In this study, six degrees of freedom linear and nonlinear dynamic models of a helicopter are identified by two state space methods on the basis of wavelet transform and Hopfield neural network. At first, nonlinear model of the helicopter is linearized by using small perturbation theory for a trim point. Daubechies wavelet is used to identify linear model of the helicopter. Then using numerical integration, the model states are obtained. After that, the system state vector are projected and approximation coefficients can be computed using input and output data based on the wavelet projection technique. Least square algorithm is used to identify system parameters. Increasing the number of system input and output data compared to the number of unknown parameters, least square algorithm can be used. The results show that wavelet has a very good performance in the presence of noise and has a high convergence rate. In the next stage, linearized equations are used for identification with Hopfield neural network method. Using Hopfield neural network algorithm and network energy function, convergence of the neural network is investigated. Using this algorithm, the system parameters can be estimated. Hopfield neural network algorithm is a recursive algorithm which integrates with respect to time. Results of Hopfield neural network show that after a few time steps, estimated parameters has a good convergence to real parameters. In used algorithms, identification process is performed off-line. Keywords: Helicopter, Wavelet, Hopfield neural network, Six degrees of freedom