and filter characteristics of the parameters, visual simple, greatly reducing the workload for the filter design optimization.

In the electricity system protection and secondary computer control, many signal processing and analysis are based on are certain types Yeroskipou and the second harmonics of the system voltage and current signals (especially at D process), are mixed with a variety of complex components, the filter has been installed power system during the critical components. Current computer protection and the introduction of two digital signal processing software main filter. Digital filter design using traditional cumbersome formula, the need to change the parameters after recalculation, especially in high filters, filter design workload. Uses MATLAB signal processing boxes can achieve rapid and effective digital filter design and simulation.

MATLAB is the basic unit of data matrix, with its directives Biaodashi mathematics, engineering, commonly used form is very similar, it is used to solve a problem than in MATLAB C, Fortran and other languages End precision much the same thing. The popular MATLAB 5.3/Simulink3.0 including hundreds of internal function with the main pack and 30 types of tool kits (Toolbox). kits can be divided into functional tool kits and disciplines toolkit. MATLAB tool kit used to expand the functional symbols terms, visualization simulation modelling, word processing and real-time control functions. professional disciplines toolkit is a stronger tool kits, tool kits control, signal processing tool kit, tool kits, etc. belonging to such communications

MATLAB users to open widely welcomed. In addition to the internal function, all the packages MATLAB tool kits are readable document and the document could be amended, modified or users through Yuanchengxu the construction of new procedures to prepare themselves for kits.

3、 Digital filter design

Digital filter design of the basic requirements

Digital filter design must go through three steps :

(1) Identification of indicators : In the design of a filter, there must be some indicators. These indicators should be determined on the basis of the application. In many practical applications, digital filters are often used to achieve the frequency operation. Therefore, indicators in the form of general jurisdiction given frequency range and phase response. Margins key indicators given in two ways. The first is absolute indicators. It provides a function to respond to the demands of the general application of FIR filter design. The second indicator is the relative indicators. Its value in the form of answers to decibels. In engineering practice, the most popular of such indicators. For phase response indicators forms, usually in the hope that the system with a linear phase frequency bands human. Using linear phase filter design with the following response to the indicators strengths：①it only contains a few algorithms, no plural operations；②there is delay distortion, only a fixed amount of delay; ③the filter length N (number of bands for N-1), the volume calculation for N/2 magnitude.

(2) Model approach : Once identified indicators can use a previous study of the basic principles and relationships, a filter model to be closer to the target system.

(3) Achieved : the results of the above two filters, usually by differential equations, system function or pulse response to describe. According to this description of hardware or software used to achieve it.

4、 Introduced FPGA

Programmable logic device is a generic logic can use a variety of chips, which is to achieve

ASIC ASIC (Application Specific Integrated Circuit) semi-customized device, Its emergence and development of electronic systems designers use CAD tools to design their own laboratory in the ASIC device. Especially FPGA (Field Programmable Gate ) generated and development, as a microprocessor, memory, the figures for electronic system design and set a new industry standard (that is based on standard product sales catalogue in the market to buy). Is a digital system for microprocessors, memories, FPGA or three standard building blocks constitute their integration direction.

Digital circuit design using FPGA devices, can not only simplify the design process and can reduce the size and cost of the entire system, increasing system reliability. They do not need to spend the traditional sense a lot of time and effort required to create integrated circuits, to avoid the investment risk and become the fastest-growing industries of electronic devices group. Digital circuit design system FPGA devices using the following main advantages

(1) Design flexible

Use FPGA devices may not in the standard series device logic functional limitations. And changes in system design and the use of logic in any one stage of the process, and only through the use of re-programming the FPGA device can be completed, the system design provides for great flexibility.

(2) Increased functional density

Functional density in a given space refers to the number of functional integration logic. Programmable logic chip components doors several high, a FPGA can replace several films, film scores or even hundreds of small-scale digital IC chip illustrated in the film. FPGA devices using the chip to use digital systems in small numbers, thus reducing the number of chips used to reduce the number of printed size and printed, and will ultimately lead to a reduction in the overall size of the system.

(3) Improve reliability

Printing plates and reduce the number of chips, not only can reduce system size, but it greatly enhanced system reliability. A higher degree of integration than systems in many low-standard integration components for the design of the same system, with much higher reliability. FPGA device used to reduce the number of chips required to achieve the system in the number printed on the cord and joints are reduced, the reliability of the system can be improved.

(4) Shortening the design cycle

As FPGA devices and the programmable flexibility, use it to design a system for longer than traditional methods greatly shortened. FPGA device master degrees high, use printed circuit layout wiring simple. At the same time, success in the prototype design, the development of advanced tools, a high degree of automation, their logic is very simple changes quickly. Therefore, the use of FPGA devices can significantly shorten the design cycle system, and speed up the pace of product into the market, improving product competitiveness.

(5) Work fast

FPGA/CPLD devices work fast, generally can reach several original Hertz, far larger than the DSP device. At the same time, the use of FPGA devices, the system needed to achieve circuit

classes and small, and thus the pace of work of the entire system will be improved.

(6) Increased system performance confidentiality

Many FPGA devices have encryption functions in the system widely used FPGA devices can effectively prevent illegal copying products were others

(7) To reduce costs

FPGA device used to achieve digital system design, if only device itself into the price, sometimes you would not know it advantages, but there are many factors affecting the cost of the system, taken together, the cost advantages of using FPGA is obvious. First, the use of FPGA devices designed to facilitate change, shorten design cycles, reduce development costs for system development; Secondly, the size and FPGA devices allow automation needs plug-ins, reducing the manufacturing system to lower costs; Again, the use of FPGA devices can enhance system reliability, reduced maintenance workload, thereby lowering the cost of maintenance services for the system. In short, the use of FPGA devices for system design to save costs.

FPGA design principles :

FPGA design an important guiding principles : the balance and size and speed of exchange, the principles behind the design of the filter expression of a large number of certification.

Here, "area" means a design exertion FPGA/CPLD logic resources of the FPGA can be used to the typical consumption (FF) and the search table (IUT) to measure more general measure can be used to design logic equivalence occupied by the door is measured. "pace" means stability operations in the chip design can achieve the highest frequency, the frequency of the time series design situation, and design to meet the clock cycle -- PADto pad, Clock Setup Time, Clock Hold Beijing, Clock-to-Output Delay, and other characteristics of many time series closely related. Area (area) and speed (speed) runs through the two targets FPGA design always is the ultimate design quality evaluation criteria. On the size and speed of the two basic concepts : balance of size and speed and size and speed of swap.

One pair of size and speed is the unity of opposites contradictions body. Requirements for the design of a design while the smallest, highest frequency of operation is unrealistic. More scientific goal should be to meet the design requirements of the design time series (includes requirements for the design frequency) premise, the smallest chip area occupied. Or in the specified area, the design time series cushion greater frequency run higher. This fully embodies the goals of both size and speed balanced thinking. On the size and speed requirements should not be simply interpreted as raising the level and design engineers perfect sexual pursuit, and should recognize that they are products and the quality and cost of direct relevance. If time series cushion larger design, running relatively high frequency, that the design Jianzhuangxing stronger, more quality assurance system as a whole; On the other hand, the smaller size of consumption design is meant to achieve in chip unit more functional modules, the chip needs fewer, the entire system has been significantly reduced cost. As a contradiction of the two components, the size and speed is not the same status. In contrast, meet the timetables and work is more important for some frequency when both conflicts, the use of priority guidelines.

Area and the exchange rate is an important FPGA design ideas. Theoretically, if a design time series cushion larger, can run much higher than the frequency design requirements, then we can through the use of functional modules to reduce the consumption of the entire chip design area, which is used for space savings advantages of speed; Conversely, if the design of a time series demanding, less than ordinary methods of design frequency then generally flow through the string and data conversion, parallel reproduction of operational module, designed to take on the whole "string and conversion" and operate in the export module to chip in the data "and string conversion" from the macro point of view the whole chip meets the requirements of processing speed, which is equivalent to the area of reproduction - rate increase.

For example. Assuming that the digital signal processing system is 350Mb/s input data flow rate, and in FPGA design, data processing modules for maximum processing speed of 150Mb/s, because the data throughput processing module failed to meet requirements, it is impossible to achieve directly in the FPGA. Such circumstances, they should use "area-velocity" thinking, at least three processing modules from the first data sets will be imported and converted, and then use these three modules parallel processing of data distribution, then the results "and string conversion," we have complete data rate requirements. We look at both ends of the processing modules, data rate is 350Mb/s, and in view of the internal FPGA, each sub-module handles the data rate is 150Mb/s, in fact, all the data throughput is dependent on three security modules parallel processing subsidiary completed, that is used by more chip area achieve high-speed processing through "the area of reproduction for processing speed enhancement" and achieved design.

FPGA is the English abbreviation Field of Programmable Gate for the site programmable gate array, which is in Pal, Gal, Epld, programmable device basis to further develop the product. It is as ASIC (ASIC) in the field of a semi-customized circuit and the emergence of both a customized solution to the shortage circuit, but overcome the original programmable devices doors circuit few limited shortcomings.

FPGA logic module array adopted home (Logic Cell ), a new concept of internal logic modules may include CLB (Configurable Logic Block), export import module IOB (Input Output Block) and internal links (Interconnect) 3. FPGA basic features are :

(1) Using FPGA ASIC design ASIC using FPGA circuits, the chip can be used,while users do not need to vote films production.

(2) FPGA do other customized or semi-customized ASIC circuits throughout the Chinese specimen films.

3) FPGA internal capability and rich I/O Yinjue.

4) FPGA is the ASIC design cycle, the shortest circuit, the lowest development costs, risks among the smallest device

5) FPGA using high-speed Chmos crafts, low consumption, with CMOS, TTL low-power

compatible

It can be said that the FPGA chip is for small-scale systems to improve system integration, reliability one of the best

Currently FPGA many varieties, the Revenue software series, TI companies TPC series, the fiex ALTERA company series

FPGA is stored in films from the internal RAM procedures for the establishment of the state of its work, therefore, need to programmed the internal Ram. Depending on the different configuration, users can use a different programming methods

Plus electricity, FPGA, EPROM chips will be read into the film, programming RAM中 data, configuration is completed, FPGA into working order. Diaodian, FPGA resume into white films, the internal logic of relations disappear, FPGA to repeated use. FPGA's programming is dedicated FPGA programming tool, using generic EPROM, prom programming device can. When the need to modify functional FPGA, EPROM can only change is. Thus, with a FPGA, different programming data to produce different circuit functions. Therefore, the use of FPGA very flexible.

There are a variety of FPGA model : the main model for a parallel FPGA plus a EPROM manner; From the model can support a number of films FPGA; serial prom programming model could be used serial prom FPGA programming FPGA; The external model can be engineered as microprocessors from its programming microprocessors.

Verilog HDL is a hardware description language for the algorithm level, doors at the level of abstract level to switch-level digital system design modelling. Modelling of the target figure by the complexity of the system can be something simple doors and integrity of electronic digital systems. Digital system to the levels described, and in the same manner described in Hin-time series modelling.

Verilog HDL language with the following description of capacity : design behaviour characteristics, design data flow characteristics, composition and structure designed to control and contain the transmission and waveform design a certification mechanism. All this with the use of a modelling language. In addition, Verilog HDL language programming language interface provided by the interface in simulation, design certification from the external design of the visit, including specific simulation control and operation.　

Verilog HDL language grammar is not only a definition, but the definition of each grammar structure are clear simulation, simulation exercises. Therefore, the use of such language to use Verilog simulation models prepared by a certification. From the C programming language, the language inherited multiple operating sites and structures. Verilog HDL provides modelling capacity expansion, many of the initial expansion would be difficult to understand. However, the core subsets of Verilog HDL language very easy to learn and use, which is sufficient for most modelling applications. Of course, the integrity of the hardware description language is the most complex chips from the integrity of the electronic systems described.
history

Verilog HDL language initially in 1983 by Gateway Design Automation companies for product development simulator hardware modelling language. Then it is only a dedicated language. Since their simulation, simulation devices widely used products, Verilog HDL as a user-friendly and practical language for many designers gradually accepted. In an effort to increase the popularity of the language activities, Verilog HDL language in 1990 was a public area. Open Verilog International (OVI) is to promote the development of Verilog international organizations. 1992, decided to promote OVI OVI standards as IEEE Verilog standards. The effort will ultimately succeed, a IEEE

1995 Verilog language standard, known as IEEE Std 1364-1995. Integrity standards in Verilog hardware description language reference manual contains a detailed description.

5、In troduction of DSP

Today, DSP is w idely used in the modern techno logy and it has been the key part of many p roducts and p layed more and mo re impo rtant ro le in our daily life.Recent ly, Northw estern Po lytechnica lUniversity Aviation Microelect ronic Center has comp leted the design of digital signal signal p rocesso r co re NDSP25, w h ich is aim ing at TM S320C25 digital signal p rocesso r of Texas Inst rument TM S320 series. By using top 2dow n design flow , NDSP25 is compat ible w ith inst ruct ion and interface t im ing of TM S320C25.

Digital signal processors (DSP) is a fit for real-time digital signal processing for high-speed dedicated processors, the main variety used for real-time digital signal processing to achieve rapid algorithms. In today's digital age background, the DSP has become the communications, computer, and consumer electronics products, and other fields based device.

Digital signal processors and digital signal processing is inseparably, we usually say "DSP" can also mean the digital signal processing (Digital Signal Processing), is that in this digital signal processors Lane. Digital signal processing is a cover many disciplines applied to many areas and disciplines, refers to the use of computers or specialized processing equipment, the signals in digital form for the collection, conversion, recovery, valuation, enhancement, compression, identification, processing, the signals are compliant form. Digital signal processors for digital signal processing devices, it is accompanied by a digital signal processing to produce. DSP development process is broadly divided into three phases : the 20th century to the 1970s theory that the 1980s and 1990s for the development of products. Before the emergence of the digital signal processing in the DSP can only rely on microprocessors (MPU) to complete. However, the advantage of lower high-speed real-time processing can not meet the requirements. Therefore, until the 1970s, a talent made based DSP theory and algorithms. With LSI technology development in 1982 was the first recipient of the world gave birth to the DSP chip. Years later, the second generation based on CMOS工艺 DSP chips have emerged. The late 1980s, the advent of the third generation of DSP chips. DSP is the fastest-growing 1990s, there have been four successive five-generation and the generation DSP devices. After 20 years of development, the application of DSP products has been extended to people's learning, work and all aspects of life and gradually become electronics products determinants.

中文翻译

数字滤波器的仿真与实现

随着信息时代和数字世界的到来，数字信号处理已成为当今一门极其重要的学科和技术领域。数字信号处理在通信、语音、图像、自动控制、雷达、军事、航空航天、医疗和家用电器等众多领域得到了广泛的应用。在数字信号处理应用中，数字滤波器十分重要并已获得广泛应用。

1、数字滤波器介绍：

模拟和数字滤波器

在信号处理、过滤功能是一个不排除部分信息,如随机噪音、提取有用的信号部分,如部分地势在一定的频率范围.

有两大类型的过滤, 模拟 以及 数字 . 他们是完全不同的物理结构,如何工作. 类比电子电路模拟用的过滤部分组成,例如由电阻、电容opamps和生产所需的过滤效果. 这种过滤器被广泛使用的电路减少噪音等方面的应用,提高视频信号、图像均衡的高科技传真系统等众多领域. 有完善的技术标准设计的模拟电路进行过滤特定要求. 在各个阶段,是一个信号,是电机电压和过滤,目前直接的物理模拟量(例如声音或视频信号和变频器生产)处理. 数码过滤用数字进行数值计算处理器的信号抽样值. 处理器的可能通用计算机等PC或专业发展计划图(数字信号处理器)芯片. 模拟信号必须先投入使用的取样和数码艺术发展局(模拟到数字转换器). 由此二元多,占抽样连续输入信号的价值,转移到处理器,进行数字计算. 这些计算通常涉及多方面的投入和增加产品价值的共同因素. 如有必要,这些计算结果,现在是抽样信号值的过滤,产出通过发展援助委员会(类比数位转换器来)信号转换回模拟形式.

看到一个数字过滤,信号是由一系列数字,而不是电压或逆流.

滤波器是指用来对输入信号进行滤波的硬件或软件。如果滤波器的输入、输出

都是离散信号，则该滤波器的冲击响应也必然是离散的，这样的滤波器定义为数字滤波器。数字滤波器的功能，就是把输入序列X通过一定的运算变换成输出序列Y。

根据数字滤波器冲激响应函数的时域特性，可将数字滤波器分为两种，即无限长冲激相应IIR滤波器和有限长冲激响应FIR滤波器。IIR数字滤波器的优点是可以利用模拟滤波器设计的结果，而模拟滤波器的设计有大量图表可查，方便简单。它的缺点是相位的非线性；若需要线性相位，则要采用全通网络进行相位校正。图象处理以及数据采集传输都要求滤波器具有线性相位特性。而FIR数字滤波器可以实现线性相位，又可具有任意幅度特性。从数字滤波器的单位冲击响应来看，可分为两大类：有限冲击响应（FIR）数字滤波器和无限冲击响应（IIR）数字滤波器。

FIR滤波器却可以得到严格的线性相位，然而由于FIR滤波器的系统函数的极点固定在原点，所以只能用较高的阶数来实现其高选择性，对于同样的滤波器设计指标，FIR滤波器所要求的阶数要比IIR高5至10倍，所以成本较高，信号延迟也较大。但是如果要求相同的线性相位，则IIR滤波器就必须加全通网络进行相位校正，同样也要增加滤波器网络的节数和复杂性。FIR滤波器可以用非递归的方法实现，在有限精度下不会产生振荡，同时由于量化舍入以及系数的不确定性所引起的误差的影响要比IIR滤波器小的多，并且FIR滤波器可以采用FFT算法，运算速度快。但是不象IIR滤波器可以借助模拟滤波器的成果，FIR滤波器没有现成的计算公式，必须要用计算机辅助设计软件（如MATLAB）来计算。由此可知，FIR滤波器应用比较广，而IIR滤波器则用在相位要求不是很严格的场合。

2、MATLAB介绍：

MATLAB是矩阵实验室（Matrix　Laboratory）之意。除具备卓越的数值计算能力外，它还提供了专业水平的符号计算，文字处理，可视化建模仿真和实时控制等功能。MATLAB作为世界顶尖的数学应用软件，以其强大的工程计算、算法研究、工程绘图、应用程序开发、数据分析和动态仿真等功能，在航空航天、机械制造和工程建筑等领域发挥着越来越重要的作用。而C语言功能丰富，使用灵活方便，目标程序效率高。既有高级语言的优点，又有低级语言的特点。因此，C语言是目前应用最广的编程语言。虽然MATLAB是一个完整的、功能齐全的编程环境，但在某些情况下，与外部环境的数据和程序的交互是非常必须而且有益的。利用MATLAB设计滤波器，可以随时对比设计要求和滤波器特性调整参数，直观简便，极大的减轻了工作量，有利于滤波器设计的最优化。

在电力系统微机保护和二次控制中，很多信号的处理与分析都是基于正旋基波和某些整次谐波而进行的，而系统电压电流信号（尤其是故障瞬变过程）中混有各种复杂成分，所以滤波器一直是电力系统二次装置的关键部件。目前微机保护和二次信号处理软件主要采用数字滤波器。传统的数字滤波器设计使用繁琐的公式计算，改变参数后需要重新计算，在设计滤波器尤其是高阶滤波器时工作量很大。利用MATLAB信号处理箱可以快速有效地实现数字滤波器的设计与仿真。

MATLAB的基本数据单位是矩阵，它的指令表达式与数学,工程中常用的形式十分相似,故用MATLAB来解算问题要比用C,FORTRAN等语言完相同的事情简捷得多.当前流行的MATLAB 5.3/Simulink3.0包括拥有数百个内部函数的主包和三十几种工具包(Toolbox)。工具包又可以分为功能性工具包和学科工具包.功能工具包用来扩充MATLAB的符号计算,可视化建模仿真,文字处理及实时控制等功能.学科工具包是专业性比较强的工具包,控制工具包,信号处理工具包,通信工具包等都属于此类。

开放性使MATLAB广受用户欢迎.除内部函数外,所有MATLAB主包文件和各种工具包都是可读可修改的文件,用户通过对源程序的修改或加入自己编写程序构造新的专用工具包。

3、数字滤波器的设计

数字滤波器设计的基本要求

数字滤波器设计要经过三个步骤：

（1）确定指标：在设计一个滤波器前，必须有一些指标。这些指标要根据应用确定。在很多实际应用中，数字滤波器常常被用来实现选频操作。因此，指标的形式一般在频域中给出幅度和相位响应。幅度指标主要以两种方式给出。第一种是绝对指标。它提供对幅度响应函数的要求，一般应用于FIR滤波器的设计。第二种指标是相对指标。它以分贝值的形式给出要求。在工程实际中，这种指标最受欢迎。对于相位响应指标形式，通常希望系统在通频带中人有线性相位。运用线性相位响应指标进行滤波器设计具有如下优点：①只包含实数算法，不涉及复数运算；②不存在延迟失真，只有固定数量的延迟；③长度为N的滤波器（阶数为N-1），计算量为N/2数量级。

（2）模型逼近：一旦确定了指标，就可利用前面学习过的基本原理和关系式，提出一个滤波器模型来逼近给定的指标体系。

（3）实现：上面两步的结果得到的滤波器，通常是以差分方程、系统函数或脉冲响应来描述的。根据这个描述用硬件或者计算机软件来实现它。

4、FPGA介绍：

可编程逻辑器件是一种可以构成各种用途逻辑的通用芯片，它是实现专用集成电路ASIC（Application Specific Integrated Circuit）的半定制器件，它的出现和发展使电子系统设计师借助于CAD手段在实验室里就可以设计自己的ASIC器件。特别是FPGA（Field Programmable Gate ）的产生与发展，使其成为继微处理器、存储器之后的为电子数字系统设计而确定的又一种新的工业标准(即可以按标准产品目录在销售市场上购到)。数字系统正朝向以微处理器、存储器、FPGA三种标准积木块构成或是它们的集成方向发展。

使用FPGA器件设计数字电路，不仅可以简化设计过程，而且可以降低整个系统的体积和成本，增加系统的可靠性。它们无需花费传统意义下制造集成电路所需大量时间和精力，避免了投资风险，成为电子器件行业中发展最快的一族。使用FPGA器件设计数字系统电路的主要优点如下：

（1）设计灵活

使用FPGA器件，可不受标准系列器件在逻辑功能上的限制。而且修改逻辑可在系统设计和使用过程的任一阶段中进行，并且只须通过对所用的FPGA器件进行重新编程即可完成，给系统设计提供了很大的灵活性。

（2）增大功能密集度

功能密集度是指在给定的空间能集成的逻辑功能数量。可编程逻辑芯片内的组件门数高，一片FPGA可代替几片、几十片乃至上百片中小规模的数字集成电路芯片。用FPGA器件实现数字系统时用的芯片数量少，从而减少芯片的使用数目，减少印刷线路板面积和印刷线路板数目，最终导致系统规模的全面缩减。

（3）提高可靠性

减少芯片和印刷板数目，不仅能缩小系统规模，而且它还极大的提高了系统的可靠性。

具有较高集成度的系统比用许多低集成度的标准组件设计的相同系统具有高得多的可靠性。使用FPGA器件减少了实现系统所需要的芯片数目，在印刷线路板上的引线以及焊点数量也随之减少，所以系统的可靠性得以提高。

（4）缩短设计周期

由于FPGA器件的可编程性和灵活性，用它来设计一个系统所需时间比传统方法大为缩短。FPGA器件集成度高，使用时印刷线路板电路布局布线简单。同时，在样机设计成功后，由于开发工具先进，自动化程度高，对其进行逻辑修改也十分简便迅速。因此，使用FPGA器件可大大缩短系统的设计周期，加快产品投放市场的速度，提高产品的竞争能力。

（5）工作速度快

FPGA/CPLD器件的工作速度快，一般可以达到几百兆赫兹，远远大于DSP器件。同时，使用FPGA器件后实现系统所需要的电路级数又少，因而整个系统的工作速度会得到提高。

（6）增加系统的保密性能

很多FPGA器件都具有加密功能，在系统中广泛的使用FPGA器件可以有效防止产品被他人非法仿制。

（7）降低成本

使用FPGA器件实现数字系统设计时，如果仅从器件本身的价格考虑，有时还看不出来它的优势，但是影响系统成本的因素是多方面的，综合考虑，使用FPGA的成本优越性是很明显的。首先，使用FPGA器件修改设计方便，设计周期缩短，使系统的研制开发费用降低；其次，FPGA器件可使印刷线路板面积和需要的插件减少，从而使系统的制造费用降低；再次，使用FPGA器件能使系统的可靠性提高，维修工作量减少，进而使系统的维修服务费用降低。总之，使用FPGA器件进行系统设计能节约成本。

FPGA设计原则：

FPGA设计的一个重要指导原则：面积和速度的平衡与互换，这个原则在后边的滤波器设计中有大量的验证体现。

这里“面积”指一个设计消耗FPGA/CPLD的逻辑资源的数量，对于FPGA可以用所消耗的触发器(FF)和查找表(IUT)来衡量，更一般的衡量方式可以用设计所占用的等价逻辑门数来衡量。“速度”指设计在芯片上稳定运行所能够达到的最高频率，这个频率由设计的时序状况决定，和设计满足的时钟周期，PADto PAD Time, Clock Setup Time, Clock Hold Time, Clock-to-Output Delay等众多时序特征量密切相关。面积(area)和速度(speed)这两个指标贯穿着FPGA设计的始终，是设计质量评价的终极标准。关于面积和速度的两个最基本的概念：面积与速度的平衡和面积与速度的互换。

面积和速度是一对对立统一的矛盾体。要求一个设计同时具备设计面积最小，运行频率最高是不现实的。更科学的设计目标应该是在满足设计时序要求(包含对设计频率的要求)的前提下，占用最小的芯片面积。或者在所规定的面积下，使设计的时序余量更大，频率跑得更高。这两种目标充分体现了面积和速度的平衡的思想。关于面积和速度的要求，不应该简单地理解为工程师水平的提高和设计完美性的追求，而应该认识到它们是和产品的质量和成本直接相关的。如果设计的时序余量比较大，跑的频率比较高，意味着设计的健壮性更强，整个系统的质量更有保证；另一方面，设计所消耗的面积更小，则意味着在单位芯片上实现的功能模块更多，需要的芯片数量更少，整个系统的成本也随之大幅度削减。作为矛盾的两个组成部分，面积和速度的地位是不一样的。相比之下，满足时序、工作频率的要求更重要一些，当两者冲突时，采用速度优先的准则。

面积和速度的互换是 FPGA设计的一个重要思想。从理论上讲，一个设计如果时序余量较大，所能跑的频率远远高于设计要求，那么就能通过功能模块复用减少整个设计消耗的芯片面积，这就是用速度的优势换面积的节约；反之，如果一个设计的时序要求很高，普通方法达不到设计频率，那么一般可以通过将数据流串并转换，并行复制多个操作模块，对整个设计采取“串并转换”的思想进行运作，在芯片输出模块再在对数据进行“并串转换”，是从宏观上看整个芯片满足了处理速度的要求，这相当于用面积复制换速度提高。

举一个例子。假设数字信号处理系统输入数据流的速率是350Mb/s，而在FPGA上设计的数据处理模块的处理速度最大为150Mb/s，由于处理模块的数据吞吐量满足不了要求，看起来直接在FPGA上实现是不可能的。这种情况下，就应该利用“面积换速度”的思想，至少复制成3个处理模块，首先将输入数据进行串并转换，然后利用这三个模块并行处理分配的数据，然后将处理结果“并串变换”，就完成数据速率的要求。我们在整个处理模块的两端看，数据速率是350Mb/s，而在FPGA的内部看，每个子模块处理的数据速率是150Mb/s，其实整个数据的吞吐量的保障是依赖于3个子模块并行处理完成的，也就是说利用了占用更多的芯片面积，实现了高速处理，通过“面积的复制换取处理速度的提高”的思想实现了设计。

FPGA是英文Field Programmable Gate 的缩写，即现场可编程门阵列，它是在PAL、GAL、EPLD等可编程器件的基础上进一步发展的产物。它是作为专用集成电路（ASIC）领域中的一种半定制电路而出现的，既解决了定制电路的不足，又克服了原有可编程器件门电路数有限的缺点。

FPGA采用了逻辑单元阵列LCA（Logic Cell ）这样一个新概念，内部包括可配置逻辑模块CLB（Configurable Logic Block）、输出输入模块IOB（Input Output Block）和内部连线（Interconnect）三个部分。FPGA的基本特点主要有：

1）采用FPGA设计ASIC电路，用户不需要投片生产，就能得到合用的芯片。

2）FPGA可做其它全定制或半定制ASIC电路的中试样片。

3）FPGA内部有丰富的触发器和I／O引脚。

4）FPGA是ASIC电路中设计周期最短、开发费用最低、风险最小的器件之一。

5）FPGA采用高速CHMOS工艺，功耗低，可以与CMOS、TTL电平兼容。

可以说，FPGA芯片是小批量系统提高系统集成度、可靠性的最佳选择之一。

目前FPGA的品种很多，有XILINX的XC系列、TI公司的TPC系列、ALTERA公司的FIEX系列等。 

FPGA是由存放在片内RAM中的程序来设置其工作状态的，因此，工作时需要对片内的RAM进行编程。用户可以根据不同的配置模式，采用不同的编程方式。

加电时，FPGA芯片将EPROM中数据读入片内编程RAM中，配置完成后，FPGA进入工作状态。掉电后，FPGA恢复成白片，内部逻辑关系消失，因此，FPGA能够反复使用。FPGA的编程无须专用的FPGA编程器，只须用通用的EPROM、PROM编程器即可。当需要修改FPGA功能时，只需换一片EPROM即可。这样，同一片FPGA，不同的编程数据，可以产生不同的电路功能。因此，FPGA的使用非常灵活。

FPGA有多种配置模式：并行主模式为一片FPGA加一片EPROM的方式；主从模式可以支持一片PROM编程多片FPGA；串行模式可以采用串行PROM编程FPGA；外设模式可以将FPGA作为微处理器的外设，由微处理器对其编程。

Verilog HDL是一种硬件描述语言，用于从算法级、门级到开关级的多种抽象设计层次的数字系统建模。被建模的数字系统对象的复杂性可以介于简单的门和完整的电子数字系统之间。数字系统能够按层次描述，并可在相同描述中显式地进行时序建模。

Verilog HDL 语言具有下述描述能力：设计的行为特性、设计的数据流特性、设计的结构组成以及包含响应监控和设计验证方面的时延和波形产生机制。所有这些都使用同一种建模语言。此外，Verilog HDL语言提供了编程语言接口，通过该接口可以在模拟、验证期间从设计外部访问设计，包括模拟的具体控制和运行。

Verilog HDL语言不仅定义了语法，而且对每个语法结构都定义了清晰的模拟、仿真语义。因此，用这种语言编写的模型能够使用Verilog仿真器进行验证。语言从C编程语言中继承了多种操作符和结构。Verilog HDL提供了扩展的建模能力，其中许多扩展最初很难理解。但是，Verilog HDL语言的核心子集非常易于学习和使用，这对大多数建模应用来说已经足够。当然,完整的硬件描述语言足以对从最复杂的芯片到完整的电子系统进行描述。

历史

Verilog HDL语言最初是于1983年由Gateway Design Automation公司为其模拟器产品开发的硬件建模语言。那时它只是一种专用语言。由于他们的模拟、仿真器产品的广泛使用，Verilog HDL 作为一种便于使用且实用的语言逐渐为众多设计者所接受。在一次努力增加语言普及性的活动中，Verilog HDL语言于1990年被推向公众领域。 Open Verilog International （OVI）是促进Verilog发展的国际性组织。1992年， OVI决定致力于推广Verilog OVI标准成为IEEE标准。这一努力最后获得成功，Verilog 语言于1995年成为IEEE标准，称为IEEE Std 1364－1995。完整的标准在Verilog硬件描述语言参考手册中有详细描述。

5、DSP简介

今天，DSP广泛应用于现代技术中，它已是许多产品的关键部分，在我们日常生活中扮演着越来越重要的角色。最近，系西北工业大学Aviation微电子研究中心完成了数字信号处理器的核心NDSP25的设计，它是以TexasTms320系列为指导来完成TMS320C25的数字信号处理器设计的目标，通过用一低端设计流，NDSP25兼容了TMS320C25的时间界面和指导内容。

数字信号处理器(Digital Signal Processor)是一种适合对数字信号进行高速实时处理的专用处理器，其主要用来实时快速地实现各种数字信号处理算法。在当今的数字化时代背景下，DSP已成为通信、计算机、消费类电子产品等领域的基础器件。

数字信号处理器与数字信号处理有着密不可分的关系，我们通常说的“DSP”也可以指数字信号处理(Digital Signal Processing )，在本文里都是指数字信号处理器。数字信号处理是一门包括了许多学科并应用于很多领域的学科，是指利用计算机或是专用处理设备，以数字形式对信号进行采集、变换、滤波、估值、增强、压缩、识别等处理，得到符合要求的信号形式。数字信号处理器是用于处理数字信号的器件，因此它是伴随着数字信号处理才产生的。

DSP发展历程大致分为三个阶段：20世纪70年代理论先行，80年代产品普及和90年代的突飞猛进。在DSP出现之前数字信号处理只能依靠微处理器(MPU)来完成。但MPU较低的处理速度无法满足高速实时的要求。因此，直到20世纪70年代，有人才提出了DSP的理论和算法基础。随着大规模集成电路技术的发展，1982年世界上诞生了首枚DSP芯片。几年后，第二代基于CMOS工艺的DSP芯片应运而生。80年代后期，第三代DSP芯片问世。90年代DSP发展最快，相继出现了第四代和第五代DSP器件。经过20多年的发展，DSP产品的应用已扩大到人们的学习、工作和生活的各个方面，并逐渐成为电子产品更新换代的决定 。

本文来源：https://www.2haoxitong.net/k/doc/80bb726348d7c1c708a145e0.html