10.2. Custom-Made Data Analysis Packages 597
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
285290295300305310315Time Domain SignalTime (s)ADC CountsFigure 10.1.21: A time series plot generated in MATLAB.10.2Custom-MadeDataAnalysisPackages
Sometimes it is not feasible to use an existing software analysis package due to
availability, cost, or customization difficulties. In such a situation one must develop
a custom analysis code. This, in fact, is not as complicated a task as it sounds,
thanks to the availability of a whole line of high level programming languages and
graphics libraries. However, before we proceed any further, one point is worth
noting. One should keep in mind that very large number of data handling and
analysis routines have already been written and are available free of cost. There
is no point in diluting one’s efforts by reinventing the wheel. For example, if one
wishes to incorporate an integration algorithm into the code, it is not necessary to
write the code from scratch since there are a host of Fortran and C language library
available that can be simply be added to the code. Now, this does not mean that
writing a data analysis package means merely adding some available bits and pieces
together. In fact, a whole package has many more components and not just the core
analysis routines. In general, any data analysis package has the following four basic
parts.
Data import/export routines.Data analysis routines.Data visualization and storage routines.