ECCM – RADAR PROBLEMS

Jammers are typically barrage noise or repeater jammers. The former try to prevent all radar detections whereas the latter attempt to inject false targets to overload processing or attempt to pull trackers off the target.

A standoff jammer attempts to protect a penetrating aircraft by increasing the level of noise in the radar’s receiver. In such an environment, the radar should be designed with electronic counter countermeasures.

These can include adaptive receive antennas (e.g., adaptive array or sidelobe canceler), polarization cancelers (defeated easily by jammer using independent jamming on horizontal and vertical polarizations), sidelobe blankers to prevent false pulses through the sidelobes, frequency and prf agility to make life more difficult for the repeater jammer, low probability of intercept (LPI) waveforms, spread spectrum waveforms that will decorrelate CW jammers, spoofer waveform with a false frequency on the leading edge of the pulse to defeat set-on repeaters or a spoofer antenna having an EIRP that covers the sidelobes of the main antenna and masks the transmitted pulses in those directions, receiver uses CFAR/Dicke-fix, guard band blanking, excision of impulsive noise in time domain, and excision of narrow-band jammers via the frequency domain, etc.

In stressing cases, the radar can employ burn through (i.e., long dwells with noncoherent integration of pulses). Bistatic radars can also be used to avoid jamming. For example, a standoff (sanctuary) transmitter can be used with forward-based netted receive-only sensors [avoid antiradiation missiles (ARMs) and responsive jammers] to located targets via multilateration.

Ultralow sidelobe antennas can be complemented with remote ARM decoy transmitters that cover the radar’s sidelobes. Adaptive antennas include both adaptive arrays and sidelobe cancelers. The adaptive array includes a number of low-gain elements whereas the sidelobe canceler has a large main antenna and one or more low-gain auxiliary elements having sufficient gain margin to avoid carryover noise degradation.

The processing algorithms are either analog (e.g., Applebaum orWidrow LMS feedback) that can compensate for nonlinearities or are digital (sample matrix inversion or various eigenvector approaches including Gram–Schmidt and singular valved decomposition (SVD)). Systolic configurations have been implemented for increased speed using Givens rotations or Householder (conventional and hyperbolic) transformations.

In a sidelobe canceller (SLC) the jamming signal is received in the sidelobe of the main antenna as well as in the low-gain auxiliary element. By weighting the auxiliary signal to match that of the main antenna and setting the phase difference to 180◦, the auxiliary signal can be added to the main channel yielding cancellation of the jammer.

The weighting is determined adaptively since the main antenna is usually rotating. Target returns in the mainbeam are not canceled because they have much higher gain than their associated return in the auxiliary antenna. Since they are pulsed vs. the jammer being continuous, target returns have little effect in setting the adaptive weight. Since the closed-loop gain of an analog canceler is proportional to jamming level, the weights will converge faster on larger jammers creating an eigenvalue spread.

To prevent the loop from becoming unstable, receiver gains must be set for a given convergence time on the largest expected jammer. Putting limiters or AGC in the loops will minimize the eigenspread on settling time. The performance of jammer cancellation depends on the nulling bandwidth since the antenna pattern is frequency sensitive and the receivers may not track over the bandwidth (i.e., weights at one edge of the band may not yield good nulling at the other end of the band).

Broader bandwidth nulling is achieved through more advanced space-time processing; that is, channelize the spectrum into subbands that are more easily nulled or, equivalently, use adaptive tapped delay lines in each element to provide equalization of the bandpasses; that is, the adaptive filter for each element is frequency sensitive and can provide the proper weight at each frequency within the band.

A Frost constraint can be included in digital implementations to maintain beamwidth, monopulse slope, etc., of the adapted patterns. If the jammers are closely spaced, mainlobe nulling may be required. Nulling the jammer will cause some undesired nulling of the target as the jammer-target angular separation decreases.

This is limited by the aperture resolution. Difference patterns can be used as auxiliary elements with the sum beam. The adaptation will place nulls in the mainlobe of the sum pattern. They are actually more like conical scan where a difference pattern is added to a sum pattern to move the beam over.

The mainbeam squints such that the jammer is placed in the null on the side of the mainbeam. Better angular resolution can be achieved by nulling with two separated array faces. The adaptive pattern can now have sharp nulls that cancel jammers with minimal target loss since the angular resolution is set by the much wider interferometric baseline.