What's new in LabVIEW 8 for Signal Processing, Analysis, and Math?

LabVIEW 8 includes a variety of additions and enhancements for mathematics, signal processing, and analysis. The new release includes a library that is more comprehensive and powerful, including over 500 functions in this area.

Curve Fitting

With a larger set of built-in models for linear parametric fitting and a powerful new interface for the Levenberg-Marquardt non-linear curve fitter, curve fitting within LabVIEW is more robust, allowing you to address a wider variety of problems.

• Improved palette organization
• New non-linear parametric models: Gaussian, Logarithmic and Power
• New robust fitting algorithms: least absolute residual and bisquare
• New quality-of-fit tools: sum-of-squared error, r-square, root-mean-squared (RMS) error and confidence intervals
• New Nonlinear Levenberg-Marquardt API (application programming interface) allows flexible model specification
• New cubic spline non-parametric fitting tool

Interpolation

New interpolation methods give you more control over the shape of the line or surface making these functions more useful for applications such as defining motion paths and generating a visually appealing surfaces.

• New interpolation methods: nearest neighbor, cubic Hermite, Lagrange, Fourier, bilinear, bicubic and bicubic spline
• Improved interface: single VI call consolidates multiple types of interpolation
• Improved usability: array-based interpolation
  
  
  

Optimization

New constrained nonlinear optimization allows you to address problems with limits on allowed input / output values. Also, new interface (API) allows you to define a larger set of problems by allowing you to specify the model with either a formula string or a VI.

• New API (application programming interface) allows flexible model specification
• New optimization tools: quadratic programming and constrained nonlinear optimization

Probability / Statistics

New and updated functions offer greater coverage for probability and statistics. New tools for inferential statistics allow you to infer information about a population, given a sample from that population. Improved descriptive statistics allow you to summarize datasets.

• New continuous probability distributions: Beta, Cauchy, Exponential, Extreme value, Gamma, Laplace, Lognormal, Pareto, Rayleigh, Triangular, Uniform and Weibull
• New discrete probability distributions: Bernoulli, Binomial, Geometric, Hypergeometric, Negative Binomial, Poisson and Uniform (Discrete)
• Complete palette of Cumulative Distribution Functions (CDFs), Inverse CDFs, Probability Density Functions (PDFs), Moments and Random generation VIs
• 18 continuous distributions for each
• 7 discrete distributions for each
• New inferential statistics: T test, Z test, correlation test, sign test (non-parametric), Wilcoxon signed rank test and rank transformation
• Improved descriptive statistics: measures of mean (arithmetic, geometric, harmonic, trimmed, median), measures of spread (standard deviation, range, mean absolute deviation, interquartile range), percentiles, covariance matrix (like a 2D variance), correlation coefficient (Pearson), Correlation coefficient (Spearman) and Correlation coefficient (Kendall's tao)

Windows

New window functions allow LabVIEW to generate/apply a more comprehensive set of windows, including three new window types and new symmetric forms of all supported windows.

• New window types: Gaussian, Chebyshev and Blackman-Nutall
• Also included: Hanning, Hamming, Triangle, Blackman, Exact Blackman, Blackman-Harris, Flat top, Kaiser-Bessel, Cosine Taped, Force and Exponential
• New Symmetric forms of all supported windows
• New complex value input for all supported windows
• New functions for calculating window properties

Math

New matrix datatype

• New fundamental datatype (wire) offers size checking for linear algebra operations

Coordinate Transforms

• Coordinate Shift
• Coordinate Rotate
• 3D Coordinate Conversion (Cartesian / Spherical / Cylindrical)

Differential Equations

New interface (API) allows you to define a larger set of problems by allowing you to specify the model with either a formula string or a VI. New solvers allow you to address a larger set of problems.

• New API (application programming interface) allows flexible formula specification
• 5 new solvers (8 total)

Digital Signal Processing (DSP)

2D DSP

• 2D FFT / Inverse 2D FFT
• 2D Discrete Sine Transform / Inverse 2D Discrete Sine Transform
• 2D Discrete Cosine Transform / Inverse 2D Discrete Cosine Transform
• 2D Autocorrelation
• 2D Crosscorrelation
• 2D Convolution

New complex-value inputs

• Digital Filtering: Digital FIR Filter, Digital IIR Filter
• FFT Power Spectrum, FFT Spectrum
• 1D/2D Autocorrelation
• 1D/2D Crosscorrelation
• 1D/2D Convolution

New parameter normalization

• 1D Autocorrelation
• 1D Crosscorrelation
• Zero-phase digital filter

New transforms

• Discrete Cosine Transform (DCT) / Inverse Discrete Cosine Transform
• Discrete Sine Transform (DST) / Inverse Discrete Sine Transform
• Chirp-Z transform

MathScript Textual Math

With the release of National Instruments LabVIEW 8, you have freedom to choose the most effective syntax for technical computing, whether you are developing algorithms, exploring DSP concepts, or analyzing results. In the new release, you can combine the intuitive LabVIEW graphical dataflow programming with MathScript, a math-oriented textual programming language that includes over 500 commonly used functions.

MathScript is an integrated part of LabVIEW and offers both interactive and programmatic interfaces. With the interactive interface, you can enter commands on a command line or through an editor window to immediately see results. As you work, a variable window updates to show the graphical / textual results and a history window tracks your commands.

The MathScript programmatic interface allows you to build custom software and instrument your math scripts by integrating them into LabVIEW graphical programming. The approach gives you the freedom to select the most appropriate syntax to define the operation of your code. For an abridged list of MathScript functions, please click on the link in the Related Links section below.

Related Links

• NI LabVIEW MathScript
• What's new in LabVIEW 8.20 for Signal Processing, Analysis and Math
  

67 ratings | 3.43 out of 5

Print

Reader Comments | Submit a comment »

Peak analysis with curved baseline correction
Chromatography is widely used in many areas of research and analysis both academic and industrial to separate compounds. Their peak location in time and peak area needs to be determined. Baselines are often curved. Ability for LabVIEW to fit such a baseline and then determine the start and end points of peaks with calculation of area under each peak would be most welcome. For me, it's one of those "I can't believe LabVIEW doesn't have this capability".
- Steve Parus, University of Michigan. sparus@umich.edu - May 25, 2006

spelling Levenberg-Marquardt
the correct spellin gis "Levenberg- Marquardt"
- Gabe Wikipil, Micro Mo Electronics. gabe.wikipil@micromo.com - Apr 12, 2006

AAA+++
- antonio@consolandi.net - Feb 24, 2006

New Function Suggestion
The signal processing suite for LV8 is a very nice improvement. Here's a suggestion for LV9: An implementation of the "Damped Richardson- Lucy Deconvolution" algorithm may be useful to people who need to sharpen 2d images. A 1d iterative deconvolution algorithm may be useful as well.
- Dennis Kessler, Ericsson. dennis.kessler@ericsson.com - Dec 12, 2005

AUC
you could probably write your own. it wouldn't be that hard if you know certain parameters such as function curve and range. if you don't know the function, then you'll have to approximate using numerical analysis/techniques. MATLAB is good for such tasks.
- Otman Estrada, Raytheon. oestrada@raytheon.com - Nov 22, 2005

area under curve function
One of the most used operations in medical data analysis is integration, obtaining the area under a curve; why isn't this part of your analysis functions??? It should be dynamic allowing the user to shift the baseline and define the start and end point of integration via mouse clicks, with the AUC appearing as an area plot as well as a numerical value.
- ROBERT FEALEY, MAYO CLINIC Foundation& . fealeybobby@msn.com - Nov 1, 2005