# Principle of PGA

In high resolution imaging of SAR, the error compensation using the INS data is necessary, especially for the airborne SAR. However, constrained by the system cost, the accuracy of the INS data may not satisfy the requirement of error compensation. Therefore, the motion errors must be estimated from the echo. PGA is the most commonly used and effective algorithm in estimating and compensating motion errors from the echo.

PGA algorithm was first used in optical processing, and was introduced in SAR signal processing by Sandia Lab of the U.S. The four practical processing steps of

PGA were presented by D.E. Wahl, which makes PGA algorithm highly suitable for real SAR data processing [4].

The principle of PGA algorithm is: As to a point target after azimuth deramping operation, its azimuth phase only contains the first-order phase that indicates its azimuth location. After correcting this phase, the azimuth phase of the point target will be a constant value. However, if there is phase error, the azimuth phase is no longer a constant value, and the phase error can be estimated by calculating the phase gradient. PGA algorithm uses the strong point target in the scene to estimate the azimuth phase error, and its performance is satisfying in the imaging of areas with multiple strong targets.

The specific steps of PGA algorithm include: