Pod development for SIGINT/ELINT.
Strategic or surveillance SIGINT missions are typically medium- to long-term peace/crisis missions. The platform can be either commercial-type aircraft with covert SIGINT capability, or dedicated aircraft custom fitted with specific SIGINT/ELINT capability. Such long-range flights, at medium to high altitude, cover an extensive territory. They are principally directed towards transmitters from air defense networks (surface or airborne), surveillance radars and military deployment and exercises.
Tactical reconnaissance missions, on the other hand, are performed by high-speed, low-flying aircraft, to reduce vulnerability in hostile territory. The processed information should be available for tactical combat decisions within a few minutes. Due to space limitations aboard such aircraft, the ELINT equipment is generally (but not necessarily) pod mounted.
Real-time ELINT applications - for threat identification, targeting and/or avoidance - require decision times on the order of a few seconds to a fraction of a minute. Wild Weasel and combat air strike applications. Obviously, only fighter aircraft are equipped with such capability.
Each type of mission will yield data to be processed for different needs; the signals to be captured and analyzed are also different depending on the mission. Subsequent processing, in real time or in deferred time, will first extract detection data (while minimizing false alarms), then physical transmitter localization, identification when possible and accurate measurement of relevant signal parameters.
For each type of signal, accurate navigation information from the platform must be merged with the incoming signal data to permit accurate positioning. Furthermore, signal parameters must be deconvoluted to exclude ambient noise (whether intentional or not), including active signals emanating from the platform, if any.
DESIGN CHOICES - RECEIVER
The system design goals have dictated certain technology choices among the different possible techniques of reception, direction finding (DF), signal processing (both hardware and software) for DF and ranging and diagnostic tools (hard and soft).
Among the different reception techniques considered, the serious contenders included the channelized receiver, the acousto-optical (Bragg cell) receiver, and the wideband compressive receiver or microscan technique.
The Multichannel Receiver
The channelized receiver, composed of a large number of narrowband channels to obtain the necessary system instantaneous bandwidth, manifests the basic block diagram shown in Figure 1. The selectivity, and thus the discriminating power of such a receiver, is directly linked to the bandwidth of each channel, implying a trade-off between the selectivity and the total instantaneous bandwidth. It is difficult to improve this trade-off without increasing the complexity of the receiving subsystem, although it is possible to reduce cost, weight and bulk of the receiver by integration of the SAW filters and the associated RF components.
This trade-off problem may be partially alleviated by adapting the multichannel receiver in a configuration of successive banks of SAW filters connected through a detection and switching matrix, each successive bank having SAW filters of decreasing channel bandwidth with respect to the preceding bank (Figure 1). In this way, an analysis can be performed by the first bank of relatively wideband filters in a low-density environment, while analysis capability in a medium-density environment will be obtained by a second bank of filters with medium bandwidth filters. In a high density environment another bank of very narrowband filters will do the job. This technique can allow such a configuration to obtain any desired performance level at the expense of additional receiver complexity and cost.
A major performance parameter for the multichannel approach is the channel-to-channel rejection, which should be typically on the order of 70 dB rejection at the 3 dB point between neighboring channels. The capabilities which may be obtained with the multichannel configuration include:
* high sensitivity (linked to the bandwidth of the individual channel)
* high dynamic range (provided the filter shape and attenuation are convenient and reproducible)
* high probability of intercept at the expense of a large number of channels and resulting complexity
* high selectivity, also dependent on a high number of channels if the total instantaneous bandwidth is to be covered, or else with a multistage approach which slows reaction time.
The major drawbacks of the multichannel approach are its complexity, bulk and cost.
The Acousto-optic (Bragg Cell) Receiver
The basic principle underlying acousto-optical processing is the Bragg effect. Figure 2 shows an application of this effect in which the acoustic waves propagating through an elastic, optically transparent medium interact with coherent, collimated light. Signals intercepted from the environment are converted into longitudinal acoustic waves in the medium; they propagate by alternately compressing and expanding the transparent material, whose optical index of refraction varies as a function of the local density (compressed or expanded). This acoustically induced refractivity modulation forms a diffraction grating which, when illuminated at the Bragg angle, deflects the optical beam through an angle proportional to the acoustic frequency. Multichannel spectral analysis of the applied signals is thus obtained by sensing the position of the beam of light on the focal plane imaging array.
The advances in wideband Bragg cells, solid-state lasers and charge coupled devices (CCD) detectors have revived interest in acousto-optic analyzers for various applications. This is particularly true for radar warning receivers (RWRs), where they offer a complement to the capabilities of classical RWRs against high duty cycle, long pulse and low peak-power radars.
The main advantages of the Bragg cell receiver architecture include its relative simplicity (at least in principle), good performance in wideband operation (typically 2 GHz) and high sensitivity. However, its selectivity is moderate and its dynamic range is limited. Another temporary (but rather major) drawback is the availability of the technology, which is not yet considered mature and free of risk. R&D is still needed for a production system based on this technology.
PRINCIPLES OF MICROSCAN RECEIVERS
An example of a microscan or compressive receiver configuration is given in Figure 3. The spectrum analysis of the incoming signal is performed by pre-multiplying the incoming signal by a "chirp" (a waveform whose frequency f is a linear function of time t) and introducing the resulting product into a dispersive delay line (DDL). The DDL acts as a compressor filter, which is matched to the input chirp characteristic to permit faithful recovery of the input signal information.
In the example shown in Figure 3, the chirp characteristic is parameterized by the duration 2T and the frequency excursion 2B. Such a chirp can be obtained by triggering a pulsed expansion filter (for very short chirps) or by using a linearized VCO. The response of the compressor filter is the Fourier transform, covering a domain in frequency space (let's call it a bandwidth) of B, obtained by integration over a time T.
The detection of the compressed pulse (which contains the information representative of the incoming signal spectrum) and the measurement of its position with respect to the chirp trigger sync gives the carrier frequency of the input signal. By generating a trigger every 2T (the chirp duration), successive elementary spectrum analyses are performed, corresponding to input signal analyses over a period T. The interception probability, for a short pulse, is therefore 50%. A 100% detection probability can thus be obtained by simply doubling the chirp.
The most general configuration includes a chirp whose characteristics are nB and nT (n>l) with a compressor filter featuring B and T. The instantaneous bandwidth analyzed every nT is (N-1)B, but the interception probability is 1/n for a pulse whose width is less than T (except if the chirp number is multiplied). It is important to note that, under all circumstances, the duration of the compressed pulse at the output of the analyzer is 1/B, which can have ramifications for the digitizing of output data and subsequent processing. Theoretical performances of three configurations with different chirp and DDL arrangements for DDL characteristics of B and T are given in Table 1.
The resolution bandwidth determines the receiver frequency selectivity, which means that two continuous, simultaneous signals of equal amplitude will be discriminated if their carrier frequencies are separated from each other by a value higher than the value listed in Figure 4. The profile of the equivalent analysis filter is steeper when weighting the compressor filter in amplitude and/or phase (subject to certain limitations that may result from the analyzed signal duration).
To choose the appropriate configuration for a specific application, a compromise must be found between the selectivity (best for 2B and 2T), the intercept probability (best for B/2 and T/2) and the frequency coverage (best for nB and nT).
The advantages of the microscan receiver include high sensitivity (narrow equivalent noise BW), high probability of intercept, high selectivity, good dynamic range and high measurement accuracy. A combination of wideband and narrowband microscan simultaneously allows a high probability of intercept (wideband) and a high-precision spectrum analyzer.
The comparison between the three receiver technologies is summarized in Table 2. The result is that perhaps in a future generation of products, acousto-optic technology will be preferred, but because of its relative lack of availability, the microscan technology appears to be the best choice.
DESIGN CHOICES - DIRECTION FINDING TECHNIQUES
As for the receiver, several competing technologies may be considered for the system design, including high-gain rotating antenna, Rothman's lens and interferometry. A brief overview and comparison follows.
High-gain Rotating Antenna
This technique affords the advantages of high gain, which increases the overall system sensitivity (typically 15 to 20 dB improvement, depending on frequency), and a spatial filtering effect, which leads to a better separation of simultaneous signals whose only difference is their direction of arrival (DOA). The direction finding (DF) accuracy can be typically 1/10 of the beamwidth when processing both amplitude and phase (two channels) at the antenna output. This represents about 1 [degrees] for a typical system using this technique.
By spinning the antenna rapidly, the probability of intercept can be increased accordingly, but the maximum rotation velocity may be limited by the antenna size. One major drawback of the spinning antenna, on the other hand, is that the size and weight of the antenna and mechanical rotation means make its use on a pod of reasonable dimensions unwieldy, except at higher frequencies (K band) where the antennas may be smaller. However this option can be used in ground stations or large SIGINT aircraft. Lenses
This technique is more often used for ESM/RWR systems due to its limited accuracy (a few degrees at best). The principle is to use a lens which performs the equivalent of an analog Fourier transform of the antenna output, producing the equivalent of a multibeam antenna with as many outputs as beams.
The interferometry principle is quite simple, consisting in sensing the phase difference between receiving antennas organized in an array. The number of antennas and the spacing between them determine the band covered by the array, the periodicity of angular position ambiguities and the accuracy of the measurement. In theory, there is no limit to the accuracy which can be achieved by an array of sufficient dimensions; however, this may be obtained only at the expense of additional ambiguities. To overcome this undesirable trade-off, simultaneous use of long-base interferometry for high accuracy, coupled with short-base interferometry to eliminate the ambiguities of the long-base measurements, affords an ideal solution.
The main advantages of interferometric techniques include wide instantaneous bandwidth, wide instantaneous field of view (120 [degrees] typically), and sub-degree accuracy. However, installation of such systems aboard an aircraft can be a delicate problem, as particular care must be taken not to induce wavefront distortions due to metallic obstacles or structures such as fuel tanks, weapons systems, etc. This topic will be discussed in more detail later.
The advantages and drawbacks of the different DF solutions are summarized in the Table 3.
Processing the very high rate data flow from modern receivers operating in a dense environment is a considerable challenge, particularly if the processing is to be performed within the pod itself, in real time and in flight for "Wild Weasel" applications.
The classical deterministic approach of identification has reached its limits due to the diversity, density and variety of signals in the modern environments. Thus, much work has been devoted over the last several years to developing expert systems and artificial intelligence (AI) methods of addressing the problem. Thomson-CSF's Analyzer Superheterodyne TACtique family, for example incorporates a resident "expert identifier.") SYSTEM ARCHITECTURE
The merits of the various techniques which could be used for the different functions of the ASTAC receivers have been compared with the various requirements. The selection parameters for the various solutions are summarized in Table 4. On the basis of the conclusions summarized in this table, the selection of the ASTAC subsystems was made. ASTAC associates a wideband/narrowband microscan receiver with a high accuracy interferometer, with an architecture which has been applied to wide-body aircraft for SIGINT applications as well as to the ASTAC ELINT pod for fighter aircraft.
BRIEF DESCRIPTION OF THE POD
The ASTAC pod is based on a wide- and narrow-band microscan receiver associated with subdegree accuracy interferometers (Figure 4). The pod itself is 4 meters long, approximately 40 cm in diameter and weighs 400 kg.
Its main characteristics include its C- to K-band frequency coverage, a DF accuracy of less than 1 [degrees] RMS, a processing speed greater than 20 rad/s, an active transmitter table containing more than 200 radars, a fixing accuracy rate of less than 1% of its range, a real-time processing capability and its capability to sort out several tens of radars per second.
For Wild Weasel operation with an operator, the ASTAC accuracy enables a rapid fix of the transmitter and designation in real time to SEAD weapons either on board or aboard other aircraft. The operator acknowledges the ASTAC output through a man-machine interface, however, the ASTAC operation, although controllable by the operator, is fully automatic. During the preliminary design phase, all precautions were taken to allow full hardware and software compatibility with future improvements in sensor performance, processing algorithms and processor speed. Its high accuracy, high probability of intercept, and ultra-fast processing make it convenient for a full range of missions from Wild Weasel to surveillance and reconnaissance.
Bradford Smith is employed with Thomson Central Research Laboratories patents department. He received undergraduate degrees in physics and psychology at the University of California, Santa Cruz in 1974 and a master's degree in experimental nuclear physics.
Patrice Thomas has been working in various capacities of EW for Thomson-CSF since 1971. He received his doctorate in physics from Paris University in 1971. Tabular Data Omitted Figuration Omitted.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||signal intelligence; electronic intercept|
|Author:||Smith, Bradford L.; Thomas, Patrice|
|Publication:||Journal of Electronic Defense|
|Date:||Apr 1, 1991|
|Previous Article:||EW: an integral part of helicopter design.|
|Next Article:||ESC weathers the Desert Storm.|