Data analysis software verification Benchmark Test 2
Using Group B signals

 

 

Function of test

 

The function of this test is to check the computation of statistics for square wave signals.

 The statistics to be computed are: 

Number of Values
Maximum signal level
Minimum signal level
Mean level
Standard deviation
Root mean square
Skew
Kurtosis

   

Test Group B Signal Details

 

The data is simulated 16 bit Analogue to Digital Converter (ADC) output. The data is scaled to a full-scale range of +/-10 Volts.

Eight signals are supplied in the group.

All files have the names starting with EIS_DATA_B_ followed by:

1_short_V1.TXT          Short duration square wave signal with
                                zero mean and integral number of waves.

1_long_V1.TXT          Long duration square wave signal with
                              zero mean and integral number of waves.

2_short_V1.TXT          Short duration square wave signal with
                                zero mean and non-integral number of waves.

2_long_V1.TXT          Long duration square wave signal with
                              zero mean and non-integral number of waves.

3_short_V1.TXT          Short duration square wave signal with
                                non-zero mean and integral number of waves.

3_long_V1.TXT          Long duration square wave signal with
                               non-zero mean and integral number of waves.

4_short_V1.TXT          Short duration square wave signal with
                                non-zero mean and non-integral number of waves. 

4_long_V1.TXT          Long duration square wave signal with
                               non-zero mean and non-integral number of waves.

Analysis parameters to be used

The data is supplied in ASCII form and each file represents one signal. When loaded into an analysis package the data should be set to a sample rate of 400 samples per second. 

In each case the analysis should be performed on the entire signal. 

Reference results

The design cases for the signals are:

Signal 1_short_V1

Positive level +7.5126 V, Negative level –7.5126 V, Frequency 4.0 Hz, 8000 samples  

Signal 1_long_V1

Positive level +7.5126 V, Negative level –7.5126 V, Frequency 4.0 Hz, 128000 samples

Signal 2_short_V1

Positive level +7.5126 V, Negative level –7.5126 V, Frequency 4.0 Hz, 8054 samples  

Signal 2_long_V1

Positive level +7.5126 V, Negative level –7.5126 V, Frequency 4.0 Hz, 128054 samples

Signal 3_short_V1

Signal 3_long_V1

Positive level +9.3906 V, Negative level –5.6346 V, Frequency 4.0 Hz, 128000 samples

Signal 4_short_V1

Positive level +9.3906 V, Negative level –5.6346 V, Frequency 4.0 Hz, 8054 samples  

Signal 4_long_V1

Positive level +9.3906 V, Negative level –5.6346 V, Frequency 4.0 Hz, 128054 samples

Based on the above the theoretical results are as shown in Table 1. 

Note:  All results have been evaluated to the same precision but trailing zeros have been suppressed for clarity.

Table 1 Reference Results

 

 

Acceptable tolerance in computed results

The tolerance on the results of the specified analyses should as follows:

Number of samples processed should be exact.

Maximum and Minimum levels should be within +/- 1 ADC level (i.e. +/- 3.0e-4V referred to a 10V full scale value)

Other results should be accurate to 4 significant figures. (This is considered to be acceptable for normal analysis purposes)

 

Typical Benchmark 2 erroneous results and their causes

If good agreement with the reference results is obtained for the short signals and poor results for the long signals the most likely cause is the use of single precision floating point computation for both cases instead of double precision.  

Results return documents.

In order to aid the working party in the development of benchmark data users are invited to submit the results obtained when employing the tests. These submissions will be assessed by the working party and summaries may be used in presentations and publications of the working party’s activities.

If you wish to submit results please use the document contained in Appendix A.

Appendix A

Benchmark 2.  Analysis results return.

Name:                                                Organisation:                                                Date:

Address:

Email:                                                 Tel:                              Fax:

 

Analysis Package used:                                                  Version:

 

Computer Hardware and operating system details

Processor:                             Speed:                                   Memory size:                       Mb:

Operating System:                     Version:

 

Signal	No. Sample Processed	Max. level	Min. level	Mean	R.M.S.	S. D.	Skewness	Kurtosis	Crest Factor
B_1_short
B_1_long
B_2_short
B_2_long
B_3_short
B_3_long
B_4_short
B_4_long

 

Comments:

 

Suggestions for further benchmark tests:

 

 

 

Return results to:  Dr. B. J .May  1, Westhall Road, Mickleover, Derby. DE3 0PA. UK

Email: verification@e-i-s.org.uk               Tel: +44 (01332) 737034