Upgrading to Release 11 (Release 11 New Features)  Release 11 New Features if ((parent.frames.lengthUpgrading to Simulink 3.0 From Simulink 2.1 Starting in Version 2.2 (Release 10), Simulink changed the way it treats direct-feedthrough loops containing triggered subsystems. Before Version 2.2, Simulink treated such loops as algebraic and attempted to use an algebraic solver to solve them. Beginning in Version 2.2, Simulink treats direct feedthrough loops containing triggered subsystems as nonalgebraic, thereby allowing use of nonalgebraic solvers. For example, Simulink 2.1 unsuccessfully attempts to simulate the following counter model, using an algebraic solver. In particular, it stops the simulation after detecting discontinuities in the algebraic solution. Simulink 2.2 and subsequent versions successfully use a variable-step discrete solver to simulate the same model. Upgrading to DSP Blockset 3.0 and Communications Toolbox 1.4 The New Complex Data Format Versions 1.0 through 2.2 of the DSP Blockset and Versions 1.0 through 1.3 of the Communications Toolbox provided complex data capability by creating a double-length real vector whose first half contained the real components of the vector's elements, and whose second half contained the imaginary components of the vector's elements. This format was generally only recognized by other blocks in the DSP Blockset and Communications Toolbox that had the complex identifier (*) at the appropriate port. Simulink 3.0 provides an intrinsic complex data type (see Chapter 8 of Using Simulink) that supplants the earlier DSP Blockset and Communications Toolbox implementation. Complex data in Simulink is now handled very much the same as complex data in MATLAB. Double-length vectors are no longer used to convey complex information. Your existing complex-data models will continue to work in Simulink 3.0 since the complex capability in older models is implemented solely by the blocks themselves. The older blocks also continue to be available to you by typing dsplib(2) and commlib(1) at the command line. However, 3.0 blocks cannot be directly intermixed with the older blocks in complex-data models. The next section explains options for upgrading. Why You Need to Update Your Models to Use the New Complex Data Format The new complex data format introduced by Simulink 3.0 is not compatible with the old complex data format implemented in the DSP Blockset 2.2 and the Communications Toolbox 1.3. Mixing new (i.e., Simulink 3.0 and its associated toolboxes and blocksets) and old (2.2) blocks in a simulation using complex data can easily lead to errors and incorrect results. For small models, we recommend completely replacing all old blocks in the complex portion of the model with their new equivalents when you begin incorporating new blocks into the model. For larger models, where the required block substitutions might be extensive, you can use the Convert Complex DSP to Simulink block to convert complex data in the old format to complex data in the new format. To convert back to the old format, use the Convert Complex Simulink to DSP block. The figure below shows how you can use these two convertor blocks to migrate part of a complex-data model to the new complex format while letting other components continue to use the old complex-data format. The convertor blocks are only needed for interfacing new blocks to the complex-data section of an older model. New blocks can be added to real-data sections of older models without any data format conversion. The Convert Complex DSP to Simulink and Convert Complex Simulink to DSP blocks are provided to facilitate the transition from 2.2 to 3.0 by allowing you to upgrade your models incrementally, as convenient. Ultimately, of course, you should try to migrate all of your models to the 3.0 complex data format. Note Within a section of model that uses the old complex format, you should continue to use the complex port identifier (*) as a guide to wiring blocks. Output ports labeled with the * symbol should only be connected to input ports labeled with the * symbol. Locating Old Blocks in a Model In a DSP Blockset model that contains both old and new blocks, you can determine which blocks are linked to a particular version by using the dsp_links utility. At the command line, type blks = dsp_links to highlight (in red) blocks in the current model that are linked to old libraries. Blocks that are linked to new libraries are highlighted in blue. The dsp_links command analyzes all levels of the model, and also displays a summary in the command window of the number of blocks linked to each library version. The function returns a structure, blks, containing separate lists of the old and new blocks in the model. Similarly, you can use the comm_links utility to perform the same analysis for the Communications Toolbox. Upgrading Optimization Toolbox 2.0 New large-scale algorithms have been incorporated into the Optimization Toolbox 2.0 functions. The new functionality improves the ability of the toolbox to solve large sparse problems. To accommodate this new functionality, many of the function names and calling sequences have changed. For more information on how to convert your old syntax to the new function calling sequences, see the Optimization Toolbox User's Guide. Upgrading to Fuzzy Logic Toolbox 2.0 Note The Fuzzy Logic Toolbox has not been updated for Release 11. If you have already updated your FIS models from a pre-2.0 version of the Fuzzy Logic Toolbox to the 2.0 level (as part of a Release 10 upgrade), you can ignore this section. In the Fuzzy Logic Toolbox 2.0, the Fuzzy Inference System (FIS) is represented as a MATLAB structure. A structure (instead of a flat matrix) is now the basic element in constructing a fuzzy logic system. This fundamental change in the way of representing fuzzy logic system makes many details of working with the constructed system easier. A Fuzzy Inference System that you created with a pre-2.0 version of the Fuzzy Logic Toolbox is still usable in 2.0, if you run the convertfis function on it. The convertfis function automatically converts pre-2.0 Fuzzy Inference Systems to work with Version 2.0. [ Previous | Next ]