In this paper, a nonminimum-phase secondary path measured inside a car is analysed and its influence on noise reduction is discussed. To improve control performance, a new nonlinear algorithm using a configuration with a bilinear filter and trigonometric expansions is proposed for a control system with a nonminimum-phase secondary path. Within the controller of an adaptive feedforward control system, the proposed algorithm achieves high-order nonlinearities and has a better ability to predict a non-Gaussian signal. The application of this algorithm in reducing the interior road noise of a car is illustrated through simulations. Since road noise is complicated, six acceleration signals from a car's body are adopted as reference signals and one loudspeaker is used as an actuator to reduce noise near one microphone. The simulation results show that the proposed algorithm achieves better results than a traditional linear algorithm.