The last two decades witness the emergence and success of exoplanet science. Thousands of exoplanets have been detected using various methods. However, there is still a lack of Earth-like planets orbiting Sun-like stars. This situation will probably be changed by a combination of the radial velocity and transit methods. To disentangle planet signals from correlated noise in radial velocity data, I have introduced the so-called "Goldilocks principle" of noise modeling, a new noise proxy called "differential radial velocities" and red noise periodograms, which are implemented in a web application called "Agatha". These new techniques have enabled discoveries of Earth-sized planets corresponding radial velocity variations as low as 0.3 m/s, reaching the detection limit of Earth analogs (0.1 m/s). The application of these new methods in the analysis of radial velocity and transit data may lead to the identification of a network of habitable worlds. 

Dr. Fabo Feng got his PhD at the Max Planck Institute for Astronomy (Heidelberg, Germany) and he is currently a postdoc fellow working at the University of Hertfordshire (UK). He has developed a framework of periodograms, called "Agatha", to disentangle periodic signals from correlated noise. This software has been used to find the HD 207947 and Tau Ceti planet candidates with radial velocity variations as low as 0.4 and 0.3 m/s, respectively.