To address scan-dependent biases stemming from copolar patterns, it is essential to rectify these discrepancies by utilizing appropriate correction values at each beamsteering angle of interest. Attaining these correction values requires a meticulous and accurate characterization of the main beam copolar patterns at the desired beamsteering angles. The noticeable cross-polar patterns lead to undesired interaction between signals from horizontally and vertically oriented fields, consequently introducing biases in the estimated polarimetric variables. To counteract this effect, a practical solution involves utilizing pulse-to-pulse phase coding within either the horizontal or vertical transmission elements. Notably, this method has undergone empirical validation and has the advantage of not necessitating detailed knowledge of the cross-polar patterns.
In its simplest form, the pulse-to-pulse phase coding applies [0°, 180°] phase shifts in one of the orthogonal channels while leaving the other channel not coded. The phase shifts can be ordered alternately or in the block manner. In the former, the code is applied as a train of successive [0°, 180°] phase shifts. This complicates the ground clutter (GC) filtering because the ground clutter signal appears not only at zero velocity but also at the ±va (where va is the unambiguous velocity) in the spectral domain. In the case of block order, the first half of the pulses are not coded while the phase is shifted by 180° for the rest of the pulses. Such code order ensures that the GC signal appears only at and around the zero velocity. Nonetheless, it is found out that, in radars with significant cross-coupling, the application of block ordered [0°, 180°] phase shifts adversely affects the GC filtering. Herein, these effects are addressed and the potential solutions are investigated.

