Data Visualisation for DSTO Over-the-Horizon Radar R&D
Written and submitted by Rod Barnes (Editor), Don Sinnott (Chief SSD), Stuart Anderson, Marina Ozerova, Ben White, Mike Wilson, Trevor Harris and Andrew Cool.
Enhancing Australia's Defence Capabilities
DSTO research assists the Australian Defence Force by
- Investigating the use of future technologies for defence applications
- Ensuring that Australia is a smart buyer of defence equipment
- Developing new defence capabilities
- Enhancing existing capabilities by increasing performance and safety, and reducing costs
DSTO has a central office in Canberra and two laboratories, the Aeronautical and Maritime Research Laboratory in Melbourne and the Electronics and Surveillance Research Laboratory in Salisbury, outside Adelaide. Research facilities linked with these laboratories are located throughout Australia.
DSTO has been assisting the defence of Australia since 1910. Early inventions include the black box flight recorder, now used around the world. There have also been many developmental projects, including the launch of Australia’s first satellite.
DSTO works in fields of science as disparate as psychology and the latest in ballistics protection, looking at everything from insect navigation to submarine combat systems. The Australian Defence Force initiates most of DSTO’s research. However, DSTO’s researchers also forecast trends in science and anticipate military needs.
More information on DSTO can be found at www.dsto.defence.gov.au.
Introduction to Over the Horizon Radar: Don Sinnott
Surveilance Systems Division within DSTO has responsibility for R&D in Over-the-horizon radar (OTHR). OTHR is an application of radar technology which exploits the phenomenon of electromagnetic wave reflection by the Earth’s ionosphere to allow the operation of radar beyond the local horizon. The ionosphere is a solar-induced shell of ionisation that exists in the upper atmosphere. Electromagnetic waves may interact with the ionosphere so that, within a range of frequencies, the waves are reflected to earth. An area or object illuminated by this electromagnetic wave returns part of its scattered energy by an (almost) reciprocal ionospherically reflected path and this signal, when received and processed, allows radar operation.
As in many other forms of radar: a suitably designed waveform is modulated onto a transmitted carrier wave and processing of the returned echoes allows definition of both group range (by determination of elapsed time) and direction (by use of directional antennas) of the region under radar surveillance. As pointed out below, a third dimension, that of Doppler, is available to the OTHR operator and is a crucial parameter in deriving information from the radar returns. In particular, Doppler information allows discrimination of moving targets from stationary (or slowly moving) clutter. (Clutter is the term used in radar parlance to describe returns from extended areas, and is normally seen as contamination to be minimised, in favour of the discrete target return, by processing. Of course in some surface-mapping OTHR applications this clutter is the object of the surveillance operation.)
Operation in the High Frequency (HF) band implies wavelengths of between 10 and 100m. It follows that transmitting and receiving systems need to be physically large in order to realise adequate directive properties, which derive from multi-wavelength apertures. Acceptable angular resolution demands antenna systems of hundreds or thousands of metres in length. The typical OTHR thus has very large, ground based array antenna systems with electronic steering, making it appear physically quite unlike conventional radar, Figures 1-3. Another point of departure from conventional radars is that most OTHR implementations use bistatic operation, whereas most conventional radars use a common antenna in which operation is monostatic (the single antenna switches between transmit and receive functions). Many OTHR systems use continuous transmission of a repetitive waveform and hence need spatially separated transmit and receive antenna sites. The arrays must be separated by sufficient distance (typically 60-100Km) so that the direct leakage to the receiving system of the transmitted signal (by line-of-sight or ground-wave propagation) does not interfere with reception. A cost-attractive architecture is a relatively compact transmitting antenna array (100 to 200m) and a receiving array about an order of magnitude larger. Figure 1 shows the transmitter array for the Jindalee Over-the-horizon radar network (JORN) in Western Australia. Figure 2 and 3 are pictures of the receiver site.
As well as being constrained to the broad frequency band necessary for ionospheric propagation, an OTHR must also allow for the fact that the ionosphere is a highly variable and unpredictable medium. For acceptable OTHR performance, continuous real-time ionospheric assessment (sounding) is necessary, as is the ability for the radar system to adapt to the changes. The day-night variability of the ionosphere requires, typically, a 5:2 change in frequency for comparable operational range but imposed on this diurnal ionospheric cycle are both longer term sunspot-related effects and short-term disturbances.
Antenna pattern modelling: Marina Ozerova and Ben White
As the introductory pictures suggest, antennas are an integral component of OTHR systems. Understanding antennas and the radiation patterns they generate is an important design issue. Antennas provide directional selectivity, which is used to estimate target positions. However the selectivity can also be used for a host of other reasons including the elimination of interference. Another key factor is the electrical gain that is produced by the coherent integration across an antenna aperture. A COTS data visualisation package has assisted greatly in the visualisation of the patterns of radiation that certain antennas radiate. Figure 4 shows the results of antenna modelling on a long wire dipole. Note that the pattern is very dependent on the frequency of transmission. Higher frequencies have shorter wavelengths and the patterns become accordingly more complicated because the current source has greater phase variation along the length of the antenna.
Radar scattering phenomenology in the HF band: Stuart Anderson
As with any radar, the ability of the OTHR sensor to derive information about entities in its field of view is dependent on the interaction of the probe signals – in this case HF electromagnetic waves – with those entities. In most HF radar applications, the interaction may be viewed as a localised electromagnetic scattering process, with the incident electromagnetic field taken to be a plane wave. As in other branches of radar, this scattering process is summarised by describing an object in terms of its radar cross section (RCS): the area for which the power from the incident plane wave would be sufficient to produce, by omnidirectional radiation, the backscattered power density observed.
OTHR wavelengths are sufficiently large that most objects of interest are comparable with or smaller than a wavelength. Scattering by conductors comparable in size to a wavelength is dominated by resonances in which the radar cross section can be a sensitive function of frequency. This means that the strength of the radar return is a poor guide to the size of an object – a wire resonant at the OTH radar’s carrier frequency can appear to the radar as an echo comparable with that of a large aircraft. For conductors much smaller than a wavelength, in the Rayleigh scattering regime, the radar cross section varies inversely as the fourth power of the illuminating wavelength .
Calculation of HF radar cross sections of metallic targets, such as aircraft and ships, can reasonably assume that the bodies are perfectly conducting. In this case there are many established analytical and numerical methods which can be used to compute the scattered field. Perhaps the most widely used technique is the Method of Moments for which there are many computer code implementations in general use. An example of such calculations is presented in Figure 6 which shows the squared magnitudes of the elements of the scattering matrix of a Boeing 727 aircraft, in free space, as defined in the V-H polarisation basis, at a radar frequency of 15MHz.
Geophysical Plots: Trevor Harris
The Earth’s ionosphere, a partial magneto-plasma, is produced by photons and particles that are emitted from the sun and which ionize a portion of the neutral upper atmosphere. The lowest altitude at which ionization is encountered is about 60km and it extends from there to many earth radii. The region between 90km and 400km is the most important for OTHR, as it is at these altitudes that electromagnetic wave reflection is significant. In this region, the bulk of the ionization is caused by the action of extreme UV and X-rays on atoms and molecules of oxygen and nitrogen.
The varying frequency bands of the sun’s emission and the physical chemistry of the atmosphere are jointly responsible for relatively distinct ionospheric regions or layers. Those most significant for OTHR operation are designated the D, E and F regions. The D-region does not provide useful reflection of HF waves but is associated with absorption. The E-region occurs from 90-150km with an ionisation peak around 105km and provides useful OTHR propagation during the day to about 1800km. Above this lies the F-region, within which there are frequently separate layers. The most important layer for OTHR, because it is the most intense layer in the ionosphere, is designated the F2 layer and peaks at around 300km. An important feature of the F2 layer is that it is sustained throughout the night by ion transport and the low ionisation recombination rate. The ability of a layer to reflect radio waves is related to its electron density, which defines the layer plasma frequency. Figure 7 shows the geographic variation of F2 layer plasma frequency from an ionospheric model used in SSD called FIRIC. Overlays include the day-night terminator (orange line), plasma frequency contours (blue lines) and the magnetic latitude (green lines). The plot has a set of bulletin board style options. Different options can be scrolled by the arrows on the left hand side of the control.
Oblique Propagation: Trevor Harris
A comprehensive knowledge of ionospheric propagation is required for military operation of OTHR and HF communication systems. One method of increasing understanding is making swept frequency measurements over fixed one way paths for a variety of geographic locations. The oblique ionogram is an image of the amplitude received as a function of frequency and delay over the path. It requires precise timing to be maintained between the two sites involved. Variation of propagation over oblique paths is often investigated by determining the variation of certain points on the ionogram. For instance the maximum observed frequency (MOF) can be collated, analysed and displayed. In Figure 8, the variation of ionogram propagation over a month, as a function of time of the day, is displayed by building up a density picture. The density value is generated by overlaying all the ionograms for that time in the month, and accumulating the occurrence statistics in each frequency-range pixel. This plot gives a feel for the ionospheric paths and their properties in range and frequency that are likely to occur at various times.
Once again a variety of widgets provide control over the display. A combination of drop down menus, slider and button widgets provide processing and display options. Arrow button widgets provide a recorder style temporal control.
Real time monitoring of the ionosphere for OTHR: Mike Wilson
The backscatter sounder is another swept frequency ionospheric sounding device. It measures the received power for a backscattered two way path. A large database of ionograms has been collected over almost 15 years from the Jindalee project. The measurements show a variety of forms due to ionospheric dynamics, Figures 9-11 are examples. Backscatter ionograms display the power of the received signal as a function of the group range and the transmitted radio frequency. Strong backscatter, which is colour coded red, corresponds to good propagation at that range and frequency. Thus, an ionogram provides a snapshot of the ionosphere’s ability to support radio wave propagation.
Sunrise in the Ionosphere: Mike Wilson
Jindalee backscatter ionograms are usually recorded for each of 8 beams, which cover a 90 degree azimuth sector. In a special experiment an ionogram was recorded for each of 32 beams so that the full 360 degree azimuth range could be monitored. 32 ionograms from one run of the experiment were processed to produce Figure 12. The figure shows a map of the highest propagating frequency at about 7am local time. High propagation frequencies, which are colour-coded red, correspond to a stronger ionosphere in that direction away from Alice Springs. The figure shows that radio wave propagation to the east is much stronger than propagation to the west. Sunlight ionises the upper atmosphere, which strengthens the ionosphere to the east. To the west, however, the depleted night-time ionosphere shows much reduced propagation. In effect this figure shows sunrise in the ionosphere.
Ionospheric sporadic-E layer: Rod Barnes
Sporadic-E (Es), is a thin, erratically occurring layer that forms in the earth’s ionosphere, and is most prevalent between 90 and 120 km altitude. Its behaviour is strongly controlled by the orientation of the magnetic field and its production mechanism is quite different at mid, low and high latitudes. At mid-latitudes, Es is chiefly a summer daytime phenomenon. Es, which can be the most intense layer in the ionosphere, has several important effects on OTHR. First, it can provide very good propagation for surface-mode detection, as it imparts very little Doppler distortion. Second, it can obscure propagation to the higher layers, which has the negative impact of limiting the OTHR maximum range to around 2000km. Third, because of its erratic temporal and spatial nature, Es is very hard to model and consequently can play havoc with the accuracy of an OTHR. Figure 14 shows an image of Es which was measured with the Jindalee radar. The radar provides very high resolution HF measurements of the structure. Animation of the layer image, easily achieved with a COTS data visualisation package, has shown how the patches of Es move with time. In some cases they drift without changing their shape, but on most occasions they form and dissipate in the one spot.