The steady state burn of fusion plasmas will require a significant increase in the amount and sophistication of feedback control with respect to present day experiments. In the last years, it has been realised that more involved real time schemes need significant advances in the signal processing and data analysis methods. Since one of the crucial issues for the control of reactor relevant configurations is the proper identification of the plasma to be controlled, various methods for the determination of the magnetic topology are being developed at JET. In addition to a real time algorithm (EQUINOX) to solve the Grad-Shafranov equation on a time scale of ms, a new approach based on Bayesian statistics is also providing very reliable and fast results. Robust methods of confinement regime identification are a prerequisite for safe, general control schemes. New identifiers based on Support Vector Machines have been developed and they have success rates exceeding 99% in determining whether plasmas are in the L or H mode. Prediction of harmful events is also an important issue in the perspective of safely operating reactor relevant devices. A new disruption predictor based on Support Vector Machines is being developed and has already provided success rates higher than 90 % in realistic real time conditions. Moreover, the generalisation capability of this new predictor has been confirmed by applying it to new experimental campaigns not used for the training. The success rate remains high even more than ten campaigns, which means more than three years, after the last one used for the training. The enormous progress of video camera technologies in the last decades has increased the range of applications of image diagnostics. Their deployment in real time requires the development of new image processing algorithms. The innovative technology of Cellular Nonlinear Networks has already been implemented successfully on JET for the real time identification of hot spots. A series of new feedback schemes has also been explicitly developed not much to control the plasma but to really improve the physics understanding of some phenomena. Particularly interesting are the simultaneous control of the safety factor and pressure profiles and the real time tracking of Toroidal Alfven Eigenmode instabilities. These advanced feedback schemes for physics understanding often require more advanced signal processing techniques like adaptive filtering.