In this paper, we compare the performances of a coherent versus incoherent Direct Sequence Optical Code Division Multiple Access (DS-OCDMA) system. Superstructured Fiber Bragg Grating (S-FBG) encoders/decoders are used to implement unipolar codes such as Prime Sequence (PS) and Extended Quadratic Codes (EQC) codes. We implement the Importance Sampling (IS) technique, which is a variant of the well-known Monte-Carlo (MC) method, to evaluate the Bit Error Rate (BER) performances of the system. Our simulation results depict that coherent system outperforms the incoherent one. The last system can be used but a BER floor is demonstrated due to the beat noise of the incoherent source. We show that increasing bit rate leads to a deterioration of the BER behavior and requiring an increase of the optical bandwidth of the signal