Abstract—Improving the resolution of Synthetic Aperture Radar (SAR) is a major issue that has gained significant attention recently, especially in the squint mode where obtaining high resolution is challenging. Generally speaking, the azimuth and range resolutions can be improvement by applying matched filtering for compressing the signal. Signal compression is a major part of the well-known algorithms such as the Range Doppler algorithm (RDA) and the Chirp Scaling Algorithm (CSA) by which high positioning accuracy and high resolution are obtained. The transmitted signal by SAR is most often in the form of linear frequency modulation (LFM). However, the chirp signal is always generated in the SAR azimuth direction at the receiver. The squint of observation angle of the radar makes the chirp signal non-linear and non-stationary. Consequently, this study particularly focuses on the azimuth signal compression and instead of the Matched Filter (MF) based on Fourier Transform (FT), an alternative Matched Filter is used for signal compression on time-frequency domain that is known as Matched on Slope Filter (MSF). The Matched on Slope Filter improves the effectiveness of the compression in both squint and non-squint modes. The simulation results indicate that the proposed method improves the resolution and decreases the Peak to Side Lobe Ratio (PSLR) in the target detection.