Page 338 - Fundamentals of Radar Signal Processing
P. 338

target  is  approaching  or  receding,  or  at  what  radial  velocity.  Furthermore,  it

               provides  no  indication  of  the  number  of  moving  targets  present.  If  multiple
               moving targets are present in the slow-time signal from a particular range the
               result will still be only a “target present” decision from the detector. On the
               other hand, MTI processing is very simple and computationally undemanding.
               Despite its simplicity, a well-designed MTI can improve the SCR by several to
               sometimes 20 or more decibels in some clutter conditions.



               5.2.1   Pulse Cancellers
               The major MTI design decision is the choice of the particular MTI filter to be
               used. MTI filters are typically low order, simple designs. Indeed, some of the
               most common MTI filters are based on very simple heuristic design approaches.
               Suppose  a  fixed  radar  illuminates  a  moving  target  surrounded  by  perfectly
               stationary  clutter.  The  clutter  component  of  the  echo  signal  from  each  pulse
               would be identical, while the phase of the moving target component would vary

               due  to  the  changing  range.  Subtracting  the  echoes  from  successive  pairs  of
               pulses would cancel the clutter components completely. The target signal would
               not cancel in general due to the phase changes.
                     This observation motivates the two-pulse MTI canceller, also referred to
               as the single or first-order canceller. Figure 5.8a illustrates the flowgraph of a
               two-pulse canceller. The input data are a sequence of baseband complex (I and

               Q)  data  samples  from  the  same  range  bin  over  successive  pulses,  forming  a
               discrete-time sequence y[m] with an effective sampling interval T equal to the
               pulse  repetition  interval.  The  discrete  time  transfer  function  (also  called  the
               system function) of this linear finite impulse response (FIR, also called tapped
                                                                                        –1
               delay  line  or nonrecursive)  filter  is  simply H(z)  =  1  – z .  The  frequency
               response as a function of analog frequency F in hertz is obtained by setting z = e             j

               2πFT :
















               FIGURE 5.8   Flowgraphs and transfer functions of basic MTI cancellers: (a)
               two-pulse canceller, (b) three-pulse canceller.
   333   334   335   336   337   338   339   340   341   342   343