To avoid the coefficient matrix ill-conditioned problem, and reduce the heavily reliance on data length of those least mean square/least square (LMS/LS) blind equalization algorithms, a novel electrical blind equalization algorithm of optical communication systems is presented by using ridge regression(RR) approach. By constructing a novel cost function of RR blind equalization, the blind equalization problems are transformed into solving an unconstrained optimization problem and the process of how to get the equalizer is shown. Meantime, a batch processing iterative method is constructed to solve this special optimization problem. Then the influence of ridge factor λ value of RR blind equalization algorithm and the algorithm complexity on the cost function and its property are analyzed. The proposed approach can place M-ray phase shift keying (MPSK) and quadrature amplitude modulation (QAM) modulation modes in a single, unified algorithm framework. Finally, simulation results verify that the performance of the proposed algorithm is superior to those existing LMS/LS blind equalization algorithms.