This content is not included in
your SAE MOBILUS subscription, or you are not logged in.
Fixed-Point Model Development Assistant Tool
Technical Paper
2016-01-0018
ISSN: 0148-7191, e-ISSN: 2688-3627
Annotation ability available
Sector:
Language:
English
Abstract
Development of the software using fixed-point arithmetic is known to be tedious and error-prone. Difficulty of selecting the correct data type can outwear software developers. The common retreats often sought after include manual calculation of the approximate ranges, exhaustive simulations with extreme input values and conservative development approach by using excessive word length. The first two retreats - manual calculation and exhaustive simulations - increase the software development time, and the third retreat - conservative development - leads to the excessive memory (RAM and ROM) utilization by the software. The model-based development environment such as the Simulink has graphical nature to the software with flow of data defined by connecting signal lines. The model-based software therefore gives an opportunity to trace signal flow in the software. Input-tracing method is presented to trace the flow of the input signals of the user selected block in the software model. Input signals are traced up to the point where data is having well defined range and resolution. Software developers often use data dictionary to define range and resolution of parameters, input signals and output signals of the software models. The interval arithmetic is then performed along the traced input signal paths to determine range and resolution of the output of the user selected block. Equations of QI - number bits for integer and QF - number of bits for fraction are then used to determine data type of the output signal of the user selected block. This process of determining data type of the output of a block can be repeated for several blocks in the model. The process is automated by using MATLAB m-script to simplify software development. Application of the assistant tool is demonstrated to give error-free software models with extremely high time and cost savings.
Recommended Content
Journal Article | A Model-Based Configuration Approach for Automotive Real-Time Operating Systems |
Technical Paper | The Role of the Data Dictionary in the Model-Based Development Process |
Citation
Kakade, R. and Basu, S., "Fixed-Point Model Development Assistant Tool," SAE Technical Paper 2016-01-0018, 2016, https://doi.org/10.4271/2016-01-0018.Also In
References
- Kakade , R. and Patel , J. Analytical Method for Efficient Development of Optimal Fixed-Point Software and its Application SAE Technical Paper 2014-01-0309 2014 10.4271/2014-01-0309
- Erkkinen , T. Fixed-Point ECU Development with Model-Based Design SAE Technical Paper 2008-01-0744 2008 10.4271/2008-01-0744
- MathWorks Simulink - Simulation and Model-Based Design http://www.mathworks.com/products/simulink/ October 2015
- Cengel Yunus A. Heat Transfer: A Practical Approach Second McGraw-Hill 2003 9780072458930
- Hickey Mothy J. , Qun Ju and Van Emden Maarten H. Interval Arithmetic: from Principles to Implementation Journal of the ACM 48 5 1038 1068 1999 10.1145/502102.502106
- Moore Ramon E. Interval Analysis 4 Prentice-Hall Englewood Cliffs 1966
- dSPACE Autoscaling of Variables http://www.dspace.com/en/inc/home/products/sw/pcgs/targetli/scaling.cfm November 2015
- MathWorks Overview of the Fixed-Point Tool http://www.mathworks.com/help/fixed-point/ug/overview-of-the-fixed-point-tool.html November 2015
- Embedded Coder: What is Code Replacement? The MathWorks Inc. http://www.mathworks.com/help/ecoder/ug/what-is-code-replacement-sc.html 2015