DSTO-GD-0248 PR - Department of Defence
Contents
1. Figure 14 Rooftop checker signal zoomed in on the time scale A spectrogram would show a dominant lineat 100 Hz 2 5 Storage of Digitised D ata Since most of the digitised data files are large the PC has a 650 MByte CD writer and 100 MByte zip drive installed This enables the data to be easily stored onto CD for later processing and archiving or copied to another PC using the zip drive The data files produced contain 16 bit little endian binary data The A D card only digitises at 12 bit but the data files are saved as 16 bit since this is a more commonly used format 3 Conclusions This set up enables simplified digitisation of the FM tracks on AQH4 tapes by combining the digitisation validation and storage process into one PC Also digitising many channels at once reduces the time used to digitise the required tracks and gives synchronous data Testing of the anti aliasing filter showed that the filter was not working at the level expected but was good enough for the settings used see Table 2 This filter is only an eight channel filter which could be upgraded to include another eight channels enabling all 16 FM channels to be digitised simultaneously DSTO GD 0248 The validation of the digitised data files has also improved by using a COTS program called Cool Edit Cool Edit enables the data to be viewed as a time series or spectrogram with the ability to zoom in on segments of data and listen to it 4 Refer2. Contents L INTRODUCTION asa 1 2 THE TRAN SCRIPTION PROCESS wi cissssissssssscisscsssedsancesctescvetsaacssscesncdsavastecadunussaveraaossies 1 2 1 Replay Of AQH A Tapes iii i 2 2 2 ANUA lasing Filter ir he ah 3 22 1 Using the Filters prorrata 3 2 22 selecting the cut Off frequency asi regel 4 23 Digitisation of AQHAFM Tracks mmm errar 6 24 Validation of Data Digitised aaa 10 2 5 Storage of Digitsed Data 00 000000 iere Lora a YR C Ya ck aa FEE cV RE CREE 14 3 CONCLUSION Sia iden 14 4 REFERENCES nn 15 APPENDIX A TESTING OFANTI ALIASING FILTER rrr 17 A 1 Experimental Set up for the Anti Aliasing Filter Test 17 A 2 Anti Aliasing Filter Responses rere nennen nnnm nnn nnns 19 List of Figures Figure T Dieitisalon Equipment set UD esuen ea 2 Figure 2 Start up window for Filter488 anne TA 3 Fig re o Filterd58 setine Wind O Wie ee 4 Figure 4 Typical Filter Response Curve with f 40kHz f 20kHz and Te LORH T 5 Fisure 5 NI Card JE exe main fon ni eo pae deeem aet usa ut ptem ma du rta Uu ette 7 Risure 6 Loading the tentado 7 PIC UES Sample data nn ee ei 9 Hetre o D rac i Data need 9 Foure Data Saved inicien iii eos 10 Figure 10 Spectrogram of a DIFAR signal using Cool Edit sss 11 Figure 11 Spectrogram of
3. 2 1 Replay of AQH 4 Tapes Data recorded during trials by the RAAF P 3C aircraft are stored on 28 track AQH4 tapes SSQ 57A SSQ 36 bathythermograph SSQ 53D DIFAR and SSQ 62B DICASS sonobuoys all have analogue electronics and are recorded on AQH4 tapes as FM signals on specific tracks during the RAAF trials see Table 1 A RACAL recorder has been set up at MOD Salisbury to replay the AQH4 tape s FM tracks Each FM track used by the RAAF has a FM card installed in the RACAL recorder This simplifies the replay of AQH4 tapes since the cards do not have to be moved around in the back of the RACAL recorder to change the tracks These FM cards are connected to the back of a BNC panel displaying the AQH4 track numbers The data can be replayed off the tapes with an analogue signal being produced for each FM track Changing tracks just requires changing the cables connected to the front of the BNC connector panels see Figure 1 These signals then need to be low pass filtered before digitisation to stop aliasing Table 1 AQH 4 Track Allocation DSTO GD 0248 Function Sensor Receiver AOHA Function Sensor Receiver Station Module Station Module Number Number Number Number 1 Pilot Headset 15 DIG SS1 5 amp 6 or ICS 2 FM SS1 2 16 FM SS1 5 3 FM SS1 3 17 DIG SS1 7 amp 8 4 FM SS1 4 18 FM SS1 5 5 FM SS1 6 19 DIG SS1 1 amp 2 6 FM SS1 7 20 FM SS1 6 7 DIG SS1 1 amp 2 21 DIG SS1 3 amp 4 8 SE
4. Chan 5 Chan 6 Chan 7 Rev 10 C Bypass C Bypass C Bypass C Bypass Filter488 Rey 8 Cutoff 1 Cutoff 1 Cutoff 1 Cutoff 1 l z Cutoff 2 Cutoff 2 Cutoff 2 C Cutoff H2 instrumentation Cutoff 2 Cutoff 2 Cutoff 2 C Cutoff 2 Output Control Immediate M On Execute Exeuute Figure3 Filter488 setting window To transmit these setting commands to the programmable filter the Output Control button has been selected in the immediate position see Figure 3 After selecting the output control to immediate the filter settings will be set and the filter is ready for use To change the filter set up select File then O pen and choose a new set up from the list Select the Output Control to immediate to set the filter to the new set up parameters 2 2 2 Selecting the cut off frequency When selecting an anti aliasing filter for use in a data acquisition application the cut off or corner frequency f of the filter is normally the highest frequency of interest in the signal being acquired see Figure 4 The attenuation of the filter at the Nyquist frequency f is what determines the level of any aliasing that appears in the signal being acquired As the upper frequency of interest is f some aliasing in the region f to f can be tolerated as long as there is sufficient attenuation in the band f f to f where f denotes the sampling frequency in the data acquisition process
5. Recon ogg sim 230 00 CI ey E O ee Ce a B Sf 4 8 oft ELE O EL EE Figure 11 Spectrogram of a Bathythermograph sonobuoy zoomed in on the frequency scale y swerlrie5tk Ddb na 40000 1 Cool Edit File E Fila Edit Mew Trenstorn Generate Anakre Ophons Help ua A acia a Oel LEN OM NE die EM am NE OR E E rur 4000716 bt Mono an T ee 12 TAN K Time 5 00 000 Figure 12 Time series of ambient noise from an SSQ 57A sonobuoy zoomed it on the amplitude scale 12 DSTO GD 0248 sw2rOrf20tk 6db40A 20000 Cool Edit m Ed EEI Transform RES Analyze Options aa E877 777 Ha P ES B i CI UC 0 6p3 47 CON AM g qp OR OO qp 3 O 5 rm eH HAS Tamm AG 104 OA 1920350 HAZ 10850 Hita 10670 Hi 115450 zoom m ow Fun 20000718 bitmono E S000000 Samples Em n Figure 13 Spectrogram of a SUS charge recorded using a 40 dB attenuated SSQ 57A Three spikes can be seen showing the shock wave and two bubble pulses To calibrate the SSQ 57 sonobuoys a rooftop checker RTC signal is recorded onto the AOHA tapes and digitised using the same equipment This signal is a 100 Hz sine wave see Figure 14 so is easily checked to see if it was digitised correctly 13 DSTO GD 0248 14 swerlrfe5tkzDdb frtc 40000 1 Cool Edit CO E Fila Edit Mew Trenstorn Generte Anev e Qphons Help IA ES ES qe m SSR Fale HEE sa828 EGER ime GG Bas bam Lam am ran
6. also monitored this input signal by connecting it to the analyser Ch 1 A input The output of the filter was monitored by connecting it to the analyser Ch 2 A input The measurement process was conducted as per the swept sine measurement process specified on pages 1 45 1 46 of the SR785 Dynamic Signal Analyser Operating Manual Table 3 specifies the start and stop frequencies used for the measurement process along with the specified filter cut off frequencies 17 DSTO GD 0248 Table 3 Settings used during the A nti Aliasing Filter M easurement Start frequency Stop frequency Filter cut off frequency Hz Hz Hz 18 DSTO GD 0248 A 2 Anti Aliasing Filter Responses The following figures are the results obtained from the swept sine measurements for the configurations listed in the Table 3 Anti aliasing filter performance fo 3000 Hz P A A re DR m PM PE EM po ee A S need A A ek ie ee EEE E BAA l l l aih 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Frequency Hz 90 Figure 16 Filter response for a filter cut off frequency of 3000 Hz Anti aliasing filter performance A 5600 Hz 90 d 0 1600 3200 4800 6400 8000 9600 11200 12800 14400 16000 Frequency Hz Figure 17 Filter responsefor a filter cut off frequency of 5600 Hz 19 DSTO GD 0248 Anti aliasing filter performance A 7100 Hz Fu So ee 3 ee A ee see
7. buffer sizes used If the time duration does not match the buffer length it will be increased to fill the last buffer There is a pre digitisation amplifier in the A D card that is set by the program The default setting is 5 volts which is the normal level required for digitising SSQ 57A sonobuoys If a different input voltage is required it can be set between 0 05 and 10 volts using the Voltage put down window The only other setting required before digitisation is the DI SE differential or single ended setting The anti aliasing filter used has a common earth so single ended is used When a number of channels are being digitised without a common earth the differential setting can be used but this reduces the number of channels to eight ref PCI E Series User Manual July 1997 The settings mentioned above could be set in any order but the Scan Rate and Duration could change if other settings are altered When all the settings are correct the AQH4 tape should be positioned to start at the section to be digitised and played When the tape is playing the Begin Sampling button should be pressed This will start the A D card going with about a one second delay The GUI will be disabled except for the Cancel Sampling button see Figure 7 w ASD aquisition HI card M v LI LI LI B M M E m LI LI al LI DT empchannell 2000 DA Temprichannel 000 DA Temprchannelz 2000 DA Temprchannel3 2000 DA Tempichannelb 2000
8. DATempichannel 000 Figure 7 Sampling data DSTO GD 0248 L an 2 T Ta Hegt sanpa C ln AS kin ele During sampling the program can be stopped by pressing the Cancel Sampling button The sampling process requires a large amount of computer resources to save the data when it digitises For this reason the screen saver or other programs that could use up computer time should be stopped The progress bar will display how much longer the sampling has to go During sampling the data is recorded in a temporary file with all the channels in one file When the sampling has finished the data needs to be unpacked from the temporary file and into the file names selected for each channel see Figure 8 This is done using a dll written in C giving a major speed improvement over Visual Basic im ASD aquisition MI card immmmmmmmicmmEmR DT emp channel 2000 DA Temp channell 000 DA Temp channel2 2000 DA Temprchannel3 2000 DA Temp channel 2000 DA Tempichannel 2000 Figure 8 Unpacking D ata Fate iilii Stop Rein DSTO GD 0248 10 After the data has been unpacked it is saved into the file names displayed as 16 bit little endian binary data The A D card only digitises at 12 bit but the data files are saved as 16 bit since this is a more commonly used format The text box then displays Data Unpacked and Saved and the files saved are displayed in green see Figure 9 These files can not be o
9. RVO 22 FM SS1 7 9 FM SS1 8 23 DIG SS1 5 amp 6 10 FM SS1 2 24 SERVO 11 FM SS1 3 25 DIG SS1 7 amp 8 12 FM SS1 1 26 FM SS1 8 13 DIG SS1 3 amp 4 27 TCG 14 FM ESM or SS2 4 28 FM SS1 1 2 2 Anti Aliasing Filter Since the data replayed from the RACAL recorder is analogue it has to be filtered before digitisation to stop aliasing An IOtech 488 eight channel programmable analogue filter bank is used as the aliasing filter The filters are 8 pole Chebyshev low pass filters with programmable cut off frequencies from 1 Hz to 50 kHz The cut off frequencies are the same for all of the eight filters and are specified using a Windows based program called Filter488 The input and output lines for the filter are connected to the BNC panels 2 2 1 Using the Filter488 program Launch the Filter488 program by double clicking on the Filter488 shortcut that appears on the window s desktop Figure 2 will then appear Filter488 Would wou like ta Open an existing setup Figure2 Start up window for Filter488 Click on the yes button and select an appropriate set up from ambient ftr bathy ftr dicass ftr strum ftr or sus ftr DSTO GD 0248 Figure 3 displays the filter settings after selecting sus ftr Filter488 File Interface Chan Contig Cutoff ee Cutoff H2 g m IL Hz 25KHz s ore FREE I Otech O Bypass O Bypass the smart Cutoff 1 Cutoff 1 Cutoff 1 Cutoff 1 approach to Window Panel
10. Signals in the band f f to f are aliased to frequencies below f so S these must be attenuated by a minimum of 50 dB DSTO GD 0248 x 30 fe j 1 In XN 60 N Js 70 M 80 w 90 NL O 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 Frequency Hz Figure4 Typical Filter Response Curve with f 40kHz f 20kHz and f 16kHz The Chebychev filters used in the programmable filter module specify an attenuation rate of 80 dB per octave The simple rule of selecting the filter cut off frequency to be 80 of the Nyquist frequency ie f 0 4 f would result in any aliased signals in the band 0 0 4 f being attenuated by a minimum of 80 log 0 6 log 0 4 47 dB which is an acceptable level of attenuation Since data obtained from SSQ 57A SSQ 36 bathythermograph SSQ 53D DIFAR and SSQ 62B DICASS sonobuoys are normally digitised at set rates see Table 2 the anti aliasing filter was characterised for each required sampling rate using the set up described in Appendix A 1 This enabled correct settings for each desired sampling frequency to be determined and applied during on going data transcription work instead of relying on the 80 dB per octave as being correct The results obtained from the measurements showed that the filters do not achieve this level of attenuation see Appendix A 2 The actual attenuat
11. Transcription Method used for P 3C AQH4 Tapes Paul A Clarke Trevor Jackson and Peter W ade M aritime O perations Division Aeronautical and M aritime Research Laboratory DSTO GD 0248 ABSTRACT Sonobuoy data is regularly obtained using the RAAF P 3C aircraft during DSTO trials This data is recorded on the aircraft using 28 track AQH4 tapes and needs to be transcribed onto CDs for storage and analysis The method used to transcribe analogue sonobuoy data was recently upgraded This document describes the equipment used and the steps involved in transcribing this data RELEASE LIMITATION Approved for public release DEPARTMENT OF DEFENCE DSTO DEFENCE SCIENCE amp TECHNOLOGY ORGANISATION Published by DSTO Aeronautical and M aritime Research Laboratory PO Box 4331 Melbourne Victoria 3001 Australia Telephone 03 9626 7000 Fax 03 9626 7999 Commonwealth of Australia 2000 AR 011 524 July 2000 APPROVED FOR PUBLIC RELEASE Transcription Method used for P 3C AQH4 Tapes Executive Summary During trials with the RAAF P 3C aircraft sonobuoy data is stored on the aircraft using 28 track AQH4 Tapes This data needs to be transcribed onto CDs for storage and analysis This transcription method for analogue sonobuoys has been improved The previous method required moving large data files between a number of different computers before it was ready to be copied onto CD and had limited quality control checking The impro
12. a Bathythermograph sonobuoy zoomed in on the frequency O RR 12 Figure 12 Time series of ambient noise from an SSQ 57A sonobuoy zoomed it on the amplit de scale eier 12 Figure 13 Spectrogram of a SUS charge recorded using a 40 dB attenuated SSQ 57A Three spikes can be seen showing the shock wave and two bubble pulses 13 Figure 14 Rooftop checker signal zoomed in on the time scale A spectrogram would show a dominant imet 100 Hd a 14 Figure 15 Experimental set up used to test the anti aliasing filter 17 Figure 16 Filter response for a filter cut off frequency of 3000 Hz esses 19 Figure 17 Filter response for a filter cut off frequency of 5600 Hz sese 19 Figure 18 Filter response for a filter cut off frequency of 7100 Hz esses 20 Figure 19 Filter response for a filter cut off frequency of 14000 Hz 20 Figure 20 Filter response for a filter cut off frequency of 15000 Hz 21 Figure 21 Filter response for a filter cut off frequency of 16000 Hz 21 Figure 22 Filter response for a filter cut off frequency of 18000 Hz 22 Figure 23 Filter response for a filter cut off frequency of 20000 HZ 22 Figure 24 Filter response for a filter cut off frequency of 22500 HZ 23 List of Tables Table Ts AOH4 Tr
13. ack Alloeaton aaa lien 3 Table 2 Recommended Filter Settings for Digitisation of Sonobuoy Signals 6 Table 3 Settings used during the Anti Aliasing Filter Measurement 18 A D BNC CD COTS DIFAR DICASS DIG EEPROM ESM FM GUI ICS PC PCI RAAF RTC SS SS2 SSQ TCG DSTO GD 0248 List of abbreviations Analogue to Digital Bayonet Nut Connector Compact Disk Commercial off the shelf Directional Frequency Analysis and Recording Directional Command Active Sonobuoy System Digital Electrically Erasable Programmable Read Only Memory Electronic Support Warfare Measures Frequency Modulated Graphical User Interface Internal Countermeasure Set Personal Computer Peripheral Component Interconnect Royal Australian Air Force rooftop checker Sensor Station 1 Sensor Station 2 Sonobuoy Time Code Generator DSTO GD 0248 1 Introduction Sonobuoy data obtained using the RAAF P 3C aircraft is recorded onto AQH4 tapes during a trial This data then needs to be transcribed to CD so it can be analysed and archived The method of transcribing analogue sonobuoy signals has been updated improving the ability to digitise multiple channels at the same time and at a higher sampling frequency The upgrade also reduced the time taken to transcribe the data to CD and improved the ability to validate the data This was done by combining the A to D card and CD writer in one PC so ti
14. are 16 FM tracks on an AQH4 tape and only eight filter channels the channel numbers on the A D card do not match the track numbers on the AQH4 tape The BNC panels are used to change which AQH4 track is recorded on which digitisation channel A 16 channel A D card was chosen so a future update to the anti aliasing filter will enable all FM tracks to be digitised simultaneously A GUI program called NI_Card_dll was written in Visual Basic to control the A D card The main window is displayed when starting the program see Figure 5 This has the 16 channels in a frame in the top left the operation buttons in the top right a frame for the type of digitisation in the bottom right and a progress bar and text box in the bottom left The text box displays the process that is occurring or that has just occurred At start up the text box should display either Calibration constants obtained from EEPROM factory defaults or Calibration constants obtained from EEPROM self calibration see Figure 5 If neither of these is displayed the calibration constants were not obtained from the A D cards EEPROM so any data obtained is not calibrated and the card should be either re calibrated or repaired DSTO GD 0248 First choose the channels required and the names of files in which data is to be saved Each channel has a tick box which tells the program which channels you want digitised When a channel is chosen you can add the file name by either clicking th
15. e button or the file name display area This will open a standard save file name window The file name chosen will be displayed after OK is pressed see Figure 6 A D aquisition NI card Figure5 NI Card dll exemain form im A D aquisition MI card D Temp schannell 2000 D Temp schannel 2000 Temp schannelz 2000 T P D Temp channel 2000 g ea eE 0 47 emp channel6 2000 opaca DA Tempichannel 000 VU m m m m st DU b Kia Figure 6 Loading the file names A tick box can be unselected if you no longer want that channel digitised If the file name has already been chosen it will be greyed out and not used see Figure 6 DSTO GD 0248 Next the Scan Rate should be chosen The Scan Rate is the digitisation rate for each channel not the combined rate ie if you want 4 channels digitised at 20kHz each the Scan Rate is 20000 while the combined rate would be 80000 The A D card is only capable of handling a combined digitisation rate of 1 25 MHz but this high speed can quickly overflow the card buffer so a max speed of 1 0 MHz has been set in the program The A D card can digitise only at set rates depending on the internal clocks in the card These set rates have been put in the program so if you select a rate the program will select the closest rate possible and display it in the scan rate text box The Duration of the digitisation can now be set This has set intervals that can be used due to the
16. ences John E Savage 1991 1992 Basic Introduction to Air ASW Acoustic Systems Mission Avionics Technology Department Naval Air Development Center Warminster PA USA PCI E Series User Manual July 1997 National Instruments Corporation Austin TX 78730 5039 USA Part Number 320945C 01 SR785 Dynamic Signal Analyser January 1998 Operating Manual and Programming Reference Stanford Research Systems Sunnyvale California 94098 USA 15 DSTO GD 0248 16 DSTO GD 0248 Appendix A Testing of Anti Aliasing Filter A 1 Experimental Set up for the Anti Aliasing Filter T est To test the performance of the anti aliasing filters the Stanford Research Systems SR 785 Dynamic Signal Analyser was used to generate a test signal which was fed through the programmable filter The generated test signal and the filter output were both monitored by the signal analyser and the filter response was obtained by process of direct comparison of the filter input and output Figure 15 shows the experimental set up all SR 785 Test Signal Filter IOTech 488 Figure 15 Experimental set up used to test the anti aliasing filter The analyser was set up to perform a swept sine measurement In this configuration the analyser provided a sweeping sine wave source that swept a sine wave between specified start and stop frequencies This swept sine signal was fed from the analyser source output into the input of the filter The analyser
17. f Chief Maritime Operations Division 15 SECONDARY RELEASE STATEMENT OF THIS DOCUMENT Approved for public release OVERSEAS ENQUIRIES OUTSIDE STATED LIMITATIONS SHOULD BE REFERRED THROUGH DOCUMENT EXCHANGE PO BOX 1500 SALISBURY SA 5108 16 DELIBERATE ANNOUNCEMENT No Limitations 17 CASUAL ANNOUNCEMENT 18 DEFTEST DESCRIPTORS Data analysis Computer programming Data processing Acoustic data Sonar signal processing 19 ABSIRACT Sonobuoy data is regularly obtained using the RAAF P 3C aircraft during DSTO trials This data is recorded on the aircraft using 28 track AQH4 tapes and needs to be transcribed onto CDs for storage and analysis The method used to transcribe analogue sonobuoy data was recently upgraded This document describes the equipment used and the steps involved in transcribing this data Page classification UNCLASSIFIED
18. formance la 20000 Hz 90 O 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 Frequency Hz Figure 23 Filter response for a filter cut off frequency of 20000 Hz 22 20 30 40 50 60 70 80 90 O 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 Anti aliasing filter performance f 22500 Hz nu o qo 2 2 qv HT 3S e GRO cw oe er ie co cO dq Q9 9 x0 ID ERO 9 0 ee ee ee ex x9 C exe que ee Xp a Frequency Hz Figure 24 Filter responsefor a filter cut off frequency of 22500 Hz DSTO GD 0248 23 DSTO GD 0248 24 Page classification UNCLASSIFIED DEFENCE SCIENCE AND TECHNOLOGY ORGANISATION a OF DOCUMENT 2 TITLE 3 SECURITY CLASSIFICATION FOR UNCLASSIFIED REPORTS THAT ARE LIMITED Transcription Method used for P 3C AQH4 Tapes RELEASE USE L NEXT TO DOCUMENT CLASSIFICATION Document U Title U Abstract U 4 AUTHOR S 5 CORPORATE AUTHOR Paul A Clarke Trevor Jackson and Peter Wade Aeronautical and Maritime Research Laboratory PO Box 4331 Melbourne Vic 3001 Australia 6a DSTO NUMBER 6b AR NUMBER 6c TYPE OF REPORT 7 DOCUMENT DSTO GD 0248 AR 011 524 General Document DATE July 2000 8 FILE NUMBER 9 TASK NUMBER 10 TASK SPONSOR 11 NO OF PAGES 12 NO OF 3 13 URL 14 RELEASE AUTHORITY http www dsto defence gov au corporate reports DSTO GD 0248 pd
19. hniques used to get the geoacoustic sediment parameters and target strength measurement Trevor Jackson Maritime Operations Division Trevor Jackson is a Senior Professional Officer in Maritime O perations Division at DSTO Salisbury Hejoined DSTO in 1988 after completing a Bachelor of Engineering in Electronics Engineering at the South Australian Institute of Technology At DSTO Trevor has been responsible for hardware and software engineering developments associated with complex trials and data acquisition and analysis systems for various sonar systems including systems in service with the RAN He has also activel y participated In a numerous trials and been involved with the analysis of data collected during trials Trevor is currently the head of a section responsible for providing data acquisition and analysis services to M OD Salisbury Peter Wade Maritime Operations Division Peer Wade is a Technical Officer Levd 3 in the Maritime O perations Division at DSTO Salisbury He joined DSTO then WRE in 1977 as an apprentice electrical fitter After completing his apprentice ship and an Associate Diploma in Electrical Engineering he worked as an electro mechanical draftsperson until 1996 when he became involved in underwater acoustic trials and data analysis Currently studying for a Degree in Computer Science he currently performs most of the data transcription associated with acoustic trials performed by M aritime O perations D ivision
20. ion rate was approximately 66 dB per octave from f to 2f with an attenuation of about 42 dB at 0 6f This would normally not be sufficient attenuation but since pre whitened sonobuoy data is being digitised with a dynamic range of about 30 dB this level of attenuation is satisfactory Table 2 shows the standard digitisation rates and filter settings used for digitising sonobuoy data DSTO GD 0248 Table 2 Recommended Filter Settings for Digitisation of Sonobuoy Signals Sampling Rate Hz Filter cut off frequency Signal type and application Hz 10000 4000 Bathythermograph buoy 20000 8000 SSQ 57A amp SSQ 57B sonobuoys for SUS runs and ambient noise ambient noise analysis DICASS analysis sonobuoys Table 2 has the standard f 0 4f for all sampling rates except 50 kHz At the 50 kHz sampling rate the filter is set to 22 5 kHz which is 0 45 f this is used since the SSQ 57A DICASS DIFAR and STRUM sonobuoys already have a designed roll off below this frequency The SSQ 57B sonobuoy signal does not contain this roll off but is filtered at the same rate for convenience The SSQ 57B sonobuoy data should only be used up to 20 kHz any higher data could have aliased data in it see Figure 24 2 3 Digitisation of AQH4FM Tracks The digitisation is done using a National Instruments PCI MIO 16E 1 16 channel 1 25 MHz 12 bit A D card installed in the PC and a BNC 2090 panel next to the other BNC panels Since there
21. le ee ea ee Cam dS ee eat ee SS Be SS Sh a0 amp ee Se Sb aa SS OS Sok SS aa ENR SS II O m m ox 90 i A O 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 Frequency Hz Figure 18 Filter responsefor a filter cut off frequency of 7100 Hz Anti aliasing filter performance f 14000 Hz 90 ad O 4000 8000 12000 16000 20000 24000 28000 32000 36000 40000 Frequency Hz Figure 19 Filter response for a filter cut off frequency of 14000 Hz 20 Anti aliasing filter performance f 15000 Hz 90 O 4000 8000 12000 16000 20000 24000 28000 32000 36000 40000 Frequency Hz Figure 20 Filter response for a filter cut off frequency of 15000 Hz Anti aliasing filter performance f 16000 Hz 20 30 40 50 60 70 80 O 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 90 Frequency Hz Figure 21 Filter responsefor a filter cut off frequency of 16000 Hz DSTO GD 0248 21 DSTO GD 0248 Anti aliasing filter performance f 18000 Hz zi ne Sage ee oe ee Rede a eor efe ee a re SURE ne ME E OET Ze SEE EE A A A RE ee pe Sees tes sl alia ae EEE a a 90 O 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 Frequency Hz Figure 22 Filter response for a filter cut off frequency of 18000 Hz Anti aliasing filter per
22. me is not wasted transferring the data between machines and off the shelf software could be used to check the data As part of the upgrade a Graphical User Interface GUI was written to simplify the digitisation process This report describes the new method used to transcribe and validate the data 2 The Transcription Process The transcription of FM data from AQH4 tape to CD has a number of steps 1 The FM tracks of the AQH4 tapes are replayed using a RACAL STOREHORSE recorder Each FM track produces an analogue data channel 2 The analogue data is then low pass filtered using a multi channel programmable filter to remove aliasing during the digitisation The current filter has eight channels so a maximum of eight channels can be digitised at the current time 3 The data is digitised and saved as a separate file for each AQH4 track of interest on the PC s hard disk 4 These files are then checked for errors by audio replay and displaying the data graphically using Cool Edit 5 The data is written onto CD s for analysis and storage The RACAL recorder filter and PC have been installed in racks and connected to a number of BNC connector panels see Figure 1 This enables the different AQH4 tracks to be easily digitised A second RACAL recorder also has been placed in the racks to enable duplication of tapes but this is only possible for the FM tracks see Table 1 DSTO GD 0248 Figure 1 Digitisation Equipment Set up
23. ns a tone that is dependent on the temperature in the water see Figure 11 e The SSQ 57 sonobuoy is omnidirectional so the signal looks similar to white noise see Figure 12 The standard SSQ 57 and 20 dB attenuated SSQ 57A sonobuoy data can have some structure due to shrimp fish and shipping noise Listening to these sonobuoys can quickly verify if the sonobuoy is working 40 DSTO GD 0248 60 and 80 dB attenuated SSQ 57A sonobuoys tend to contain mostly electrical noise unless there is a strong signal source in the water e Any SUS or DICASS signals in the data can also be used verify if the sonobuoy is working see Figure 13 sw2rlrf44k04difB 50000 1 Cool Edit File Edit Wew Tronstorn Generiese Anse Options Help u sa ee Laia is 7 alil LEX 1e 14070 105001 OX Miri TES u T u i m i T i 1 40 T i E i ZI 4 mm i 120 l 34b 200 4 zu i m zoom m somortebirMono Beg 000 000 Ent 3 15000000 Gamples End IA Te A i AAA A AA CC CC i IA A B 25 c E lt O mg a ZI B4 Ei El 15 12 4 Figure 10 Spectrogram of a DIFAR signal using Cool Edit 11 DSTO GD 0248 swari rfl Zik Abt 10000 1 Cool Edit Oi E Fila Edit Mew Trenstorn Generate Anakre Ophons Help muje Ao ace a SSCP ad Co ee IESESESENI E NEI E bm ola on om ow ob io en ie em iw ie o ala Tai EI IT RE Ber 0000M 1509000 94 les End Lupum E Preuss
24. vement required the purchase of an analogue to digital A D converter card and a number of FM cards for the AQH4 tape recorder Transcription of data from the AQH4 tapes onto CDs now involves the following steps Replay of the AQH4 Tape Anti alias filtering of the analogue signal Digitisation of the analogue signal Quality checking the data files Writing the data onto CD for storage and analysis This new method now has the digitisation validation and storage processes in one PC thus simplifying the transcription process In addition a commercial off the shelf COTS software program called Cool Edit was purchased to check the data files This has improved the validation process since it is user friendly fast and has a good Graphic User Interface GUI to check the data quality This document describes the new transcription process the required filter settings and the techniques for using the A D card software Authors Paul Clarke Maritime Operations Division Paul Clarke is a Professional Officer Class 2 in the Maritime Operations Division at DSTO Salisbury Hejoined DSTO in mid 1995 after completing a Bachelor of Science degree in Oceanography and Meteorology at Flinders University and a Bachelor of Engineering degree in M echanical Engineering from the University of South Australia At DSTO Paul has worked on underwater acoustics including work on the effect the environment has on acoustic propagation various inversion tec
25. ver written until some change is made to the settings This stops you accidentally pressing the Begin Sampling button clearing the data already obtained im ASD aquisition MI card B x Iv Ehamel0 n T emptchannel0 2000 Iv FEhanne in Temp channel1 2000 J Channel 2 0 Tempichannel2 2000 u Channel 3 In Tempschannel3 2000 P Lance Samplifa Channel 4 m Channel 5 u Iv FEhannelB 0 Tempichannel 2000 u Stop l rjnieiri tiri Iv FEhannel D T emp channel7 2000 Channel E Channel 3 n E xit Channel 10 E Channel 11 u Channel 12 20000 Scan Rate Hz 0 10 Duration MMM S5 Channel 13 25 V Voltage Channel 14 Data Unpacked and Saved se DI SE Channel 15 Figure 9 Data saved 2 4 Validation of Data Digitised The digitised data is validated using Cool Edit Cool Edit enables the data to be viewed as a time series or spectrogram with an ability to zoom in on specific sections of the data The data can also be listened to on speakers connected to the PC These combined features enable the data to be quickly checked The different types of sonobuoys have specific features which can be used to verify if the sonobuoy was functioning correctly Some examples are e The DIFAR and DICASS sonobuoys have two tones in the recorded signal at 7 5 kHz and 15 kHz ref J Savage 1991 These can be seen on a spectrogram see Figure 10 e The SSQ 36 bathythermograph retur
Download Pdf Manuals
Related Search