Introduction
Wide-area surveillance of Australia's northern approaches prevailed as a strategic defense priority following World War II due to renewed national concerns about Australia's vulnerability to air attack. Attention focussed on a vast region known to as the “sea–air gap” that separates the sparsely populated Australian mainland to the north and northwest from the closest nonsovereign land mass. The impracticalities of realizing any kind of comprehensive wide-area surveillance capability well-matched to Australian defense requirements using existing line-of-sight (LOS) radar technology available at the time posed a seemingly intractable problem for a geographically large nation of relatively small population.
Early Australian exposure to British radar developments and the successive migration to Australia of British radar experts following World War II created a vibrant Australian radar research community in the 1950s. As defense strategists mulled over the surveillance challenge, a small team of far-sighted scientists in an Australian government laboratory that is now DSTG began to explore the potential of high frequency (HF) radar exploiting reflection from the ionosphere to deliver a highly cost-effective capability for wide-area surveillance linked to the defense requirement. This small team went on to lead the development of the first Australian over-the-horizon radar (OTHR) demonstrator systems designated as Jindalee that have since been evolved by three generations of professionals in government, industry, and academia to deliver Australia's leading-edge JORN capability.
The Australian OTHR story begins in the 1960s when there was not a lot of high-level support for OTHR in senior defense circles. Cooperation with the United States was essential in providing exposure to U.S. OTHR technological development to that time and equally significant in building credibility within Australia's strategic Defense community. Australia's first OTHR project was established due to the great personal resolve of its principal proponent to overcome official skepticism and have the Jindalee developmental program funded through a series of phases in the 1970s. Following the successful completion of this developmental program in the 1980s, the Jindalee Facility Alice Springs (JFAS) OTHR was formally transitioned to the Royal Australian Air Force (RAAF) as an operational system in the early 1990s. The intent to develop an OTHR network was announced by the Australian government in 1986 and the three-station Australian system termed JORN was commissioned on 2 April 2003.
OTHR is still viewed today as a mixture of technology and “magic.” JORN can detect and track aircraft and ships over an area of approximately 15 million square kilometers, which is about twice the total land area of Australia. While Australian OTHR systems were not the first to be built internationally, a crowning achievement is that they are the first OTHR systems to be adopted as part of defense operational capability. JORN stands out as a prime example of how scientific innovation can be transitioned to meet a strategic defense requirement vital to securing a nation's borders. JORN can claim to be the most advanced OTHR system in the world and no small part of the credit for this goes to Australian defense science smarts. It has been said that one day satellite surveillance would make JORN obsolete, but that prediction has not yet come to pass. JORN is currently undergoing a substantial upgrade program that will extend the network's operational life to beyond 2040.
OTHR development in Australia is well captured in several documents including [1], [2], [3], [4], [5]. The motivation for this article is to enhance the visibility of the Australian OTHR program within the AESS community and to reflect on what is one of Australia's most impressive technological feats relevant to celebrating the Society's 50th anniversary. This article is structured as follows. The “Fundamental Principles” section reviews the operational concept and key characteristics of OTHR systems. The “Historical Perspective” section provides a brief chronological account of OTHR development in Australia. The “Jindalee Operational Radar Network” section discusses existing and planned technology upgrades in the three most recent phases of JORN. The “Reflections and Prospects” section shares insights on the evolution of OTHR in Australia and future opportunities.
Fundamental Principles
Skywave OTHR operates in the HF band (3–30 MHz) and exploits signal reflection from the ionosphere over a two-way propagation path to detect and track air or surface targets at ranges an order of magnitude greater than is possible with a ground-based microwave radar. This section briefly reviews the fundamental principles, operational characteristics, and indicative capabilities of a nominal OTHR system. A more detailed coverage of OTHR systems may be found in [6] and [7].
Operational Concept
Figure 1 illustrates the OTHR principle of operation. A transmitter radiates the HF radar signal as a directional beam toward the ionosphere at oblique incidence. At one or more heights in the ionosphere, the signal is reflected and propagates downward to illuminate the Earth's surface and airspace above it well beyond the horizon. This down-looking geometry eliminates shadow regions such that targets cannot exploit mountains or valleys to deliberately avoid illumination.
OTHR principle of operation illustrated for E-layer propagation. The range coverage of 1000–3000 km includes F-layer propagation.
Targets illuminated by the transmitter scatter the incident signal and a small fraction of each target echo propagates back to a highly sensitive receiver. The receiver acquires target echoes superimposed with clutter backscattered from the Earth's surface, interference, and noise. Signal processing is used to detect target echoes against powerful disturbance signals and to estimate target parameters including range, azimuth, and relative velocity. Doppler processing is essential in OTHR to discriminate moving target echoes that compete for detection with much stronger clutter returns. A real-time air and surface surveillance picture of confirmed target tracks on a geographic display is the final OTHR output.
Skywave Propagation
The ionosphere is a region of the Earth's upper atmosphere where the density of naturally occurring plasma created by the sun's radiation affects the propagation of radio waves. The electron density height profile in the ionosphere varies with time and location due to the complex interaction of plasma production, loss and transport processes that govern ionospheric structure, and its morphology [8]. Unlike microwave signals that penetrate through the ionosphere with negligible impact on propagation, HF signals may experience absorption, refraction, and reflection at heights roughly between 60 and 600 km.
Figure 2 illustrates a notional electron density height profile of a hypothetical day-time mid-latitude ionosphere. The net effect of an atmosphere with varying concentrations of different chemical species as a function of height and a spectrum of solar ionizing radiation (spectral lines and continuum radiation) is an electron density height profile that exhibits local maxima in regions known as the D, E, and F regions. Despite there not being any direct solar radiation at night, the ionosphere maintains sufficient ionization at night to enable continuous OTHR operation.
Notional electron density height profile. Key spectral components of solar radiation, ionized chemical species, and layer height regions are shown.
Figure 3 traces the propagation of a fan of signal rays through a model ionosphere at frequencies of 20 and 30 MHz. The high-elevation rays escape into space and this results in a so-called “skip-zone” where HF signals are not returned to ground. The focusing of reflected rays results in enhanced illumination intensity immediately following the skip zone and this effect gradually decreases with increasing ground range.
Ray tracing through a model ionosphere at 20 and 30 MHz. The plasma frequency may be most simply approximated as 9 times the square root of electron density.
At a particular signal frequency, skywave propagation will produce a useful range interval on the ground where the signal power density illuminating the targets is sufficient for effective OTHR operation. This useful range interval moves further out in ground range as the signal frequency increases. An operator may therefore control the placement of the OTHR surveillance region in range by appropriately selecting the signal frequency.
System Characteristics
OTHR systems are typically two-site (bistatic) systems that use continuous-wave signals to enhance sensitivity. A distance of about 100 km between the transmit and receive sites provides sufficient isolation and is small relative to the OTHR range coverage (1000–3000 km). This quasi-monostatic configuration provides near-reciprocal two-way ionospheric propagation paths between the radar and targets. This greatly simplifies optimal frequency selection as well as the conversion of tracks from radar to geographical coordinates.
Linear frequency modulated (LFM) continuous waveforms provide desirable ambiguity function characteristics, effective control of out-of-band emissions, constant-modulus envelopes suitable for solid-state amplifiers, and computationally efficient matched filtering. Depending on the OTHR mission for air or surface surveillance, LFM waveform parameters may have a bandwidth of 5–50 kHz, coherent processing interval of 1–40 s and waveform repetition frequency of 2–60 Hz.
The vertical log-periodic antenna element strikes a good balance between broadband performance, radiation properties, and array cost for OTHR transmissions. Separate high- and low-band arrays are typically used to cover the HF spectrum. A mesh ground screen increases antenna gain at low elevation angles. JORN transmit arrays containing 14 low-band and 28 high-band elements are about 160 m long. This transmit aperture length trades off high gain (sensitivity) with beam width (coverage).
A waveform generator and power amplifier per antenna element provides versatility to manage the transmit resource, reduces radiated phase noise, and enables graceful degradation to component failure. The average power of an OTHR may be 200–600 kW. Indicatively, the transmit antenna gain of an OTHR may vary from 15 to 25 dB and the transmit beam may be 8º–12ºwide in azimuth. Antennas with high gain in elevation up to about 45º support to increase range coverage. The JORN Laverton radar transmit arrays are shown in Figure 4.
The Laverton radar uses two pairs of high- and low-band transmit antenna arrays oriented at 90º to provide 180º coverage in azimuth.
The Laverton receive antenna shown in Figure 5 consists of two very wide aperture uniform linear arrays (ULAs). Each ULA consists of 480 elevated-feed twin-monopole elements and has an aperture 2970 m long. An OTHR receive antenna may have a gain in the order of 25–35 dB and fine azimuth resolution of 0.2º–2º. This enhances sensitivity against noise and clutter, and improves angle estimation accuracy for target tracking. A wide-band digital receiver per element may be used to simultaneously sample the entire HF band. This allows multiple bandwidths to be digitally down converted and extracted for signal processing using a common aperture.
The JORN Laverton receive antenna consists of two very wide aperture ULAs oriented at 90ºto provide 180ºcoverage in azimuth.
In an externally noise-limited environment, receive elements that are not well matched at lower frequencies in the HF band do not compromise signal-to-noise ratio. The receive element design and aperture length tradeoff detection and tracking performance with system cost, processing capacity, and operator workload. As shown in Figure 6, the receive antenna electronically steers simultaneous high-resolution “finger-beams” (blue) to acquire echoes from the surveillance region illuminated by the broader transmit antenna beam (red). The range depth is limited by the unambiguous range, processing capacity, and ionospheric support. Representative dimensions of a notional OTHR surveillance region and resolution cells are indicated in Figure 6.
Indicative dimensions of a surveillance region and resolution cells for a frequency
Nominal Capabilities
Figure 7 illustrates the OTHR track-while-scan capability. The azimuth coverage of 90
Overall potential coverage of an OTHR and a surveillance region within it illuminated simultaneously by a single frequency.
The OTHR illuminates a desired surveillance region in the coverage using an appropriate choice of signal frequency and transmit beam direction. The radar dwells on the surveillance region for a coherent processing interval required for Doppler processing. Target detections are used to update existing tracks or to form new tracks in the surveillance region. Tracks are formed in radar coordinates and subsequently converted to geographical coordinates.
An OTHR interrogates different surveillance regions in the coverage by selecting the signal frequency (range control) and electronically steering the antenna beam (azimuth control). Surveillance regions are searched in a scanning sequence using a suitable region revisit rate for target tracking. Active surveillance regions scheduled concurrently on the OTHR timeline may be interpreted as the real-time coverage of the system for target detection and tracking purposes.
A distinguishing feature of OTHR is the requirement for a real-time frequency management system (FMS) [11]. The FMS consists of a network of geographically distributed sounders and sensors that collect and analyze environmental data to provide operators with automated real-time advice on optimal frequency selection for a given radar mission. The FMS also supports the construction of a real-time ionospheric model (RTIM) to enable the conversion of target tracks from radar to geographic coordinates (i.e., coordinate registration).
The FMS monitors clear channel availability and directional background noise as a function of frequency across the HF band. A dedicated miniradar system evaluates ionospheric propagation support in terms of backscattered signal power and clutter spectral purity as a function of frequency, range, and azimuth. Measurements are combined to advise on optimal frequency selection. A geographically distributed network of oblique and vertical incidence sounders provides real-time information on ionospheric structure to construct the RTIM. Numerical ray tracing through the RTIM supports multipath track association and coordinate registration.
Surveillance Roles
Figure 8 shows the JORN coverage map and radar site locations near Alice Springs (Northern Territory), Laverton (Western Australia), and Longreach (Queensland). The network is controlled from a JORN Coordination Center (JCC) located near Adelaide in South Australia.
An underlying reason for long-standing interest in OTHR is that its surveillance capabilities are complementary to those of LOS sensing modalities with higher precision and more circumscribed coverage. The chief advantage of OTHR resides in its capability to provide persistent and cost-effective surveillance over wide and potentially remote geographical areas where microwave radar coverage is either not feasible or convenient.
The capability of OTHR to fill surveillance gaps in space and time is albeit at lower resolution and accuracy compared to alternative LOS sensors. OTHR can therefore not be viewed as a complete solution to a nation's surveillance requirements. The operational value of OTHR needs to be measured in terms of the contribution it provides as one element of a portfolio of assets managed as part of a comprehensive layered surveillance system.
OTHR tracks aggregated over time can support defense forces by providing a strategic picture of the normal pattern of activity in a region of interest. During military operations, the real-time OTHR output provides a live regional air and surface picture to inform mission planning and tactical deployment of defense assets to best advantage. The integration of OTHR in a command and control network leads to more efficient use of surveillance assets and potentially a reduction in the number of systems that need to be procured and maintained for an effective response. As both the speed and stealth of threat technology increases [7], surveillance systems capable of alerting and cueing forward-based LOS sensors on military platforms with early warning of potential threats can support to manage emission control and minimize the element of surprise.
Historical Perspective
During World War II, the British Chain Home radar systems operated using HF LOS propagation. Echoes were sometimes received from much longer ranges and recognized to involve reflection from the ionosphere, but target detection well beyond the horizon was not demonstrated. This insight was not lost and the first Australian experiments on target detection by virtue of skywave HF radar backscatter commenced in the 1950s. These attempts failed due to insufficient system sensitivity. Success would need to await for enhancements in signal generation and processing technology before OTHR could become a reality.
Unknown to Australia at the time, the U.S. Naval Research Laboratory were also developing OTHR. By 1961, the “MADRE” system had succeeded in demonstrating aircraft detection at ranges of thousands of kilometers by exploiting skywave propagation [9], [10]. Through a four-nation technical cooperation program established in 1965, U.S. participants became aware of Australian HF radar research in 1968. This led to a small team of Australian scientists being briefed into the U.S. program and subsequently collaboration on OTHR.
Geebung [1971–1972]
Following the success of OTHR in the United States, it was necessary to confirm that ionospheric propagation conditions in Australia would be suitable for target detection. Project Geebung involved an experimental trial to measure path loss and Doppler stability for one-way and two-way ionospheric propagation paths between Mirikata on the Woomera Rocket Range and Broome. A frequency-modulated continuous-wave transmit and receive system of U.S. design along with communications equipment and antennas of local design were used to make measurements on the 1850-km path over an 11-month period.
The final report noted that “No evidence has been found to suggest that the operational value of an OTHR in Australia would be reduced to unacceptable level by inherent physical limitations in the ionosphere.” This finding and cooperation with the United States led to a proposal for a pilot program to develop the case for an operational capability. The concept of a pilot radar led to birth of the Australian OTHR project named Jindalee, an indigenous word meaning “bare bones.”
Due to great effort from the team's prime mover, John Strath [21], to convince defense leaders of the potential capability, the Jindalee program was approved in April 1974 as a two-stage project. Stage A was to demonstrate the capability to detect targets using a simple non-scanning radar configuration. Stage B was to develop a much more ambitious system that demonstrated operational capabilities including automatic track-while-scan over a 90º sector. There was to be an extended hold between the two stages to allow for critical review of results.
Jindalee Stage A [1975–1978]
The Jindalee Stage A system was a relatively low-cost OTHR design to demonstrate aircraft detection. Stage A utilized four 5-kW transmitters to drive a 16-element vertical log-periodic antenna array loaned from the United States at Harts Range (100 km NE of Alice Springs). The receiver located at Mt Everard (28 km NW of Alice Springs) was a locally designed ULA of 128 whip antennas 640 m long. A hardware beamformer provided a narrow sector of coverage centered along the A76 international air route to the northwest. The waveform generator and signal processor were developed by local teams. The strategy was to foster Australian innovation while building on experience gained from the U.S. program.
Jindalee Stage A was turned on in 1976 and detected its first aircraft during the initial checkout phase [1]. Figure 9 shows the Stage A receive array and an A-scan display in which the signal level was displayed as a function of Doppler for each range cell with ground clutter at the center. The Stage A radar demonstrated a credible aircraft detection capability over a two-year period. A large amount of environmental data was also recorded to guide future OTHR development. Although not expected for the Stage A system, trials in December 1977 demonstrated the detection of ships off the northwest coast of Australia. The fine Doppler resolution of OTHR also showed the potential to resolve minor speed variations in aircraft.
Stage A receive antenna array (left) and A-scan radar display of signal level versus Doppler for each range cell (right).
A particularly significant event was a live demonstration to a high-level defense delegation in April 1977. A dedicated aircraft was scheduled to fly a planned set of maneuvers. The visitors were advised of what they would see on the radar A-scan displays. The target was detected as expected but failed to perform the maneuver as predicted. To the relief of all present (and perhaps the future of OTHR in Australia), the target was observed to perform the planned maneuvers just before the party left the control room. There had been confusion over times at the radar site in Central Australia and the aircraft in Western Australia. This real-time demonstration was critical in gaining funding approval for Stage B development in 1978.
Jindalee Stage B [1978–1987]
The Stage B system was to be a substantially larger and more ambitious OTHR to demonstrate operational capabilities. It was designed entirely by the former Weapons Research Establishment (WRE) that is now DSTG. A number of key decisions, contrary to advice from the U.S. community, were made for Stage B that proved to be crucial to the success of the program. Among these were: a) full automation of radar operation; b) inclusion of a calibrated FMS; and c) the development of an arithmetically oriented (ARO) vector processor.
The Stage B transmitter used the 16-element log-periodic array from Stage A and each antenna was fed by a 20-kW servo-tuned transmitter. This enabled transmit beams with high power and gain to be steered electronically over a 90º sector. The transmit antenna is shown in Figure 10. The receive antenna was expanded to 462 dual-fan elements spanning a 2.8-km-long ULA. The receive antenna was configured as 32 fully overlapped subarrays. Each subarray was connected to a hardware beamformer distributed in underground bunkers along the array and each sub-array output was connected to a heterodyne receiver. The receive antenna is shown in Figure 11. The ability to divide the transmit and receive arrays in half and operate them as two independent radars was a novel feature of the Stage B system. This enabled the single radar dwell detection probability to be traded with region revisit time for tracking as ionospheric conditions varied.
The Jindalee OTHR transmit antenna array consisted of 16 vertical log-periodic antenna elements each fed by a 20-kW power amplifier.
Stage B receive array employed fan antennas to broaden frequency response and a galvanized wire ground screen to increase low elevation gain.
Stage B pioneered the use of a real-time and site-specific FMS [11] to provide automated advice on optimal frequency selection. This included backscatter sounder, miniradar, HF spectrum monitor, and oblique sounder subsystems. The radar operator defined the radar mission and coverage regions, while the FMS data were used by the software to determine the optimum operating frequency for each radar region. There was one oblique sounder path, but no method to determine the ionospheric reflection height. Conversion of radar tracks to geographic coordinates was a manual operation in Stage B.
A minimum Stage B system configuration was deployed in April 1982. Aircraft were first detected in June 1982 and ship detections were achieved in February 1983. Enhancements to the minimum system followed in rapid order with selectable surveillance regions added in October 1982, dual-frequency capability in June 1983, and the introduction of an automated tracking system in February 1984.
The Stage B radar was software controlled to a much greater extent than Stage A. The display capability of Stage A was also replaced with enhanced operator interfaces. As shown in Figure 12, this enabled visualization of raw detections on azimuth–range–Doppler video displays. A second geographic display showed automatically generated tracks that could be manually confirmed by operators.
ARO vector processor (top) was critical to the success of Stage B. Raw video (left) and geographic track displays (right). The raw video display showed signal intensity as a function of Doppler (horizontal) and range (vertical) with azimuth (finger beams) nested in the vertical dimension.
The key to realizing a real-time track-while-scan capability over 90º in Stage B was the WRE-developed ARO processor and its associated multiport memory. Commercial array processors that were just emerging only provided a simple vector capability and required interaction between the host processor and vector processor on a function by function basis. The ARO processor incorporated a novel multivector capability with an internal control processor to avoid these significant overheads. This proved to be critical in achieving real-time performance for both the radar and FMS.
The transition from Stage A to Stage B took roughly six years to complete. A series of Joint Service Evaluation Trials (JSET) were conducted between 1984 and 1986 to demonstrate capability for various air and maritime military targets. The JSET trials were the first time that RAAF operators controlled the radar consoles. The trials evaluated detection and tracking performance in areas of maneuver handling, multiple aircraft discrimination, slow-speed targets, long-range targets, low radar cross-section targets, and the cuing of interceptors.
Following the success of Stage B, the ongoing uncertainty about the future of OTHR in Australia was resolved by the recommendation in the 1987 Defence White Paper (DWP) that Australia build an OTHR operational network based on the Jindalee development.
Jindalee Stage C (JORN Phase 3) [1987–2003]
A defense acquisition project office was stood up following the 1987 DWP recommendation and Jindalee Stage C was later renamed to JORN Phase 3. The Stage B radar previously known as the Jindalee Experimental Facility was transferred from DSTG to the Project Office and renamed the JFAS. JFAS was to have two roles: i) allow defense to gain operational experience with OTHR and ii) support ongoing research relevant to future JORN radars.
The initial refurbishment of JFAS was completed in 1990 and military personnel from the RAAF and Royal Australian Navy (RAN) assumed responsibility for day to day operations. On 1 January 1993, the radar was officially transferred to the RAAF as an operational system with the establishment of the No 1 Radar Surveillance Unit (1RSU). Figure 14 shows the JFAS control room following the Stage C upgrade.
Jindalee Stage B RF equipment rack including receivers, waveform generators, and timing systems.
DSTG in collaboration with industry partners continued to improve the functionality and performance of JFAS during JORN Phase 3. The servo-tuned transmitters were replaced with solid-state amplifiers that allowed the system to rapidly change frequencies while scanning between dwell regions at different ranges. A low-band transmit antenna array was added to improve performance when the radar was operating below 12 MHz. Obsolescence issues led to an RF upgrade including a replacement of the receivers that enabled the range coverage of a single region to be doubled, and new waveform generators that provided both improved performance and flexibility.
Rapid advances in commercial computers allowed signal processing to be rehosted from the ARO processor to DEC Alpha computers. Display hardware was similarly migrated to DEC Alpha workstations using X windows. These changes made it possible to rapidly implement functionality changes requested by operators. Software changes were made to radar calibration, signal processing, automated tracking as well as operator displays. An initial demonstration to remotely operate JFAS from the RAAF base at Edinburgh near Adelaide in South Australia was conducted in 1995. 1RSU relocated its operational staff to RAAF Edinburgh in July 1999.
The JFAS upgrades enhanced signal processing and tracking to improve sensitivity, resolution, and accuracy. This included a range of adaptive processing techniques to mitigate against clutter, interference, and noise, many of which are classified. The JFAS capability advanced significantly due to the ongoing research and development in Stage C. These capabilities would be subsequently transitioned as future enhancements into the new JORN radars being designed and built during this time.
Jindalee Operational Radar Network
Two new OTHR sites were selected for JORN. Radar 1 with 90º coverage is at Longreach in Queensland. Radar 2 with 180º coverage is at Laverton in Western Australia. JFAS with 90º coverage is designated as Radar 3. The overlapping coverage in the highest priority area over the Timor Sea in Figure 8 minimizes the ability of targets to avoid detection by flying tangentially relative to a single radar.
JORN Phase 4 [1991–2003]
The contract specification for the JORN radars was mostly limited to capabilities that had already been demonstrated on the Stage B radar with two notable exceptions: i) the use of a receiver per element architecture rather than a receiver per subarray design to provide future capability for enhanced signal processing algorithms, and ii) an automated process for the conversion of tracks from radar to geographic coordinates. The Alice Springs radar was not initially integrated into the network at a command and control level, but provision was made for surveillance information at the track level to be passed to the JCC near Adelaide in South Australia.
Defense Science and Technology staff were actively engaged in the specification of the JORN requirements in support of the Jindalee Project Office. The contract to build a network of two radars and a central command center was awarded to Telecom Australia in June 1991 with Marconi and Telstra (a Telecom Australia and RLM joint venture) as major subcontractors. The project was initially scheduled for completion in June 1997 but was plagued with development delays. Lockheed Martin Australia assumed full responsibility for the project in 1997 and final operational capability was achieved in 2003.
The RAAF accepted JORN into service in April 2003. Figure 15 shows the network operations control room shortly after the acceptance of JORN into service. Radars 1 and 2 were maintained by Lockheed Martin Australia while Radar 3 was maintained by BAE Systems Australia. A general description of the JORN architecture and its surveillance capabilities is given in [12], while an insightful case study of the acquisition and management of JORN as a defense procurement project appears in [13].
Network operations at the JCC after the RAAF accepted JORN into service in April 2003.
The risk-managed nature of the JORN specification, and the ongoing research and development program on the JFAS radar meant that the initial JORN radars would be subject to an incremental improvement program. The large capital cost involved in the RF hardware made it essential that performance was compliant from the outset. Emphasis was placed on spectral purity, calibration, and receiver performance. Detailed analysis of the Marconi initial receiver design using the long-term database acquired by the Stage B FMS revealed that the receiver would be incapable of supporting radar operation. Marconi had to undertake significant redesign to achieve the specified performance. Had this database not existed, the failure of the initial receiver design would almost certainly have led to termination of the project at initial operating capability.
The JORN transmit antenna, shown in Figure 16, is a ULA of vertically polarized log-periodic antenna elements. The high-band array is a 162.5-m-long ULA composed of 28 (14 + 14) elements spaced 5.75 m apart for operation between 11 and 32 MHz. The low-band array is a 155-m-long ULA composed of 14 (7 + 7) elements spaced 12.5 m apart for operation between 5 and 13 MHz. Both high- and low-band arrays may be operated in full-aperture mode or as two independent half-aperture transmitters. Each element of the array is fed by a 20-kW amplifier with a pair being combined for the high-band array to yield a total maximum transmit power of 560 kW.
JORN Radar 1 transmit antenna array at Longreach, Queensland. High- and low-band arrays provide high-efficiency operation over the HF band.
The JORN receive antenna, shown in Figure 17, is a ULA of 480 elevated-feed monopole-doublet elements spaced 6.2 m apart with a digital receiver per element. The receive aperture is 2970 m long and may have a spatial resolution between 0.2º and 2º over the design frequency range.
JORN Radar 1 receive antenna array. Fat monopole antennas improve broadband performance and minimize wind-driven aeolian noise.
The Laverton radar consists of a single set of transmitters and receivers that can be switched between two orthogonal transmit and receive arrays of identical design on a dwell by dwell basis. This provides the ability to flexibly place radar dwell regions over a wider area with 180º coverage in azimuth. This coverage is albeit with the same scan revisit constraints as a single 90º radar.
The JORN OTHR sites are remote and thousands of kilometers away from the JCC. The JORN real-time control and management software needs to adapt to dynamic propagation conditions and communicate in real time with the remote transmit and receive systems. The limited bandwidth of the communication links made it necessary to perform high data volume signal processing at the receiver sites with transfer of intermediate data to the JCC for tracking and display.
JORN Phase 5 [2003–2007]
Capability enhancements driven by DSTG's research and development program coupled with the evolving operational requirements derived from service use culminated in JORN Phase 5. The main objectives of Phase 5 were to 1) rationalize maintenance and operations across all three radar systems; 2) further progress the integration of JFAS as Radar 3 in a national OTHR command and control network; 3) introduce new capabilities developed by DSTG on JFAS during the JORN Phase 3/4 period; 4) improve the distribution of surveillance information from the network to national agencies; and 5) continue to undertake further research and development of OTHR technology. JORN Phase 5 was jointly delivered by BAE Systems Australia and Lockheed Martin Australia.
Among the JORN enhancements detailed in [14] is a “stare” capability where the radar repeatedly observes the same geographic area with high revisit rate to provide operators with improved tracking performance for maneuvering targets. Automatic conversion of target tracks from radar to geographic coordinates was improved through real-time interpretation of ionospheric sounder measurements and the introduction of 2-D numerical ray tracing that increases the accuracy of coordinate registration. Phase 5 integrated JORN and JFAS human–machine interfaces to provide consistency for operators and reduce training overheads.
The legacy signal processor was replaced by large clusters of commodity processors capable of sharing large volumes of data with low latency that allowed compute capacity to be scaled and increased as required to host new real-time algorithms. The enhanced signal processing capacity paved the way to improve range-depth coverage. It also enabled the reduction of signal processing losses in achieving twice the coverage capability using data extrapolation techniques [15]. A practical example of this technique is illustrated in Figure 18. A particularly significant addition in Phase 5 was a cognitive processing suite incorporating multiple simultaneous streams of advanced signal processing capabilities [16], [17], [18].
Top panel shows an intensity-modulated range–Doppler display using the central 64 sweeps of an OTHR air-mode dwell that contained 128 sweeps. Multipath echoes from an aircraft with low relative velocity are poorly resolved from clutter near 0 Hz Doppler frequency (target A). An echo from a different aircraft with higher relative velocity is well resolved from the clutter but has a much lower SNR (target B). The middle panel shows the range–Doppler map after the same 64 sweeps of real data are extrapolated to 128 sweeps prior to coherent processing. The enhancement in Doppler resolution and SNR gain enables targets A and B to be more clearly distinguished from clutter and noise, respectively. The bottom panel shows the range–Doppler map resulting when all 128 sweeps of collected data are coherently processed. The similarity between the middle and bottom panels illustrates that the technique can be used to halve the radar dwell time without compromising performance.
JORN Phase 6 [2018-Ongoing]
The JORN Phase 6 Mid-life Upgrade program aims to extend the JORN life-of-type to beyond 2040. The planned upgrades will address obsolescence issues and component commonality across the radar network. They will also enable significant enhancements to future operational capability. The contract for Phase 6 was awarded to BAE Systems Australia in April 2018. The scope of the Phase 6 program includes all system components except the antennas and cabling of the radars and remote ionospheric sounders. An exception is that the JFAS receive antenna will be fitted with a receiver per element to upgrade the receiver per subarray architecture.
The Phase 6 specifications draw heavily on DSTG and will introduce improvements developed by DSTG during Phase 5. The DSTG research and development program was driven by a careful analysis of the enhancements required to a) deliver improved detection of small targets, slow targets, and maneuvering targets in disturbed environments; b) increase real-time surveillance coverage; c) improve target geolocation capability; and d) reduced operator workload. The program benefited from unparalleled access to the JORN radars for DSTG activities and from close interaction with the operational community. Some of key system upgrades are described in the following.
Next-Generation Common Aperture Receiver
Advances in high-speed data acquisition systems with sufficient dynamic range and linearity will lead to the replacement of the JORN heterodyne receivers with broadband devices that capture the entire HF spectrum and digitally down convert the desired signal bandwidths [4]. Although simple in concept, the design of a direct digital receiver with suitable noise figure, dynamic range, and linearity to ensure OTHR performance remains limited by the external environment is complex and demanding. This is partly due the HF large signal environment and its significant variation with location and time.
The DSTG research program developed and tested a series of wide-band direct digital receivers and waveform generators. This led to the delivery of an exemplar “Common Aperture” OTHR receiver shown at the top of Figure 19. BAE Systems Australia were provided with this exemplar as Government Furnished Material and further developed their receiver to exceed the JORN Phase 6 specification. This design placed a large emphasis on manufacturability, given the need for several thousand receivers in JORN Phase 6. This technology enables common aperture operation, where the full receive aperture may be used for multiple simultaneous missions at different frequencies. Unlike the traditional JORN heterodyne receiver design, common aperture operation allows use of the full receive array aperture for the following:
half-radar transmit operation, which increases sensitivity by 3 dB and doubles the angular resolution;
high-fidelity channel evaluation for optimum frequency selection using operational signal processing routines;
potential bistatic operation with very large baselines using transmissions from one or more other JORN radars.
Next generation digital receiver (top) and waveform generator (right) and portable roadie racks (left) containing receivers and waveform generators.
Transmit Capability Enhancements
Although not originally planned for Phase 6, the development of state-of-the-art high-power amplifiers (HPAs) by Australian company Schach RF will lead to the replacement of HPAs at all JORN sites. The JORN Phase 6 waveform generators, also shown in Figure 19, will provide exceptional levels of spectral purity.
As shown in Figure 20, DSTG has been involved in the testing and evaluation of high-precision cryogenically cooled sapphire controlled oscillators developed by the University of Adelaide and commercialized by QuantX Labs. Exploiting this technology provides long-term timing stability and levels of phase noise traditionally seen only in national standards laboratory settings. This new reference source removes prior limitations to dependent (transmit and receive) components.
Signal Processing Enhancements
OTHR systems are not as constrained by the space, weight, and power restrictions that limit processing capacity for conventional radar systems on military platforms. The much lower data rates and early-warning capability of OTHR relative to microwave radars also reduces demand on processing latency. It is therefore possible to implement sophisticated signal processing techniques with high computational complexity in real-time OTHR systems that may not be feasible in other types of radar systems. The exponential growth in computer power since the original JORN specification will allow advanced signal processing algorithms developed by the DSTG research program to be transitioned in Phase 6.
Sensitivity and resolution in both range and azimuth will be significantly enhanced through bandwidth and aperture extrapolation techniques. Coupled with the technique used in Phase 5 to halve the radar dwell time, this suite of algorithms significantly increases the coverage area that can be kept under surveillance by a given radar with minimal compromise in detection and tracking performance. Simultaneous processing of radar dwells along with active and passive channel evaluation dwells provides enhanced capabilities for both radar and frequency management functions.
Coordinate Registration Improvements
JORN Phase 6 will replace transmit and receive hardware of all ionospheric sounders and transponders. The spatial structure of small- and medium-scale ionospheric disturbances that affect radar performance are captured over an extended and more densely sampled geographical coverage. This increases the fidelity of the RTIM and creates the potential for short- and medium-term ionospheric forecasting for radar tasking purposes. The use of 3-D numerical ray tracing to model geomagnetic-field induced “out-of-plane” ray-path deviations and multipath propagation including high-ray modes using the enhanced RTIM will improve coordinate registration and hence track accuracy.
Reflections and Prospects
Sovereign development of an operational capability provides advantages for defense as well as wider national benefits. In the case of JORN, this includes the ability to tailor system performance to Australian requirements, more deeply understand capabilities and gaps, control life extension, and create pathways that accelerate the transition of innovative concepts to operational capabilities. This grows the science and technology ecosystem and the economy. It is instructive to reflect on some key factors contributing to the success of OTHR development in Australia with reference to the three pillars of the More, Together Defence Science and Technology Strategy 2030 [19]. These pillars are 1) brilliant people, collaborative culture; 2) outstanding infrastructure powering innovation; and 3) one-defense capability.
Brilliant People, Collaborative Culture
It is widely recognized that although defense industry has played an important role in the construction and maintenance of JORN systems, DSTG has been the originator of most, if not all, of the innovation that has transitioned to operational capability. One factor that has contributed to this has been strong leadership combined with highly motivated scientists, engineers, and technicians from a wide range of disciplines, most of whom have devoted their entire working careers to the development of OTHR capability. A hallmark of the Australian HF radar program has been the high degree of teamwork and close interaction between different disciplines through a shared culture within a focussed branch focussed branch of DSTG that has been singularly dedicated to OTHR development over the long term.
Outstanding Infrastructure Powering Innovation
The Australian OTHR program is strongly founded in a sophisticated experimentation philosophy. This is supported by outstanding physical and digital research environments that drive innovation. The ability to develop, field, and test experimental HF equipment and systems has enabled the Australian HF radar enterprise to rapidly advance scientific knowledge, validate innovative concepts, and demonstrate new capabilities for defense clients. The courage to take risks in trying new and unorthodox approaches to meet defense requirements has characterized Australian OTHR development from the outset. Failures along the way have been valued as learning opportunities upon which to build.
One-DEFENSE Capability
A contributing factor to making JORN a unique operational capability has been the close collaboration between DSTG and the RAAF operational community. This formally commenced with the JSET in 1986 and continues to the present day. Through this partnership, DSTG staff enjoy direct access to operational OTHR systems to conduct ongoing research and development activities, which in turn are driven by a close understanding of RAAF operational priorities. Real-time capability demonstrator systems managed by DSTG are configured side-by-side with operational OTHR systems to process live data feeds on a noninterference basis. This facilitates rapid prototyping, development, and testing of robust techniques in operationally relevant environments.
Partnering with industry has been essential to scale capacity and to leverage complementary capabilities on focal projects. Corporate risk needs to be lowered in order to incentivize and achieve innovative solutions. Trust is an essential component of successful partnerships and the JORN experience showed that partnerships thrived when all parties demonstrated value to each other. The successful achievement of future partnering models will be critically dependent on the ability of defense industry to develop the deep expertise that has characterized the DSTG program over many decades. Early steps are being taken with the contracting of industry to supplement DSTG resources with an embedded workforce.
To train the next generation of HF radar scientists, engineers, and technicians, a partnership between defense, BAE Systems Australia, and The University of Adelaide has been formed with the intent to establish a Center for Advanced Research in HF Technologies. Through world-class education and research environments, this initiative aims to grow and nurture a “home-grown” talent pipeline to sustain the Australian HF enterprise by providing university engineering scholarships, internships, and undergraduate industry placements. The intent is to create new defense-focused courses and support-targeted research activities using dedicated laboratories and small-scale OTHR systems in a framework open for other academic institutions and industry partners to become involved.
Future Prospects
The Australian government has recently announced the intent for an expansion of the JORN to provide wide-area surveillance of Australia's eastern approaches [20]. Although timescales have not been announced, multiple location and configuration options are currently under consideration by DSTG, each with its associated benefit and cost. Modeling studies are currently underway to inform decisions.
Although JORN is the most advanced OTHR network, its surveillance coverage is still limited to geographical areas identified in the original system design. DSTG continues to develop modular, scalable, and moveable HF radar capabilities exploiting multiple modalities and processing techniques for the detection and tracking of challenging targets. This provides agility to meet short-term surveillance requirements that are not feasible for JORN.
Conclusion
Of all Australia's significant defense assets, the JORN is probably the least visible to Australians for delivering safety and security to the nation. It grew from humble origins in the desert, where steel for antennas was initially fabricated using home-spun technology by the iconic Hills Hoist clothesline company, to become, more than half a century on, the finest OTHR system in the world and a key plank in defense operational capability for wide-area surveillance.
Several nations experimented with OTHR before and during the era of Project Jindalee. Some superb systems were built at huge cost, but none of these systems were ever transitioned to defense operational capability. Australia could not afford anything on that scale for Project Jindalee. The pioneers of OTHR in Australia had to overcome significant credibility and technical challenges to make it work.
DSTG has since made an enormous investment over several decades in brainpower to advance OTHR capability. Close collaboration between defense, industry, and international partners has been instrumental. The JORN is an astounding outcome, and should be acknowledged as one of Australia's finest defense science and technology achievements.
No comments:
Post a Comment