In this thesis, the application of simulation-based methods in Radar detection is studied. In these methods, parameter estimation and calculation of the integrals are done by generating random samples. Specially, two detectors based on Importance Sampling are presented. In these detectors, which are called Particle Detector (PD), the approximated likelihood ratio is calculated by Monte Carlo sampling. In the first detector, the unknown parameters are first estimated and are substituted in the likelihood ratio. In the second detector, the unconditioned likelihood ratio is obtained by integrating out the unknown parameters. Due to numerical approach in these methods, they can be applied in many detection problems which do not have closed form solutions. Simulation results show that for problems in which the Generalized Likelihood Ratio Test (GLRT) can be applied, the performance of the proposed detectors and the GLRT are approximately the same. On the other hand, the proposed detectors can be used in many problems in which the Maximum Likelihood (ML) estimates of unknown parameters do not exist or some prior information about the parameters is available. In addition to the simulation-based methods, a new approach for detecting a target with an unknown amplitude in clutter is proposed. The received signal models under two hypotheses, H 0 and H 1 , are assumed to be the same except that the target amplitude is zero under H 0 . Using the Bayesian approach, it is shown that the likelihood ratio can be calculated as the ratio of the prior and the posterior probabilities of the target amplitude. Based on this relation, a new method for target detection in Gaussian clutter is presented, namely, Prior Posterior Ratio Test (PPRT). This method is applied to white and colored clutter with known or unknown statistics. Simulation results show that the proposed detector has much better performance compared to the conventional GLRT detectors. It is also shown that with some modifications in detection rule, the proposed detector exhibits Constant False Alarm Rate (CAFR) property.