Matlab产生独立可运行C代码

关于Matlab R2009a里，提到的比较多就是Real-time Workshop (虽然以前也有很多这方面的帮助）

最近跟Mathworks工程是will聊天，彻底询问了一下关于Matlab转换成独立可编译C代码的问题，得到的答复仍然是：

1：Simulink模型转换成real-time C代码已经很完美（包含比较新的Stateflow等）

2：Matlab的代码，如果是基本的代码的话，可以转换成独立运行的C代码，但是很多工具箱还不支持。

如果需要把比如说image processing, neural network等，需要用Matlab提供的基本函数改写这些代码（似乎很麻烦，所以尽量开发的时候使用Simulink）。

现在大家关心的问题是，那些函数可以直接转化成C代码，以下是函数列表：

大家可以看到，除了基本函数意外，对信号处理的函数支持的特别全面。

这也很明显，因为很多都是embedded system,信号处理是最常用的功能。

一般的算法，可以通过基本语句完成。

Aerospace Blockset Arithmetic operators Casting Complex numbers Derivative and integral Discrete math Exponential Filtering and convolution Fixed-Point Toolbox HDL code generation Histogram Input and output Interpolation and computational geometry Logical operators

Matrices and arrays Polynomials Relational operators Rounding and remainder Set Signal processing Special values Specialized math Statistics String Structure Trigonometry

具体支持的Maltab函数列表

Function Product Remarks/Limitations abs MATLAB — abs Fixed-Point Toolbox — acos MATLAB Generates an error during simulation and returns NaN for Real-Time Workshop targets when the input value x is real, but the output should be complex. To get the complex result, make the input value complex by passing in complex(x). acosd MATLAB — acosh MATLAB Generates an error during simulation and returns NaN for Real-Time Workshop targets when the input value x is real, but the output should be complex. To get the complex result, make the input value complex by passing in complex(x). acot MATLAB — acotd MATLAB — acoth MATLAB — acsc MATLAB — acscd MATLAB — acsch MATLAB — add Fixed-Point Toolbox — all MATLAB — all Fixed-Point Toolbox — and MATLAB — angle MATLAB — any MATLAB — any Fixed-Point Toolbox — asec MATLAB — asecd MATLAB — asech MATLAB — asin MATLAB Generates an error during simulation and returns NaN for Real-Time Workshop targets when the input value x is real, but the output should be complex. To get the complex result, make the input value complex by passing in complex(x). asind MATLAB — asinh MATLAB — assert MATLAB Generates specified error messages at compile time only if all input arguments are constants or depend on constants. Otherwise, generates specified error messages at runtime. atan MATLAB — atan2 MATLAB — atand MATLAB — atanh MATLAB Generates an error during simulation and returns NaN for Real-Time Workshop targets when the input value x is real, but the output should be complex. To get the complex result, make the input value complex by passing in complex(x). beta MATLAB — betainc MATLAB — betaln MATLAB — bin2dec MATLAB — bitand MATLAB Does not support floating-point inputs. The arguments must belong to an integer class. bitand Fixed-Point Toolbox Not supported for slope-bias scaled fi objects. bitandreduce Fixed-Point Toolbox — bitcmp MATLAB Does not support floating-point input for the first argument. The first argument must belong to an integer class. bitcmp Fixed-Point Toolbox — bitconcat Fixed-Point Toolbox — bitget MATLAB — bitget Fixed-Point Toolbox — bitor MATLAB Does not support floating-point inputs. The arguments must belong to an integer class. bitor Fixed-Point Toolbox Not supported for slope-bias scaled fi objects. bitorreduce Fixed-Point Toolbox — bitreplicate Fixed-Point Toolbox — bitrevorder Signal Processing Toolbox Requires Signal Processing Blockset™ license. bitrol Fixed-Point Toolbox — bitror Fixed-Point Toolbox — bitset MATLAB Does not support floating-point input for the first argument. The first argument must belong to an integer class. bitset Fixed-Point Toolbox — bitshift MATLAB Does not support floating-point input for the first argument. The first argument must belong to an integer class. bitshift Fixed-Point Toolbox — bitsliceget Fixed-Point Toolbox — bitsll Fixed-Point Toolbox bitsra Fixed-Point Toolbox — bitsrl Fixed-Point Toolbox — bitxor MATLAB Does not support floating-point inputs. The arguments must belong to an integer class. bitxor Fixed-Point Toolbox Not supported for slope-bias scaled fi objects. bitxorreduce Fixed-Point Toolbox — bsxfun MATLAB — cart2pol MATLAB — cart2sph MATLAB — cast MATLAB — cat MATLAB — ceil MATLAB — ceil Fixed-Point Toolbox — char MATLAB — chol MATLAB Does not allow two output arguments. circshift MATLAB — class MATLAB — compan MATLAB — complex MATLAB — complex Fixed-Point Toolbox — cond MATLAB — conj MATLAB — conj Fixed-Point Toolbox — conv MATLAB — conv2 MATLAB — convergent Fixed-Point Toolbox — cos MATLAB — cosd MATLAB — cosh MATLAB — cot MATLAB — cotd MATLAB — coth MATLAB — cov MATLAB — cross MATLAB If supplied, dim must be a constant. csc MATLAB — cscd MATLAB — csch MATLAB — ctranspose MATLAB — ctranspose Fixed-Point Toolbox — cumprod MATLAB — cumsum MATLAB — cumtrapz MATLAB — deconv MATLAB — det MATLAB — detrend MATLAB If supplied and not empty, the input argument bp must satisfy the following requirements: Be real Be sorted in ascending order Restrict elements to integers in the interval [1, n-2], where n is the number of elements in a column of input argument X , or the number of elements in X when X is a row vector Contain all unique values diag MATLAB If supplied, the argument representing the order of the diagonal matrix must be a real and scalar integer value. diag Fixed-Point Toolbox If supplied, the index, k, must be a real and scalar integer value that is not a fi object. diff MATLAB If supplied, the arguments representing the number of times to apply diff and the dimension along which to calculate the difference must be constants. disp Fixed-Point Toolbox — divide Fixed-Point Toolbox Any non-fi input must be constant; that is, its value must be known at compile time so that it can be cast to a fi object. Complex and imaginary divisors are not supported. dot MATLAB — double MATLAB — double Fixed-Point Toolbox — eig MATLAB QZ algorithm used in all cases. Consequently, for the standard eigenvalue problem (B identity), results will be similar to those obtained using the following code in MATLAB: [V,D] = eig(A,eye(size(A)),'qz')However, V may represent a different basis of eigenvectors, and the eigenvalues in D may not be in the same order. Options 'balance', 'nobalance', and 'chol' are not supported. Outputs are always of complex type. ellipke MATLAB — end Fixed-Point Toolbox — eps MATLAB — eps Fixed-Point Toolbox Supported for scalar fixed-point signals only. Supported for scalar, vector, and matrix, fi single and fi double signals. eq MATLAB — eq Fixed-Point Toolbox Not supported for fixed-point signals with different biases. erf MATLAB — erfc MATLAB — erfcinv MATLAB — erfcx MATLAB — erfinv MATLAB — exp MATLAB — expint MATLAB — expm MATLAB — expm1 MATLAB — eye MATLAB Dimensions must be real, nonnegative, integer constants. factorial MATLAB — false MATLAB Dimensions must be real, nonnegative, integer constants. fft MATLAB Length of input vector must be a power of 2. fft2 MATLAB Length of input matrix dimensions must each be a power of 2. fftshift MATLAB — fi Fixed-Point Toolbox Use to create a fixed-point constant or variable. The default constructor syntax without any input arguments is not supported. The syntax fi('PropertyName',PropertyValue...) is not supported. To use property name/property value pairs, you must first specify the value v of the fi object as in fi(v,'PropertyName',PropertyValue...). Works for all input values when complete numerictype information of the fi object is provided. Works only for constant input values (value of input must be known at compile time) when complete numerictype information of the fi object is not specified. numerictype object information must be available for non-fixed-point Simulink inputs. filter MATLAB — filter2 MATLAB — fimath Fixed-Point Toolbox Fixed-point signals coming in to an Embedded MATLAB Function block from Simulink are assigned the fimath object defined in the Embedded MATLAB Function dialog in the Model Explorer. Use to create fimath objects in Embedded MATLAB code. fix MATLAB — fix Fixed-Point Toolbox — fft2 MATLAB Dimensions 1 and 2 of the input matrix must have power of two length flipdim MATLAB — fliplr MATLAB — flipud MATLAB — floor MATLAB — floor Fixed-Point Toolbox — freqspace MATLAB — gamma MATLAB — gammainc MATLAB — gammaln MATLAB — gcd MATLAB — ge MATLAB — ge Fixed-Point Toolbox Not supported for fixed-point signals with different biases. get Fixed-Point Toolbox The syntax structure = get(o) is not supported. getlsb Fixed-Point Toolbox — getmsb Fixed-Point Toolbox — gt MATLAB — gt Fixed-Point Toolbox Not supported for fixed-point signals with different biases. hex2dec MATLAB — hilb MATLAB — hist MATLAB Histogram bar plotting not supported; call with at least one output argument. If supplied, the second argument x must be a scalar constant. Inputs must be real. histc MATLAB — horzcat Fixed-Point Toolbox — hypot MATLAB — idivide MATLAB opt string must be in lowercase. For efficient generated code, MATLAB rules for divide by zero are supported only for the 'round' option. ifft MATLAB Length of input vector must be a power of 2. Output of ifft block is always complex. Does not support the 'symmetric' option. ifft2 MATLAB Length of input matrix dimensions must each be a power of 2. Does not support the 'symmetric' option. ifftshift MATLAB — imag MATLAB — imag Fixed-Point Toolbox — ind2sub MATLAB The first argument should be a valid size vector. Size vectors for arrays with more than intmax elements are not supported. inf MATLAB Dimensions must be real, nonnegative, integer constants. int8, int16, int32 MATLAB — int, int8, int16, int32 Fixed-Point Toolbox — interp1 MATLAB Supports only linear and nearest interpolation methods. Does not handle evenly spaced X indices separately. X must be strictly monotonically increasing or strictly monotonically decreasing; does not reorder indices. interp1q, see interp1 MATLAB X must be strictly monotonically increasing or strictly monotonically decreasing; does not reorder indices. intmax MATLAB — intmin MATLAB — inv MATLAB Singular matrix inputs can produce nonfinite values that differ from MATLAB results. invhilb MATLAB — ipermute MATLAB — isa MATLAB — ischar MATLAB — iscolumn Fixed-Point Toolbox — isempty MATLAB — isempty Fixed-Point Toolbox — isequal MATLAB — isequalwithequalnans MATLAB — isfi Fixed-Point Toolbox — isfimath Fixed-Point Toolbox — isfinite MATLAB — isfinite Fixed-Point Toolbox — isfloat MATLAB — isinf MATLAB — isinf Fixed-Point Toolbox — isinteger MATLAB — islogical MATLAB — isnan MATLAB — isnan Fixed-Point Toolbox — isnumeric MATLAB — isnumeric Fixed-Point Toolbox — isnumerictype Fixed-Point Toolbox — isreal MATLAB — isreal Fixed-Point Toolbox — isrow Fixed-Point Toolbox — isscalar MATLAB — isscalar Fixed-Point Toolbox — issigned Fixed-Point Toolbox — issorted MATLAB — isstruct MATLAB — isvector MATLAB — isvector Fixed-Point Toolbox — kron MATLAB — lcm MATLAB — ldivide MATLAB — le MATLAB — le Fixed-Point Toolbox Not supported for fixed-point signals with different biases. length MATLAB — length Fixed-Point Toolbox — linspace MATLAB Number of points N must be a constant that is positive, real, and integer valued log MATLAB Generates an error during simulation and returns NaN for Real-Time Workshop targets when the input value x is real, but the output should be complex. To get the complex result, make the input value complex by passing in complex(x). log2 MATLAB — log10 MATLAB — log1p MATLAB — logical MATLAB — logical Fixed-Point Toolbox — logspace MATLAB — lowerbound Fixed-Point Toolbox — lsb Fixed-Point Toolbox Supported for scalar fixed-point signals only. Supported for scalar, vector, and matrix, fi single and double signals. lt MATLAB — lt Fixed-Point Toolbox Not supported for fixed-point signals with different biases. lu MATLAB — magic MATLAB — max MATLAB — max Fixed-Point Toolbox — mean MATLAB — median MATLAB — meshgrid MATLAB — min MATLAB — min Fixed-Point Toolbox — minus MATLAB — minus Fixed-Point Toolbox Any non-fi input must be constant; that is, its value must be known at compile time so that it can be cast to a fi object. mldivide MATLAB — mod MATLAB — mode MATLAB Does not support third output argument C (cell array) mpower MATLAB — mpy Fixed-Point Toolbox — mrdivide MATLAB — mrdivide Fixed-Point Toolbox — mtimes MATLAB — mtimes Fixed-Point Toolbox Any non-fi input must be constant; that is, its value must be known at compile time so that it can be cast to a fi object. NaN or nan MATLAB Dimensions must be real, nonnegative, integer constants nargin MATLAB — nargout MATLAB — nchoosek MATLAB — nchoosek MATLAB — nchoosek MATLAB — nchoosek MATLAB — ndgrid MATLAB — ndims MATLAB — ndims Fixed-Point Toolbox — ne MATLAB — ne Fixed-Point Toolbox Not supported for fixed-point signals with different biases. nearest Fixed-Point Toolbox — nextpow2 MATLAB — norm MATLAB — normest MATLAB — not MATLAB — nthroot MATLAB — numberofelements Fixed-Point Toolbox Returns the number of elements of fi objects in the Embedded MATLAB subset (works the same as numel for fi objects in Embedded MATLAB). numel MATLAB Returns the number of elements of fi objects in the Embedded MATLAB subset, rather than always returning 1. numerictype Fixed-Point Toolbox Fixed-point signals coming in to an Embedded MATLAB Function block from Simulink are assigned a numerictype object that is populated with the signal's data type and scaling information. Returns the data type when the input is a non-fixed-point signal. Use to create numerictype objects in Embedded MATLAB code. ones MATLAB Dimensions must be real, nonnegative, integer constants or MATLAB — pascal MATLAB — permute MATLAB — permute Fixed-Point Toolbox — pi MATLAB — pinv MATLAB — planerot MATLAB — plus MATLAB — plus Fixed-Point Toolbox Any non-fi input must be constant; that is, its value must be known at compile time so that it can be cast to a fi object. pol2cart MATLAB — poly MATLAB Does not discard nonfinite input values Complex input always produces complex output polyfit MATLAB — polyval MATLAB — pow2 Fixed-Point Toolbox — power MATLAB Generates an error during simulation and returns NaN for Real-Time Workshop targets when both X and Y are real, but power(X,Y) is complex. To get the complex result, make the input value X complex by passing in complex(X). For example, power(complex(X),Y). Generates an error during simulation and returns NaN for Real-Time Workshop targets when both X and Y are real, but X .^ Y is complex. To get the complex result, make the input value X complex by using complex(X). For example, complex(X).^Y. prod MATLAB — qr MATLAB — quatconj Aerospace Blockset™ — quatdivide Aerospace Blockset — quatinv Aerospace Blockset — quatmod Aerospace Blockset — quatmultiply Aerospace Blockset — quatnorm Aerospace Blockset — quatnormalize Aerospace Blockset — rand MATLAB Does not support the V5 generator. May not match MATLAB results if seeded with negative values. randn MATLAB May not match MATLAB results if seeded with negative values range Fixed-Point Toolbox — rank MATLAB — rcond MATLAB — rdivide MATLAB — rdivide Fixed-Point Toolbox — real MATLAB — real Fixed-Point Toolbox — reallog MATLAB — realmax MATLAB — realmax Fixed-Point Toolbox — realmin MATLAB — realmin Fixed-Point Toolbox — realpow MATLAB — realsqrt MATLAB — reinterpretcast Fixed-Point Toolbox — rem MATLAB — repmat MATLAB — repmat Fixed-Point Toolbox — rescale Fixed-Point Toolbox — reshape MATLAB — reshape Fixed-Point Toolbox — rot90 MATLAB — round MATLAB — round Fixed-Point Toolbox — sec MATLAB — secd MATLAB — sech MATLAB — shiftdim MATLAB Second argument must be a constant Class of second argument must be single or double sign MATLAB — sign Fixed-Point Toolbox — sin MATLAB — sind MATLAB — single MATLAB — single Fixed-Point Toolbox — sinh MATLAB — size MATLAB — size Fixed-Point Toolbox — sort MATLAB — sort Fixed-Point Toolbox — sortrows MATLAB — sosfilt Signal Processing Toolbox Requires Signal Processing Blockset license sph2cart MATLAB — squeeze MATLAB — sqrt MATLAB Generates an error during simulation and returns NaN for Real-Time Workshop targets when the input value x is real, but the output should be complex. To get the complex result, make the input value complex by passing in complex(x). sqrt Fixed-Point Toolbox Complex and [Slope Bias] inputs error out. Negative inputs yield a 0 result. std MATLAB — strcmp MATLAB Arguments must be computable at compile time. struct MATLAB — sub Fixed-Point Toolbox — sub2ind MATLAB The first argument should be a valid size vector. Size vectors for arrays with more than intmax elements are not supported. subsasgn Fixed-Point Toolbox — subspace MATLAB — subsref Fixed-Point Toolbox — sum MATLAB — sum Fixed-Point Toolbox — svd MATLAB — tan MATLAB — tand MATLAB — tanh MATLAB — times MATLAB — times Fixed-Point Toolbox Any non-fi input must be constant; that is, its value must be known at compile time so that it can be cast to a fi object. toeplitz MATLAB — trace MATLAB — trapz MATLAB — transpose MATLAB — transpose Fixed-Point Toolbox — tril MATLAB If supplied, the argument representing the order of the diagonal matrix must be a real and scalar integer value. tril Fixed-Point Toolbox If supplied, the index, k, must be a real and scalar integer value that is not a fi object. triu MATLAB If supplied, the argument representing the order of the diagonal matrix must be a real and scalar integer value. triu Fixed-Point Toolbox If supplied, the index, k, must be a real and scalar integer value that is not a fi object. true MATLAB Dimensions must be real, nonnegative, integer constants typecast MATLAB Value of string input argument type must be lower case You may receive a size error when you use typecast with inheritance of input port data types in Embedded MATLAB Function blocks. To avoid this error, specify the block's input port data types explicitly. uint8, uint16, uint32 MATLAB — uint8, uint16, uint32 Fixed-Point Toolbox — uminus MATLAB — uminus Fixed-Point Toolbox — uplus MATLAB — uplus Fixed-Point Toolbox — upperbound Fixed-Point Toolbox — vander MATLAB — var MATLAB — vertcat Fixed-Point Toolbox — wilkinson MATLAB — xcorr Signal Processing Toolbox Does not support the case where A is a matrix Does not support partial (abbreviated) strings of biased, unbiased, coeff, or none Requires Signal Processing Blockset license xor MATLAB — zeros MATLAB Dimensions must be real, nonnegative, integer constants