|Back to Bulletins index||Back to INAG-60 Contents page|
9. COMPUTER CONTROL OF AN IPS-42 IONOSONDE
J.E. Titheridge, Physics Department,
University of Auckland, New Zealand.
9.1. The Hardware
Our IPS-42 ionosonde, purchased from KEL Aerospace in 1983, continues to give yeoman service. On average we lose 0.1% of possible recording time through ionosonde failures and servicing, and 1.2% through camera, film and processing problems. Six years ago we began work on a computer-controlled system to collect the ionograms in digital form, and to store, display and scale these as required. This project was initially quite ambitious, with facilities for specifying the frequencies to be sounded and the number of pulses transmitted. We have now settled on a simplified version for routine operation, as outlined below.
The IPS-42 sounds 3 times on each frequency. Echoes are detected by a voltage comparator, and clocked into a 1024--bit shift register which stores data from the first 800 km of effective height. The data is re-circulated during the 2nd and 3rd soundings, on the same frequency, to zero any bits which do not correspond to a consistent return. The result is then passed out to the video display -- at which point we grab it along with the associated clock signal, as indicated at the top left of Fig. 1. We also extract Scan-on, transmitter-on (Xmr) and display-on (Yb) signals from inside the IPS-42. All these, plus a Min signal, are accessed by clipping leads on to existing edge connectors. One further wire is soldered to the "Monitor Sweep" button, so that the computer can `push' this as required.
The signals are processed on a small printed-circuit board installed at the rear of the ionosonde. The three control and timing signals are combined into a single control line (Y), which is sent to the computer in true and inverted form. At the computer, which can be 1--20 metres away, differential processing reconstitutes the Y signal free of interference. Incoming clock and control signals are also processed to remove any rapid fluctuations. The serial data stream is converted to parallel form and fed to the computer as 64 8-bit bytes at each frequency (the top right of Fig. 1). The software also carries out numerous error checks, and disables the receiving system (using the "inhibit" line) during transmission periods. As a result we get perfect records at all times, except during power failures; and recording resumes automatically as soon as power is restored.
9.2. The Software
When the computer is switched on it connects to the ionosonde, and waits for a one-minute time signal. This is used to readjust the computer's clock, over a range of ñ30 sec, to agree with the ionosonde. (Synchronisation is also carried out every hour throughout the recording interval, to maintain accurate timing.) The main recording routine is then entered to obtain ionograms at specified intervals. These appear on the monitor display and on the computer screen. Any film program can proceed independently; e.g. you could collect digital ionograms every 5 minutes, and film ionograms every half hour. The last ionogram is always displayed on the computer, and is available in a separate file for remote retrieval and viewing if required.
Each ionogram is stored with header, date and time information. The height resolution of 1.6 km is about twice that available on film, or from KEL's DBD-43 unit, and requires 37 kB to store each ionogram in its raw state. This is reduced by a factor of about 5 by compacting the data with a type of run-length encoding, designed to match the data. A further factor of 2 is obtained if the ionograms are first "cleaned" by deleting the date-time numerals, all information below 70 km, all of the graticules (apart from 4 reference marks), and most isolated dots. The graticules and date-time information are readily restored from the ionogram header when required, as shown in Fig. 2.
Hourly ionograms are left uncleaned to provide a full check on ionosonde operation. For 5 minute ionograms this processing reduces the average size to typically about 3.5 kB, so that the 288 ionograms for one day occupy about 1.0 MB. Every five days a standard data-compression program reduces the files by a further factor of about 1.8, and copies them to tape. Thus the final storage requirements for high-resolution ionograms, recorded every 5 mins, is about 15 to 20 MB per month.
A separate software package is used for display and scaling of the stored data. Arrow keys give a fast forward or backward scan of successive ionograms, at rates of 2 or 3 per second. Other keys scan the data in steps of 30 minute or two hours; jump to the first or last ionogram of this day (or adjacent days); or jump immediately to the ionogram at the same time on preceding or following days. This last facility is invaluable for examining strange effects near sunrise or sunset.
Scaling of data is carried out using a computer aided approach, giving fast, accurate and fully checked results. The computer displays, as coloured lines, the critical frequencies and minimum heights scaled from the previous ionogram. These are adjusted as required using the cursor keys. Tapping `return' then saves these values and displays the next ionogram.
A `site' number is included in the header information for each file. This is used to identify the station, and to select the correct value of gyrofrequency (for each layer) so that fo and fx are both displayed with the correct separation. This greatly increases the ease and accuracy with which critical frequencies can be determined, as in Fig. 2 where foF2 is defined best by fx (broken line). Height lines are shown at h, 2h and 3h, for checking against multiple echoes. Further
developments will involve interactive real-height calculations using POLAN, to display the ionogram, the scaled data, the calculated profile and the corresponding (calculated) virtual heights.
Virtual--height variations at a fixed frequency can also be readily extracted from digital ionograms. A program VARION plots echo height against time for any required frequencies, for periods of 1 to 20 days. Echoes are displayed only when there is a return on at least 2 of the 3 channels closest to each given frequency, to reduce noise and avoid problems caused by blanking of individual channels. For greater clarity a leading-edge detection algorithm can be invoked. An example of this, for echoes in the height range 100 to 180 km over a period of 16 days, is shown in Fig. 3. Two of these plots cover one month, giving an excellent summary of the occurrence and characteristics of sporadic E.
An obsolete XT computer with a 40 MB hard disk is ideal for running the Auckland `Digion' system; this will allow storage of all programs plus 5-min ionograms for about six weeks. Such computers can be obtained for less than US$250, or $500 with a new streaming tape drive for long-term storage. We can supply copies of our system (with plug-in printed circuit boards for the computer and the ionosonde, plus the cable and all programs) for about US$1,600. Data is stored in a simple but efficient format, and source code is provided for all programs so that a user can get whatever he wants from the data. The advantages of changing to computer control have proved compelling, and may be summarised as:
Þ Recording-- No film to buy, expose, process and store. Make copies in your computer.
Þ Viewing-- Scan data rapidly at your desk. Check quickly against adjacent ionograms, or at the same time on other days.
Þ Scaling-- More detail and resolution than with film ionograms, and the position of each dot is known exactly (with no calibration). The display of fx calculated from fo, and lines at h, 2h and 3h, allows the use of all information.
Þ Savings-- To store 5 minute ionograms costs about $50 per year. Two minute ionograms would cost another $1 per week.
Þ New Uses-- Plot h'(t) at any number of frequencies for a clear display of ionospheric changes, TID's and gravity waves. Do interactive real-height calculations.
Back to Bulletins index Back to INAG-60 Contents page