Abstract—To overcome the problem that single input single output (SISO) Volterra cannot model subharmonic components of ultrasound contrast agent (UCA), we propose the multiple-input-single-output (MISO) Volterra in this paper. The Volterra kernel estimations are carried out using singular value decomposition (SVD) method from numerical simulation data of representative of subharmonic based on the Church’s model. The performance of the method is evaluated using the normalized mean-square errors (NMSE) for the first-, secondand third-order Volterra model. Accuracy of the outputs from Volterra model is shown both in time domain and frequency domain. The resulting estimation from the third-order MISO Volterra model provides the best achievable improvement at the NMSE of -12.01 dB for 0.6 MPa excitation pressure (optimal of subharmonic generation).