USER’S MANUAL__________________________________________________________________
70 __________________________________________________________________ M211322EN-D
samples of the COHO are obtained at the fixed sampling rate, and the (I,Q)
data are digitally locked downstream in the RVP900 IF-to-I/Q processing
chain (see Figure 13 on page 39). The procedure is identical to the manner
in which phase is recovered in a magnetron system, except that the COHO
signal is used in place of a sample of the transmit burst.
There are two special concerns that may come up when the RVP900 is used
in the above manner within a synchronous radar system. Both concerns are
the result of the IFDR sampling clock being asynchronous with the radar
system clock. These concerns are:
- RVP900 Generates the Radar Trigger—The trigger signals
supplied by the RVP900 are synchronous with the IFDR data
sampling clock. This is accomplished by a clock recovery PLL that
provides on-board timing, which is identical to the sampling clock in
the IFDR. Since the IFDR sampling clock is asynchronous with the
radar clocks, the RVP900 trigger outputs are similarly asynchronous.
The result is that each transmitted pulse envelope is triggered
independently of the COHO phase. The transmitted pulse is still
synchronous, but the precise alignment of the amplitude modulated
envelope varies.
In almost all cases, the exact placement of the transmitter’s amplitude
envelope does not affect the overall system stability, nor the ability of
the RVP900 to reject ground clutter and to process multi-mode return
signals. For this reason, a synchronous radar system, which is
triggered using the RVP900 triggers, still performs optimally using
the standard digital COHO locking techniques. In spite of this, some
system designers may still prefer that the amplitude envelope be
locked to the COHO.
- RVP900 Receives the Existing Radar Trigger—When an external
trigger is supplied to the RVP900, the processor synchronizes its
internal range bin selection circuitry to that external trigger. The
placement of the range bins themselves, however, is always
synchronous with the IFDR selectable sampling clock. The result is
that 27.8 nanoseconds of jitter is introduced in the placement of the
RVP900 range bins relative to the transmitted pulse.
The effect of this synchronization jitter is that targets appear to be
fluctuating in range by approximately 4.2 m. Although this is small,
relative to the range bin spacing, and does not affect the range
accuracy of the data, the effect on overall system stability is more
severe. Using both numerical modeling and actual field
measurements, we have found that sub-clutter visibility of a μsec
pulse may be limited to approximately 43 dB as a result of this 27.8
nanoseconds range jitter. This falls quite short of the usual
To Next Page
Loading...