| Version 6.0 (R14) Signal Processing Blockset™ Software Release Notes | |
This table summarizes what's new in Version 6.0 (R14):
| New Features and Changes | Version Compatibility Considerations | Fixed Bugs and Known Problems | Related Documentation at Web Site |
|---|---|---|---|
| Yes Details below | Yes—Details labeled as Compatibility Considerations, below. See also Summary. | Fixed bugs | No |
New features and changes introduced in this version are
DSP Blockset has been renamed. The new name is Signal Processing Blockset™.
For this release, significant support for fixed-point development has been added to the Signal Processing Blockset product.
The following new blocks support fixed-point data types:
The following blocks now support fixed-point data types:
Digital Filter — more structures now support fixed-point data types
Matrix Scaling
The following blocks supported real fixed-point data types in the last major release. They now also support complex fixed-point data types:
Many of the Signal Processing Blockset blocks that support fixed-point data types have a new, easier-to-use interface. For more information, see Setting Block Parameters in the Signal Processing Blockset User's Guide.
Many fixed-point capable Signal Processing Blockset blocks allow you to set intermediate data types via block mask parameters. The Accumulator, Product output, and Output parameters on many such blocks have a new Inherit via internal rule setting. When you select Inherit via internal rule, the accumulator, product output, or block output word and fraction lengths will be automatically calculated for you. In general, all the bits are preserved in the internal block algorithm for quantities using this selection. That is, the accumulator, product output, or block output word and fraction lengths are selected such that
No overflow occurs
No precision loss occurs
Internal rule equations specific to each block are given in the block reference pages.
A number of fixed-point capable Signal Processing Blockset blocks now support the logging of simulation minimums and maximums and autoscaling via the Fixed-Point Settings interface.
This section gives a brief description of each of the new blocks.
The CIC Decimation and CIC Interpolation blocks are in the Filtering/ Multirate Filters library. These blocks decimate or interpolate a signal using a Cascaded Integrator-Comb filter.
The G711 Codec block is in the Quantizers library. This block encodes a linear, pulse code modulation (PCM) narrowband speech signal using an A-law or mu-law encoder. The block decodes index values into quantized output values using an A-law or mu-law decoder. The block converts between A-law and mu-law index values.
The Inverse Short-Time FFT block is in the Transforms library. This block recovers the time-domain signal by performing an inverse short-time, fast Fourier transform operation.
The LPC to/from Cepstral Coefficients block is in the Linear Prediction library. This block converts linear prediction coefficients (LPCs) to cepstral coefficients (CCs) or cepstral coefficients to linear prediction coefficients.
The Offset block is in the Signal Operations library. This block truncates vectors by removing or keeping beginning or ending values.
The Peak Finder block is in the Signal Operations library. This block finds the local maxima and/or minima of an input signal.
The Scalar Quantizer Decoder block is in the Quantizers library. This block converts each index value into a quantized output value.
The Scalar Quantizer Encoder block is in the Quantizers library. This block encodes each input value by associating it with the index value of a quantization region.
The Short-Time FFT block is in the Transforms library. This block computes a nonparametric estimate of the spectrum using the short-time, fast Fourier transform method. The Short-Time FFT block that was located in the Power Spectrum Estimation library has been renamed the Periodogram block.
The Vector Quantizer Decoder block is in the Quantizers library. This block finds the vector quantizer codeword that corresponds to a given, zero-based index value.
The Vector Quantizer Design block is in the Quantizers library. This block designs a vector quantizer using the Vector Quantizer Design Tool (VQDTool).
The Vector Quantizer Encoder block is in the Quantizers library. This block finds the index of the nearest codeword based on a Euclidean or weighted Euclidean distance measure.
The Waterfall block is in the DSP Sinks library. This block enables you to view vectors of data over time.
The Zero Crossing block is in the Signal Operations library. This block counts the number of times a signal crosses zero.
This section gives a brief description of each of the block enhancements.
The Count data type parameter of the Counter block now supports signed and unsigned integers.
The Digital Filter block now supports these additional filter structures:
FIR
Direct form symmetric
Direct form antisymmetric
IIR Biquad (SOS)
Direct form I
Direct form I transposed
Direct form II
Every filter structure now supports fixed-point data types.
Biquad (SOS) filter structures support interstage floating-point and fixed-point scale values.
The Matrix Viewer block parameters dialog box has been upgraded.
You can now use the Scalar Quantizer Design Tool to create Scalar Quantizer Encoder and Scalar Quantizer Decoder blocks inside your models.
The Sort block now supports an additional sorting algorithm. Now, for the Sort algorithm parameter, you can choose either Quick sort or Insertion sort. Previously, only the quick sort algorithm was supported.
The Short-Time FFT block that was located in the Power Spectrum Estimation library has been renamed the Periodogram block. This block computes a nonparametric estimate of the spectrum. All instances of the old Short-Time FFT block have been replaced by the Periodogram block.
| Demo Name | Signal Processing Demo Library Location | Launch Command |
|---|---|---|
Adaptive filter convergence | Adaptive Processing | lmsxyplot |
CELP speech coder | Audio Processing | dspcelpcoder |
G711 A-law and A-Mu-A conversion | Audio Processing | dspg711amua |
G711 Mu-law and Mu-A-Mu conversion | Audio Processing | dspg711muamu |
G711 and PCM encoding | Audio Processing | dspg711cmp |
Phase vocoder | Audio Processing | dsppitchtime |
Plucked string | Audio Processing | dsppluck |
Radar tracking demonstration | Aerospace | aero_radmod_dsp |
Short-Time Spectral Attenuation | Spectral Analysis | dspstsa |
Vector quantizer design | Miscellaneous | dspvqtwodim |
The Short-Time FFT demo in Spectral Analysis demo library is now the Periodogram demo.
The Acoustic Noise Canceler demo (dspanc) is now supported on all platforms. It also has a fixed-point version (dspanc_fixpt).
The Signal Processing Blockset product has a new demo library called Fixed-Point. This library contains demo models that support fixed-point data types.
Signal Processing Blockset blocks now support triggered subsystems. The exceptions are
Blocks with multiple sample times
The Signal Processing Blockset product has extended support of constant sample times to its blocks. The output of blocks with constant sample times does not change during the simulation. You can remove all blocks having constant sample times from the simulation "loop" by setting the Inline parameters option. If you select the Inline parameters check box on the Optimization pane of the Configuration Parameters dialog box, the parameters of these blocks cannot be changed during a simulation, and simulation speed is improved.
Signal Processing Blockset source blocks capable of inheriting their sample time obey a new Simulink® inherited sample time parameter. To view this parameter, open the Configuration Parameters dialog box. In the Select pane, expand Diagnostics and click Sample Time. The new parameter, Source block specifies -1 sample time appears in the left pane. This parameter can be set to none, warning (default), or error.
The Random Source block is the only block that does not obey this parameter. If its Sample time parameter is set to -1, the Random Source block inherits its sample time from its output port and never produces warnings or errors.
You can use the Signal & Scope Manager to create and view signals without using blocks. The Signal Processing Blockset product provides signal generators and viewers that you can associate with your model using the Signal & Scope Manager. To view these generators and viewers, right-click in your model, and select Signal & Scope Manager. From the Generators and Viewers lists, expand Signal Processing.
For information on how to use the Signal & Scope Manager, see Introducing the Signal and Scope Manager in the Simulink documentation.
If you have a multirate model that you want to run in MultiTasking mode and your model contains any of the blocks listed below, your reset event can be delayed as much as one reset time interval so your model behaves deterministically:
To minimize delay in multirate models, run them in SingleTasking mode.
The following blocks no longer support different sample rates at their input ports:
The Scalar Quantizer block has been replaced by the Scalar Quantizer Encoder and Scalar Quantizer Decoder blocks.
As of Version 6.0 (Release 14) of the Signal Processing Blockset product, DSP Blockset Versions 2.2 (Release 10) and earlier are obsolete and no longer supported. DSP Blockset Version 3.x (Release 11) might also be obsoleted in a future release.
Models that contain blocks from Versions 2.2 and earlier will have broken links when loaded into Simulink 6.0 (Release 14). If you have models that contain blocks from DSP Blockset Versions 2.2 or earlier, replace the older blocks by blocks from DSP Blockset Versions 4.0 (Release 12) or later before upgrading to Signal Processing Blockset 6.0 software (Release 14). Use the command dsp_links to facilitate this process.
| Version 6.0.1 (R14SP1) Signal Processing Blockset™ Software | Compatibility Summary for Signal Processing Blockset™ Software | |
| © 1984-2008- The MathWorks, Inc. - Site Help - Patents - Trademarks - Privacy Policy - Preventing Piracy - RSS |