Abstract: ‍ ‍A near-field （NF） multiple parameter estimation algorithm based on trilinear decomposition is proposed for a bistatic multiple-input multiple-output （MIMO） radar system， configured with three orthogonal dipoles at the transmitter and receiver. The algorithm first constructs a third-order parallel factor （PARAFAC） model using the output data from the matched filter of the receiver. To accelerate the convergence speed and avoid initial value sensitivity for the proposed algorithm， the COMplex parallel FACtor analysis （COMFAC） algorithm is employed to estimate the steering matrices of the transmit and receive arrays. Subsequently， rough estimates of the unambiguous direction-of-departure （DOD） and direction-of-arrival （DOA） for each target are obtained from the structural characteristics of the transmit and receive array steering matrices， and based on the geometric relationships of the bistatic MIMO radar array system， rough estimates of the ranges from the transmit array to the target and from the target to the receive array are also obtained. Finally， based on the rough estimates of the NF multi-dimensional parameters， unambiguous spatial phase estimates of the transmit and receive arrays are obtained， leading to more precise estimates of the DOD， DOA， and the ranges from the transmit array to the target and from the target to the receive array. The proposed algorithm is applicable to uniform linear arrays with an element spacing of half a wavelength， and the proposed MIMO array system reduces array aperture loss compared to traditional NF localization， which requires a quarter-wavelength separation between array elements. Simultaneously， the proposed algorithm does not require spectral peak searching or additional parameter pairing operations. Simulation results show that the MIMO radar direction-finding algorithm based on the three orthogonal dipole arrays has higher parameter estimation accuracy compared to the traditional scalar array algorithm.