Two Procedures to Flag Radio Frequency Interference in the UV Plane

We present here two algorithms to identify and flag radio frequency interference (RFI) in radio interferometric imaging data. The first algorithm utilizes the redundancy of visibilities in the gridded UV plane to identify corrupted data. In this scheme, the detection threshold is adjusted with radial distance in the UV plane to detect fainter RFI at the highest UV and avoid losing source structure at the lowest UV. In the second algorithm, we propose a scheme to detect fainter RFI in the visibility time-channel plane of individual baselines. Low frequency observations are plagued by ripples in the residual (source-subtracted) visibilities due to incomplete subtraction of the strongest sources in the field. This can be due to a variety of reasons including primary beam asymmetries, pointing jitter, a non-isoplanatic ionosphere and other baseline effects. In order to detect faint RFI in the presence of these ripples in the baselines, we consider a window in the time-channel plane, and subtract the strongest components of it's fourier transform to mimimize the ripples. The RFI is detected in the ripple-free inverse fourier transformed data, but are flagged in the original visibilities. The subtraction of the ripples improves the detection of faint RFI and the application of flags to the original data avoids the possibility of modifying residual fringe structure in the baselines. Application of these algorithms to 5 different 150 MHz datasets from the GMRT results in a significant improvement in image noise (20 - 50 %) throughout the field along with a reduction in systematics and a corresponding increase in the number of detected sources.