The traditional Multiple Signal Classification (MUSIC) super-resolution time delay estimation algorithm has the characteristics of high resolution and can effectively solve the problem of multipath time delay estimation, but the estimation accuracy is poor at low signal-to-noise ratio (SNR). Noncircular signal is a common signal in modern communication systems. Using the second-order noncircular property of noncircular signal can increase the available information and improve the performance of parameter estimation algorithm. In this paper, we introduce the noncircular property of the signal into the MUSIC time delay estimation algorithm and propose a super-resolution time delay estimation algorithm joint signal noncircular characteristic. The new algorithm uses both the covariance matrix and the conjugate covariance matrix of the array output to improve the accuracy of the time delay estimation. Simulation results show that the new algorithm has the characteristics of steep spectral peak and better estimation accuracy.