This section aims to provide a reference for the functionality of the GDMS. The main functionality of this package is provided through the following menus. When the application is first run, you are presented with a blank window with File and Help menus. To open a data set, open the File menu and select Open. This can also be done by using the keyboard shortcut CTRL-O.

imggdmsFirstOpen.png

While a data set is being loaded, a progress window appears.

imggdmsProgress.png

Once the data set has been loaded, the application looks like:

imggdmsData.png

To close a dataset, select File - Close. Note that this will not terminate the application. To do do this, see Exit below. To save a data set after processing, select File - Save. The following window appears.

imggdmsSaveAs.png

Choose the desired location and file name and click OK. To exit the application, select File -> Exit or use the keyboard shortcut, CTRL-Q. This functionality has not been included in this release. The Edit menu exists for future enhancements such as Undo, Redo and Delete. The undo functionality however, is inherit due to the way GDMS creates a new dataset after every mathematical operation. The View Menucan be used to configure the look and feel of the user interface to suit individual users. Please refer to the User Interface Configuration chapter for full details on these features. The data processing functionality provided by the GDMS is provided in the Maths menu. All mathematical operations on a data set result in at least one additional dataset being created. This enables easy reference to both the original and processed data for comparison. Furthermore, any dataset that can be viewed within the application can be saved for future reference. The only time domain modelling currently available is Least Squares regression. This can be accessed by selecting Maths - Least Squares. The available options in the least squares menu are Variance Co-variance and Random Walk. When this option is presented, you are given the option of using the automatic re-weighting feature. If you choose to use automatic re-weighting then our algorithms will weight out any outliers in the dataset, then output a least squares regression. The regression outputs two new data sets; the first shows the dataset with the least squares regression line with the weighted points rendered in a different colour. The second dataset loaded is the residuals, also with weighted out data points displayed in a different colour. If NO is selected when prompted for automatic re-weighting, the option is given to load a weighting matrix from file for each direction. If NO is selected here, a data range may be specified before continuing with the VCV analysis by pressing Do VCV. Selecting YES to load a matrix for a file will result in a prompt request on the file's location. Once a matrix is chosen, the analysis will again provide both the least squares and residual datasets. Random walk functionality has not yet been activated in the GDMS engine. This is mostly implemented, however no agreement has yet been reached with regard to the exact format that this function will take. The available options in the Interpolation menu are Nearest Neighbour, Linear Interpolation, Natural Spline, Cubic Spline and Divided Differences. To carry out interpolation, select the Maths menu followed by the Interpolation menu and from here select the desired method. Enter the required sample rate when prompted and click OK to proceed or CANCEL to return to the main window. The available options in the Windowing menu are Rectangular, Hanning, Hamming, Blackman and Dolph-Chebyshev. To window data, go to the Maths menu followed by the Windowing menu and from here select the desired method. Each method prompts for a window size and an overlap. The window size must be number satisfying size=2n and the standard overlap is 50% of the window size. In addition, Dolph-Chebyshev prompts for the Normalised Transition Bandwidth, the values for which are restricted to the range 0.0 - 0.5. In application these numbers must be restricted to the range 0.02 -0.499 to avoid division by zero. Enter the information and click OK to proceed. When windowing is carried out on a dataset, all windows are displayed. For flexibility, the GDMS allows individual windows to be displayed. To do this, open the goto menu and navigate through the windows as desired. Select the Maths menu, select Fast Fourier Transform which will generate a PSD plot. Note, before successfully carrying out operations in the frequency domain, the data must be interpolated. The Failure to do so will result in an error message from the GDMS. The error message gives the first three dates in the dataset that caused the failure. In this release the Help menu contains only an About box. This is an area for future expansion.