I present the current status of the theoretical forecasts and progress in the analysis of GMRT reionization data. Existing constraints from WMAP3 suggest a low reionization redshift, z \sim 10, which is accessible to current and future telescopes, including GMRT, LOFAR and MWA. I describe the collaborative programme with GMRT to install a new software correlator, and the search for reionization. This has allowed us to overcome the previous RFI limitations.