Intel IPP信号处理开发者参考手册

需积分: 0 8 浏览量更新于2024-06-22 收藏 1.74MB DOCX 举报

身份认证购VIP最低享 7 折!

领优惠券(最高得80元）

"Intel IPP (Integrated Performance Primitives) 是一套高度优化的库，旨在加速信号处理、图像处理、数据压缩和其他计算密集型任务。开发者参考文档分为多个部分，包括对特殊情况的处理、已移除的函数、参考文献、词汇表等。文档还包含了注意事项和免责声明，指出Intel技术可能需要启用的硬件、软件或服务激活，并且没有任何产品或组件能够绝对安全。Intel不提供任何明示或暗示的知识产权许可，并对产品可能存在的设计缺陷或错误（即errata）做出了声明，这些错误可能导致产品偏离发布的规格。Intel对所有明示和暗示的保修，包括适销性、特定用途适用性的保修，均不承担责任。" 详细知识点： 1. **Intel IPP (Integrated Performance Primitives)**：这是一个由Intel开发的软件库，提供了大量预先优化的算法，用于提升计算密集型应用的性能，尤其是在信号处理领域。这些函数通常针对Intel处理器进行了专门的优化，以充分利用硬件特性。 2. **Signal Processing**：Intel IPP的主要关注点之一是信号处理，包括滤波、转换、编码和解码等多种操作，适用于音频、视频和其他时间序列数据的处理。 3. **Developer Reference**：这份文档是为开发者提供的，包含详细的技术信息，帮助他们在自己的应用中有效利用IPP库。开发者可以通过参考文档了解每个函数的功能、参数、返回值以及如何正确使用它们。 4. **Appendix A: Handling of Special Cases**：这部分可能详细阐述了在处理特殊输入或边缘条件时，IPP库是如何应对的。这通常包括异常情况、边界条件和不常见的输入值。 5. **Appendix B: Removed Functions for Signal Processing**：这部分列出了不再支持或者已经被移除的信号处理函数，可能是由于性能改进、功能重叠或者存在新版本替代。 6. **Appendix C: Bibliography for Signal Processing**：这个附录可能包含有关信号处理的参考文献列表，为开发者提供了更深入学习和理解相关领域的资源。 7. **Appendix D: Glossary**：词汇表是理解技术文档的关键，它定义了文档中使用的专业术语和缩写，帮助读者准确理解内容。 8. **Notices and Disclaimers**：Intel强调其技术可能需要特定的硬件、软件或服务激活才能运行，并提醒用户没有任何产品或组件可以确保绝对安全。此外，他们明确表示不提供任何形式的知识产权许可，并且不保证产品的适销性和特定用途的适用性。 9. **Desig Defects or Errors (Errata)**：Intel承认产品可能存在设计缺陷或错误，这些可能会影响产品符合公开的技术规格。如果需要，用户可以请求获取当前已知错误的详细信息。 10. **Limited Warranty**：Intel明确声明不承担任何明示或暗示的保修责任，包括但不限于适销性和特定用途适用性的保修，用户应自行评估和承担使用风险。以上就是Intel IPP文档涉及的核心知识点，对于使用IPP库进行信号处理的开发者来说，这些信息至关重要。

资源详情

资源推荐

2 Intel

Integrated Performance Primitives Developer Reference, Volume 1: Signal Processing

For example, names without modifiers: Add, Threshold, FirGenLowPass; with modifiers:

ippsFFTInv_CToC, Threshold_LT.

Data Types

The datatype field indicates data types used by the function, in the following format:

<bit depth><bit interpretation>,

where

bit depth = <1|8|16|32|64>

and

bit interpretation<u|s|f>[c]

Here u indicates “unsigned integer”, s indicates “signed integer”, f indicates “floating point”, and c indicates

“complex”.

Intel IPP supports the data types of the source and destination for signal processing functions listed in the

table below.

NOTE

In the lists of function parameters, the Ipp prefix is added to the data type. For example, 8-bit signed

data is denoted as Ipp8s type. These Intel IPP-specific data types are defined in the respective library

header files.

Data Types Supported by Intel IPP for Signal Processing

Type

Usual C Type

Intel IPP Type

unsigned char

Ipp8u

signed char

Ipp8s

16u

unsigned short

Ipp16u

16s

signed short

Ipp16s

16sc

complex short

Ipp16sc

32u

unsigned int

Ipp32u

32s

signed int

Ipp32s

32f

float

Ipp32f

32fc

64s

complex float

int64 (Windows*) or long

long

(Linux*)

Ipp32fc

Ipp64s

64f

double

Ipp64f

64fc

complex double

Ipp64fc

For functions that operate on a single data type, the datatype field contains only one of the values listed

above.

If a function operates on source and destination signals that have different data types, the respective data

type identifiers are listed in the function name in order of source and destination as follows:

<datatype> = <src1Datatype>[src2Datatype][dstDatatype]

For example, the function ippsDotProd_16s16sc_Sfs computes the dot product of 16-bit short and 16-bit

complex short source vectors and stores the result in a 16-bit complex short destination vector. The

dstDatatype modifier is not present in the name because the second operand and the result are of the

same type. The result is scaled and saturated.

There are several data types, namely 24u, 24s and 16f that are not supported by Intel IPP, but can be

readily converted to the supported data types for further processing by the library functions.

Intel

Integrated Performance Primitives Concepts 2

For the unsigned 24u data, each vector element consists of three consecutive bytes represented as Ipp8u

data types. It has a little-endian byte order when a lower order byte is at the lower address. These data may

be converted to and from 32u or 32f data types by using the appropriate flavors of the Intel IPP function

ippsConvert.

For the signed 24s data, each vector element consists of three consecutive bytes represented as Ipp8s data

types. It has a little-endian byte order when a lower order byte is at the lower address. The sign is

represented by the most significant bit of the highest order byte. These data may be converted to and from

32s or 32f data types by using the appropriate flavors of the Intel IPP function ippsConvert.

For the 16f format, 16-bit floating point data (half type) can represent positive and negative numbers,

whose magnitude is between roughly 6.1e

-5

and 6.5e

, with a relative error of 9.8e

-4

; numbers smaller

than 6.1e

-5

can be represented with an absolute error of 6.0e

-8

. All integers from -2048 to +2048 can be

represented exactly.

The figure below illustrates the bit-layout for a half number:

s is the sign-bit, e is the exponent, and m is the significand.

These data may be converted to and from 16s and 32f data types by using the appropriate flavors of the

Intel IPP function ippsConvert.

Descriptors

The descriptors element further describes the operation. Descriptors are individual characters that indicate

additional details of the operation.

The following descriptors are used in signal processing functions:

If the function has more than one descriptor, they are presented in the function name in alphabetical order.

Many functions have no descriptors listed above. Such functions operate with the default behavior.

Parameters

The parameters element specifies the function parameters (arguments).

Descriptor

Description

Sfs

Operation is in-place (default is not-in-place).

An in-place operation is guaranteed to be correct if and only if the Intel IPP function

performing the operation has the descriptor I in its name. In addition to the

descriptor I, all function declarations for in-place Intel IPP functions contain a

pointer to an in-place buffer (for example, pSrcDst) in the parameter list. Before

the function call, the in-place buffer contains the source data. After the function call,

the in-place buffer contains the result of the operation.

Saturation and fixed scaling mode (default is saturation and no scaling).

Operation is performed for the specified number of vectors.

2 Intel

Integrated Performance Primitives Developer Reference, Volume 1: Signal Processing

typedef struct {

Ipp32f

re;

Ipp32f im;

} Ipp32fc;

The order of parameters is as follows:

• All source operands. Constants follow vectors.

• All destination operands. Constants follow vectors.

• Other, operation-specific parameters.

A parameter name has the following conventions:

• All parameters defined as pointers start with p, defined as double pointers start with pp, for example,

pPhase, pSrc, ppState. All parameters defined as values start with a lowercase letter, for example, val,

src, srcLen.

• Each new part of a parameter name starts with an uppercase character, without underscore; for example,

pSrc, lenSrc, pDlyLine.

• Each parameter name specifies its functionality. Source parameters are named pSrc or src, in some

cases followed by names or numbers, for example, pSrc2, srcLen. Output parameters are named pDst

or dst followed by names or numbers, for example, pDst2, dstLen. For in-place operations, the input/

output parameter contains the name pSrcDst or srcDst.

Extensions

The extension field denotes an Intel IPP extension to which the function belongs. The following extensions

are supported in Intel IPP Signal Processing functions:

Extension

Description

Example

Intel IPP platform-aware functions

ippsMalloc_16u_L

See Also

Platform-Aware Functions for Signal Processing

Structures and Enumerators

This section describes the structures and enumerators used by Intel IPP for signal and data processing.

Library Version Structure

The IppLibraryVersion structure describes the current Intel IPP software version. The main fields of this

structure are:

• integer fields major and minor, containing version numbers;

• integer field majorBuild, containing update number;

• integer field build, containing build revision number;

• string field Name, containing the Intel IPP version name, for example, “ippSB SSE4.1”;

• string field Version, containing the version description, for example, “7.1.0 (r93873)”.

• string field BuildDate, containing the build date.

Complex Data Structures

Complex numbers in Intel IPP are described by the structures that contain two numbers of the respective

data type. They are real and imaginary parts of the complex number. For example, a single precision complex

number is described by the Ipp32fc structure as follows:

The following complex data types are defined: Ipp16sc, Ipp32fc, Ipp64fc.

Intel

Integrated Performance Primitives Concepts 2

typedef enum {

ippCmpLess,

ippCmpLessEq,

ippCmpEq,

ippCmpGreaterEq,

ippCmpGreater

} IppCmpOp;

typedef enum {

ippRndZero,

ippRndNear,

ippRndFinancial

} IppRoundMode;

typedef enum {

ippAlgHintNone,

ippAlgHintFast,

ippAlgHintAccurate

} IppHintAlgorithm;

typedef enum {

/* Enumeration:

ippCpuUnknown = 0x0,

ippCpuPP,

ippCpuPMX,

Processor:

/* Intel(R) Pentium(R) processor

/* Pentium(R) processor

with MMX(TM) technology

/* Pentium(R) Pro processor

/* Pentium(R) II processor

/* Pentium(R) III processor

and Pentium(R)III Xeon(R) processor

ippCpuPPR,

ippCpuPII,

ippCpuPIII,

Function Context Structures

Some Intel IPP functions use special structures to store function-specific (context) information. For example,

the IppsFFTSpec structure stores twiddle factors and bit reverse indexes needed in the fast Fourier

transform.

Two different kinds of structures are used:

• specification structures that are not modified during function operation; they have the suffix Spec in their

names

• state structures that are modified during operation; they have the suffix State in their names.

The function context interpretation is processor dependent. Therefore, these context-related structures are

not defined in the public headers, and their fields are not accessible. Intel IPP provides no option of

modifying these structures or creating a function context as an automatic variable.

Enumerators

The IppStatus constant enumerates the status values returned by the Intel IPP functions, indicating

whether the operation is error-free. See section Error Reporting in this chapter for more information on the

set of valid status values and corresponding error messages for signal processing functions.

The IppCmpOp enumeration defines the type of relational operator to be used by threshold functions:

The IppRoundMode enumeration defines the rounding mode to be used by conversion functions:

The IppHintAlgorithm enumeration defines the type of code to be used in some operations: faster but less

accurate, or vice-versa, more accurate but slower. For more information on using this enumeration, see Hint

Arguments.

The IppCpuType enumerates processor types returned by the ippGetCpuType function:

2 Intel

Integrated Performance Primitives Developer Reference, Volume 1: Signal Processing

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/

PerformanceIndex.

Notice revision #20201201

Product and Performance Information

typedef enum {

ippWinBartlett,

ippWinBlackman,

ippWinHamming,

ippWinHann,

ippWinRect

} IppWinType;

ippCpuP4, /* Pentium(R) 4 processor

and Intel(R) Xeon(R) processor */

ippCpuP4HT, /* Pentium(R) 4 processor with HT Technology */

ippCpuP4HT2, /* Pentium(R) 4 processor with Intel(R)

Streaming SIMD Extensions 3

ippCpuCentrino, /* Intel(R) Centrino(R) processor technology */

ippCpuCoreSolo,

/* Intel(R) Core(TM) Solo processor

ippCpuCoreDuo,

/* Intel(R) Core(TM) Duo processor

ippCpuITP

= 0x10, /* Intel(R) Itanium(R) processor

ippCpuITP2

/* Intel(R) Itanium(R) 2 processor

ippCpuEM64T = 0x20, /* Intel(R) 64 Instruction Set

Architecture(ISA) */

ippCpuC2D, /* Intel(R) Core(TM) 2 Duo processor */

ippCpuC2Q, /* Intel(R) Core(TM) 2 Quad processor */

ippCpuPenryn, /* Intel(R) Core(TM) 2 processor with

Intel(R) SSE4.1

ippCpuBonnell,

/* Intel(R) Atom (TM) processor */

ippCpuNehalem,

/* Intel (R) Core(TM) i7 processor

ippCpuNext,

ippCpuSSE

= 0x40, /* Processor supports Pentium(R) III

processor instruction set */

ippCpuSSE2,

/* Processor supports Intel(R) Streaming SIMD

Extensions 2 instruction set

ippCpuSSE3,

/* Processor supports Intel(R) Streaming SIMD

Extensions 3 instruction set

ippCpuSSSE3,

/* Processor supports Supplemental Streaming

SIMD Extensions 3 instruction set

ippCpuSSE41,

/* Processor supports Intel(R) Streaming SIMD

Extensions 4.1 instruction set

ippCpuSSE42,

/* Processor supports Intel(R) Streaming SIMD

Extensions 4.2 instruction set

ippCpuAVX,

/* Processor supports Intel(R) Advanced Vector

Extensions instruction set */

ippCpuAES,

/* Processor supports Intel(R) AES

new instructions

ippCpuX8664 = 0x60, /* Processor supports 64 bit extension

} IppCpuType;

The IppWinType enumeration defines the type of window to be used by the FIR filter coefficient generating

functions:

剩余643页未读，继续阅读

task_task

粉丝: 0
资源: 2

会员权益专享

Intel IPP信号处理开发者参考手册

IPP ipps.pdf

ipps_0.pdf

intel ipps 使用手册

浪潮ipps产品考试

工艺规划与车间调度集成( ipps)问题matlab

偏振纯度指数理论及应用研究进展.docx

快速傅里叶变换(FFT)算法C++实现代码.docx

论文研究-多群体混合进化算法求解IPPS问题.pdf

ipps-8.1.pdf

30天学会医学统计学你准备好了吗

213ssm_mysql_jsp 图书仓储管理系统_ruoyi.zip（可运行源码+sql文件+文档）

城市二次供水智慧化运行管理经验分享

mysql选择1232

《Java编程思想》学习笔记1(操作符、控制语句、对象、初始化与清理).doc

vscode配置c:c++环境教程.md

27-1访问者模式.py

人工智能基础大作业-基于python+PyQt5实现的双人五子棋源码+文档说明(课程设计)

《面向对象程序设计JAVA语言程序设计》期末考试试题及部分答案.doc

基于Matlab Simulink 2DPSK仿真使用simulink工具进行2dpsk编

本科毕业设计-基于python实现的人工智能辅助糖尿病遗传风险预测源码+文档说明+数据

会员权益专享

最新资源