Pulse radars extend target range detection by increasing the transmitted pulse width. On the other hand, target resolution is enhanced by reducing the system pulse width. These dichotomous requirements led to the invention of chirp radar systems which achieve greater target resolution for a given detectable range by frequency modulating the carrier frequency of the transmitted pulse. Along with the advent of chirp radars came the extremely simple and reliable technique of chirp signal generation known as "passive generation". However, one of the undesirable features of "passive generation" lies in the infinite time required for transmission of the resultant pulse, This means that some chirp radar systems may require time gating before transmission of the pulse. Time gating becomes necessary when the time-bandwidth product (Dispersion Factor) is less than 60 because chirp radar systems with time-bandwidth products greater than 60 which do not employ time gating have provided satisfactory operation.
The waveform distortion introduced by those systems which employ time gating creates unwanted range sidelobes in the received signal since a mismatch now exists between transmitter and receiver. Range sidelobes are undesirable because when a number of targets appear close to each other the range sidelobes of the individual returns can give rise to a resultant structure which erroneously indicates a target to be present. However, past attempts at sidelobe reduction using the paired-echo technique have not provided satisfactory results because the nature of the response was unknown.
Therefore, this paper will present a derivation of the receiver response and a method for sidelobe suppression. The results of such a derivation clearly indicate that the received waveform possesses an F-M structure. It is this aspect of the results which can be used to possibly explain the reasons that the paired-echo technique has not provided satisfactory results3 because it is not possible with an F-M structure to produce advanced and delayed replicas of the radar return that have a phase structure which can be used to cancel the range sidelobes of the radar return.
Having ascertained a possible reason for the failure of the paired-echo method, a new method for sidelobe suppression had to be considered. The new technique requires modification of the known received signal to achieve a desired waveshape and a physically realizable sidelobe suppression filter. First of all, the filter must be limited in bandwidth to approximately that of the overall system. This implies that all the spectral shaping of the de-chirp output must be limited to that portion of the signal spectrum located within the system bandwidth. Since the shaping of the spectrum within the system bandwidth eliminated all the sidelobes except those near the main-lobe, one is led to consideration of pulse width widening because it reduces the signal bandwidth and incorporates the first and largest sidelobe into the main-lobe. The completion of this aforementioned task indicated the region of the altered spectrum most affected by the remaining sidelobes. With this information at hand, successive attempts at smoothing this portion of the spectrum led to range sidelobes levels of approximately - 34 db. The resultant filter, of course, is realizable in practice. However, along with the desired reduction of sidelobes came such undesirable features as increased pulse width and reduced signal-to-noise ratio. For the example considered in the text, the increase in pulse width amounted to approximately 18% and the decrease in SNR was - .75 db. Finally, these effects are illustrated by graphs depicting the resultant output waveform and the filter required to achieve these results.
|
![[NJIT logo]](http://archives.njit.edu/vhlib/images/njit_logo.gif){kind=logo}
![[Van Houten Library logo]](images/vhl-top-image.jpg)
