Patent Application
20140159955
1. Field of the Invention
The present invention relates to radar transmissions and, more particularly, to the suppression of sidelobe signals.
2. Description of the Related Art
Radar applications often require that the energy that is transmitted in certain directions be reduced the amount typically transmitted by the radar antenna sidelobes. One method of reducing sidelobe transmission is to simultaneously transmit a nearly equal and opposite signal through an auxiliary antenna. The amplification (or attenuation) and phase applied to the signal in the auxiliary channel, or channels, if a broad angle and/or wide band null is desired, to achieve cancelation are determined from knowledge of the complex antenna pattern of the main antenna, i.e., the amplitude and phase patterns. This process is termed “transmit nulling” and sometimes referred to as “open loop.” The achieved null depth of open loop transmit nulling is limited, however, by any errors in measuring the main antenna pattern, the auxiliary pattern, and the positioning of the main and auxiliary antennas. Furthermore, the measurements will degrade with time and the effects of the measurement environment often differ from that of the operational environment.
Another method of transmit nulling, referred to as “closed loop,” uses scattering from an opportunistic sidelobe scatterer in a feedback loop that includes the radar receiver, whereby the auxiliary channel transfer function amplitude and phase weights are adjusted until the signal is nulled to the noise level. This processing is similar to that employed in adaptive sidelobe cancellation of noise jamming, or other sidelobe interference, by which a sidelobe null is placed in the receive antenna pattern. An appropriate scatterer, however, is not always available and, in cases where one is present, it often is at too long range to yield sufficient signal strength for nulling. The radar receive antenna sidelobe is in the direction of the scatterer, as well, which further limits signal strength. For agile beam phased array antenna radars, the receive beam can be pointed toward the transmit sidelobe direction during the setting of auxiliary channel cancellation weights, and then repointed toward the targeted direction for normal operation. This increases signal strength, but usually not by enough to offset the substantial range loss that is proportional to range to the forth power.
The present invention comprises a method for adaptively generating a sidelobe null in a radar transmit antenna pattern. The method involves the positioning of a small air vehicle, either manned or unmanned, and typically a helicopter, along the radial from the radar that corresponds to the sidelobe to be nulled. The position of the air vehicle is generally constrained to an area that is just beyond the far field range of the transmit antenna. The air vehicle is fitted with a receiver and antenna facing the radar, as well as a GPS device for maintaining the designated position. The vehicle further includes a communication link to the platform of the main radar transmitter to form a closed loop that enables adjustment of the attenuators and phase shifters of the auxiliary channels to suppress signals transmitted in the sidelobe to be nulled. In place of the communication link, the vehicle can contain a transponder.
The method of the present invention is generally applicable to situations where a deep null must be maintained for a limited time. Because of the finite bandwidth and possible large main antenna aperture, the sidelobe response may be non-uniform throughout the bandwidth. For such cases, multiple time delay taps separated by fixed time delays with independent amplitude and phase controls can be added to the auxiliary channels. Additional auxiliary channels can be implemented, as well. However, these additional degrees-of-freedom introduced in the feedback path will increase the convergence time of the loop. Further, if only one auxiliary channel and tap is desired, the signals can be transformed to the frequency domain and divided into subbands. A distinct weight then is computed for each subband, and the nulling signals are reconstructed from their subband constituents. Finally, multiple simultaneous nulls can also be formed by using an additional air vehicle for each sidelobe to be nulled.
In the present application, a time delay tap within an auxiliary antenna path will be referred to as a “degree of freedom (DOF).” Thus, two auxiliary antennas with two taps each would comprise a four DOF system. The “main channel” refers to the radar main antenna path. In the case of subbanding, each subband per channel would be a DOF.
The present invention will be more fully understood and appreciated by reading the following Detailed Description in conjunction with the accompanying drawings, in which:
Referring now to the drawings, wherein like reference numerals refer to like parts throughout, there is seen in
As further seen in
Thus, system 10 forms a closed feedback loop that allows controller 32 to adjust attenuators 24 and phase shifters 26 of the auxiliary channels to suppress signals transmitted in the sidelobe and thus null the signals based on information received from communication transmitter 22 of vehicle 18 about signals received by RF receiver 20 of vehicle 18. This closed loop approach according to the present invention allows for adjustment of the auxiliary channel transfer function amplitude and phase “weights” based on signals received from the air vehicle until the superposition of the signals (auxiliary and main) is nulled to the noise level.
The method of the present invention is generally applicable to situations where a deep null must be maintained for a limited time. Because of the finite bandwidth and possible large main antenna aperture, the sidelobe response may be non-uniform throughout the bandwidth. For such cases, multiple time delay taps separated by fixed time delays 30 with independent amplitude and phase controls can be added to the auxiliary channels. The taps, if needed, should be spaced approximately c/2 BW apart, where c denotes the speed of light and BW the signal bandwidth. The entire span of the taps should exceed the maximum expected multipath delay spread of the main channel signal transmitted from the radar.
Additional auxiliary channels can also be implemented, and these additional degrees-of-freedom introduced in the feedback path will increase the convergence time of the loop (or “latency,” as discussed below). Multiple simultaneous nulls can also be formed by using an additional air vehicle for each sidelobe to be nulled.
The present invention requires that the signals transmitted through the taps and auxiliary antennas be distinguished at the vehicle, and that the vehicle antenna has not moved a significant part of a wavelength in the direction of the radar during collection of the data needed to form a pattern null. The aux and main channel transmissions can be separated in time (TDMA), in frequency (FDMA), or by coding (CDMA). For an FDMA implementation the main and auxiliary signals are sampled simultaneously. For example, a 1 GHz radar with 1 MHz bandwidth requires only one DOF and only one sample each of the main and of the auxiliary channel. The 1 MHz bandwidth implies 1 μs is sample time. For 20 dB nulling (that is, 20 dB below the quiescent sidelobe level), the vehicle antenna down range movement must not exceed 3 mm in 1 μs. This implies that the vehicle down range velocity not exceed 3 km/s or 6700 mi/hr. Perhaps more pertinently, platform vibrations must not exceed this rate. In the example of a helicopter comprising the vehicle 18, the rotor rate is typically 450 RPM. For a (huge) peak to peak antenna vibration of 1 m, the movement is less than 1 m in 0.5 rev or 15 m/s, well below 3 km/s.
The solution for the weights from the data samples may be determined as follows. Let w denote the column vector of complex weights, x the column vector of complex samples (measured at the vehicle) of the signals transmitted through all but the main channel, x0 the sample of the main channel signal, and superscript H conjugate transpose. The data x and x0 are functions of time. These data can be sampled at multiple time points separated by 1/bandwidth over an interval of time (“time sampling interval”). The weight vector satisfies the formula:
Minimize E{|wHx−x0|2}
where E{ } denotes expectation.
The solution is found by setting the partial derivatives, with respect to the elements of wH to zero, holding the elements of w constant, and solving the resulting equations for w. The same w results if the process were repeated by partial differentiating with respect to the elements of w and solving for wH. The solution is given by
w=E{xx
H}−1E{xx0*} where * denotes conjugate.
The expectations generally would be estimated by averaging the respective data over the time sampling interval. The longer the interval, the more accurate the solution but the longer the latency, which is the time required to determine and apply the nulling weights. Some systems may require that the nulling weights be determined quickly. Fortunately, because the range of the vehicle from the radar is relatively short and the vehicle can be fitted with antennas containing 10 or 15 dB gain, the signals received by the vehicle and applied in determining the nulling weights will be strong, i.e., well above system noise. The large signal to noise ratio (SNR) will reduce the averaging time. Further, the radar bandwidth is often narrow and multipath delay spread small. In such cases, effective nulling may be performed with only one DOF, further reducing averaging time.
There is seen in
Transmissions from communication transmitter 22 are received by communication transmitter 16, filtered at the communication carrier frequency, f2, using a bandpass filter 62, amplified by a low noise amplifier 64, mixed with an oscillator 66, and then A/D sampled 68. The digitally sampled signal is then processed to determine the appropriate nulling weight 70, which may be stored 72 for later use 74. Note that the weight, w, is given by α=−x0/x1 times a correction phase shift needed to account for the propagation phase delay [θ1 denotes the correction phase, =2πΔp where Δ denotes the aux transmission offset frequency and p denotes the propagation delay between radar platform and vehicle. θ1 can be determined by transmitting a low level signal exp(j2πf0t)+exp(j2π(f0+Δ)t) through the aux channel and recording the phase of α. This can be done during any unused part of the pulse repetition interval (T) and updated as needed. Note that Δp is likely to be small so that the correction, if needed at all, need only be estimated.] Thus w, when multiplied (via mixing) with the aux signal, results in the aux channel transmitting the negative of the main channel signal. This, in turn, results in sidelobe signal cancellation in the direction of the vehicle.
Also in
Finally,
This work derives from research under Government Contract W15P7T-08-C-V004. The U.S. Government has rights in this invention.
