AutoSignal Release Notes v1.5

This Release Notes section contains information that was not included in the printed documentation:

(1) Updating from Release 1.0 (Select version)

(2) AutoSignals Screen Saver

(3) Memory Issues with STFT and CWT Analyses

(4) Prony Fit of Multicomponent Exponential Decays

(5) Requesting Enhancements to AutoSignal

(6) Importing Time Series from Multicolumn Data Sources

(1) Updating from Release 1.0 (Select version)

If you are updating from AutoSignal Select (v1.0), the following enhancements and features have been added to v1.5:

Most algorithms within AutoSignal now offer a powerful automation facility that enables the unattended processing of large numbers of data sets stored in an Excel file. Also available is the means to tie in real-time acquisition and processing using a custom automation DLL. The export in the automation can be to an RTF file containing graphs and numeric summaries as well as to an Excel 95 or Excel 97 file containing the numeric results of the analyses.

The Fourier Spectra with Data Window Comparison procedure enables up to three windowed spectra to be simultaneously graphed. This option can assist in selecting the best tapering window for the specific data set being processed.

The AR Spectrum with Algorithm Comparison allows up to three AR procedures to be simultaneously processed and plotted. This option can assist in selecting the optimum AR procedure and signal space setting.

The Deconvolve Gaussian Response Function option manages the instance where a signal is smeared by a one or two-sided Gaussian response function. The deconvolution seeks to recover the true signal that would have been measured using an ideal sensing system.

The Deconvolve Exponential Response Function option is for instances where a signal is smeared by a first order or exponential response function. This is always a one-sided deconvolution that seeks to recover the true signal absent the delay of the measurement system.

(2) AutoSignals Screen Saver

AutoSignal automatically places an attractive wave-based screen saver in your Windows directory. This screen saver creates letters from multicomponent sinusoids as well as multi-sinusoid dancing waveforms. To select this screen saver, right click on an empty region on the desktop and select Properties. Click the Screen Saver tab and select the AutoSignals option from the drop down list. The screen saver can be removed by deleting AutoSignals.scr.

(3) Memory Issues with STFT and CWT Analyses

The Short-Time Fourier Transform and Continuous Wavelet Transform options are particularly memory intensive. For the STFT, a separate FFT is generated for each segment. Similarly, separate FFTs are made for each scale or frequency in the CWT. For memory reasons, AutoSignal limits the number of STFT segments (time snapshots) to a maximum that will be in the vicinity of 512 and the number of evaluated CWT frequencies is limited to a maximum of 100.

The STFT is especially prone to using up available memory. In the CWT, zero padding is only used to prevent wraparound effects in the convolution. No additional memory is used as a consequence of zero padding. In the STFT, however, zero padding results in additional spectral frequencies whose magnitudes must be stored in memory. In both cases, the spectral data is fitted to a bicubic B-spline for 3D rendering and surface integration. AutoSignal limits the total rendering grid to 16,384 elements. If the spectral grid is larger than this, an averaging decimation is used. An STFT or CWT surface is then stored as a grid of B-spline coefficients that consumes yet more memory.

The amount of physical memory (RAM) free for AutoSignal's use is shown in the main status window in the Mem field. When dealing with large data sets, particularly WAV files, it is not difficult to exhaust this memory. When this happens, Windows uses the hard disk for memory operations. Excessive disk activity and extremely slow processing and procedure closure times will result if the physical memory is insufficient.

To best conserve memory resources with the STFT, zero pad the segments only when absolutely necessary, and do so only to the extent this is needed for frequency resolution. Further, avoid high overlap settings which result in such a large number of segments. The relation with overlap is not linear. A modest increase in overlap may significantly increase the number of segments and the amount of memory required. Although AutoSignal permits overlaps as high as 90%, there is usually little benefit beyond 50-70%.

For the CWT, the 35 frequency default is usually adequate. When very low frequencies are present, 35 logarithmic frequencies generally suffice. When both very low and high frequencies are present, a higher frequency count may be required. In the CWT, the memory relationship is linear. Doubling the frequency count doubles the amount of physical memory needed. Typically, there is little to gain beyond 50-60 CWT frequencies.

If these guidelines are insufficient to prevent the hard disk thrashing and drastically diminished performance that results from exhausting physical memory, you can try breaking up the large data stream into smaller separate data sets. Given the relatively low cost of RAM, upgrading to 64, 96, or 128 Mb may be a good investment if you will be doing a good deal of non-stationary analysis of large data streams.

(4) Prony Fit of Multicomponent Exponential Decays

The Prony Spectrum procedure is useful for fitting waveforms consisting of damped sinusoids as well as fitting data that consist of multiple simultaneous first order exponential decays. Since a damped sinusoid requires two eigenmodes of signal space to be represented, and an exponential decay requires only a single eigenmode, AutoSignal seeks to automatically set the component count in the Prony procedure based upon the type of signal elements present. If a component's frequency is less than 1e-8*Nyquist, it is treated as a real exponential. If the Allow Real Exp option is checked, the component is assigned one eigenmode of signal space. Otherwise, the component is not included in the Prony fit. If a component's frequency is non-zero, it is assumed to be a damped or undamped sinusoid, and two eigenmodes of signal space are assigned.

When modeling continuously decreasing multicomponent radioactive decay data, use one unit of signal space for each component present. For three different components that have appreciably different half-lives, the signal space should be set to 3 rather than 6.

(5) Requesting Enhancements to AutoSignal

Although the non-linear optimization in this first version is limited to modeling sinusoids and exponentially-damped sinusoids, we can readily add parametric component models. Please advise us of models that have proven useful to you in modeling signal oscillations.

In this initial release, the algorithms are limited to the most important 2D time-signal analysis and processing procedures. We anticipate addressing multiple dimensions, multiple channels, cross-spectra, and higher order spectral procedures in a subsequent release. Please advise us of specific procedures in these areas that you have found to have particular importance in your signal analysis work.

Further, this initial version processes real time domain data only. While all data begin this way, we recognize that time-domain intermediates can be complex. If processing complex time-domain data is important to you, please advise us of procedures you would like to see implemented.

Send enhancement suggestions directly to support@clecom.co.uk.

(6) Importing Time Series from Multicolumn Data Sources

There are instances where multiple time series are stored in a multicolumn data source, and yet no column of X (time) values is present. While we recommend including the time values in the first column (when all series have the same sampling rate and duration), or in alternating columns (when the series have different sampling rates and/or durations), you can import only the series data and have the time values automatically assigned. In the Import procedure, select the same column for the X and Y values. In order to generate the X (time) values, answer Yes when queried if you want to specify the X spacing (sampling interval) and X starting value (initial time).