Typumwandlung und Quantisieren von Daten

Quantisieren in Festkomma- und Gleitkomma-Datentypen, Typumwandlung zwischen Datentypen

Bei der Entwicklung von Festkomma-Algorithmen können Sie cast, zeros, ones, eye und subsasgn verwenden, um den Kernalgorithmus von den Datentyp-Definitionen zu trennen. Mit diesen Funktionen können Sie Double-Precision-, Single-Precision- und Festkomma-Datentypen im selben Code verwenden. Manual Fixed-Point Conversion Best Practices beschreibt, wie Sie ausgehend von generischem MATLAB®-Code eine effiziente Festkomma-Implementierung erstellen können.

Um Arithmetik mit voller Präzision mit Doubles zu simulieren und nur die Ausgabe des Algorithmus zu quantisieren, verwenden Sie quantizenumeric.

Um Gleitkommadaten mit beliebiger Wort- und Exponentenlänge zu simulieren, verwenden Sie das Objekt quantizer mit dessen zugehörigen Objektfunktionen.

Funktionen

alle erweitern

eyeCreate identity matrix with fixed-point properties
onesCreate array of all ones with fixed-point properties
zerosCreate array of all zeros with fixed-point properties
castCast variable to different data type
cast64BitFiToIntCast fi object types that can be exactly represented to a 64-bit integer data type
cast64BitIntToFiCast 64-bit integer types to an equivalent fi object type
castFiToIntCast fi object to equivalent integer data type
castFiToMATLABCast fi object type to an equivalent built-in MATLAB data type
castIntToFiCast an integer data type to equivalent fi type
subsasgnSubscripted assignment
subsrefSubscripted reference
quantizenumericQuantize numeric data
quantizerCreate quantizer object
quantizeQuantize numeric data using quantizer object
assignmentquantizerCreate quantizer object with fi object attributes
copyobjMake independent copy of quantizer object
randquantGenerate uniformly distributed, quantized random number using quantizer object
resetReset objects to initial conditions
unitquantizeQuantize numeric data using quantizer object except numbers within eps of +1
unitquantizerCreate unitquantizer object
denormalmaxLargest denormalized quantized number for quantizer object
denormalminSmallest denormalized quantized number for quantizer object
epsQuantized relative accuracy for fi or quantizer objects
exponentbiasExponent bias for quantizer object
exponentlengthExponent length of quantizer object
exponentmaxMaximum exponent for quantizer object
exponentminMinimum exponent for quantizer object
fractionlengthFraction length of quantizer object
getProperty values of object
lsbScaling of least significant bit of fi object, or value of least significant bit of quantizer object
rangeNumerical range of fi or quantizer object
realmaxLargest positive fixed-point value or quantized number
realminSmallest positive normalized fixed-point value or quantized number
setSet or display property values for quantizer objects
tostringConvert fi, fimath, numerictype, or quantizer object to string
wordlengthWord length of quantizer object
logreportQuantization report
maxlogLog maximums
minlogLog minimums
noperationsNumber of quantization operations by quantizer object
noverflowsNumber of overflows
nunderflowsNumber of underflows
resetlogClear log for fi or quantizer object
bin2numConvert two's complement binary string to number using quantizer object
hex2numConvert hexadecimal string to number using quantizer object
num2binConvert number to binary representation using quantizer object
num2hexConvert number to hexadecimal equivalent using quantizer object
num2intConvert number to signed integer using quantizer object
isequalDetermine whether real-world values of two fi objects are equal, or determine whether properties of two fimath, numerictype, or quantizer objects are equal
isfixedDetermine whether input is fixed-point data type
isfloatDetermine whether input is floating-point data type
isquantizerDetermine whether input is quantizer object
errmeanMean of quantization error
errpdfProbability density function of quantization error
errvarVariance of quantization error
roundRound fi object toward nearest integer or round input data using quantizer object
fixed.QuantizerQuantize fixed-point numbers
quantizeQuantize fi values using fixed.Quantizer object

Themen

Enthaltene Beispiele

Compute Quantization Error

Compute Quantization Error

Compute and compare the statistics of the signal quantization error when using various rounding methods.

Implement FIR Filter Algorithm for Floating-Point and Fixed-Point Types Using cast and zeros

Implement FIR Filter Algorithm for Floating-Point and Fixed-Point Types Using cast and zeros

Convert a finite impulse-response (FIR) filter to fixed point by separating the fixed-point type specification from the algorithm code.

Detect Limit Cycles in Fixed-Point State-Space Systems

Detect Limit Cycles in Fixed-Point State-Space Systems

Analyze a fixed-point state-space system to detect limit cycles.