Signal Deformation Monitoring (SDM) is an important component of fault-monitoring of Satellite-Based Augmentation System (SBAS), performing real-time detections to find out potential Evil WaveForms (EWFs) hence protect safety-critical users against hazardous misleading information (HMI). The current mainstream of operational SDMs are based on multi-correlator techniques. Evaluation of an SDM algorithm under design should be carried out by simulations, for which Evil Correlation Functions (ECFs) ergodic given Threat Space (TS) under specified Threat Model (TM) should be generated. Conventionally, ECFs are generated in accordance with a group of analytical formulae derived at the end of last century, nearly equivalent to corresponding cross-correlation functions by EWFs and nominal replica. Basing on rapid development of parallel computing, massive simulations become possible to more authentically generate ECFs by first simulating corresponding EWFs with accordance to TM definition and then cross-correlating processes. Both generation methods are detailed in this article, then compared theoretically and qualitatively. Comparison results from massive simulations show that the simulate-based method has different effects to different types of correlation functions of GPS L1 C/A signals than traditional analytic-based method, at ranging criteria of either single or dual-frequency SBAS. A further performance margin of about 0.2 dB towards the weighted population of pseudo-random noise (PRN) codes could be provided to multi-correlator-based SDM by the simulative-generation method.