In recent years, localization systems have gained significance in the indoor environment because they are used at airports, highrise buildings, and parking garages. The performance of traditional localization technologies like global navigation satellite system (GNSS) degrades in the indoor environment because of the strong presence of multipath components, low received signal strength, and strong signal attenuation. Additionally, indoor localization techniques like trilateration and triangulation have limitations because these techniques require a direct line-of-sight environment and need multiple access points (APs). Thus, WLAN fingerprinting-based indoor localization has gained popularity due to its stable performance and being widely available. The general fingerprinting approach is received signal strength-based, which has performance limitations due to signal fluctuations and multipath components. Channel state information (CSI) based-fingerprinting has proven more stable in indoor localization. In this paper, we present the performance study of CSI based fingerprinting in three different scenarios: no, fixed, and moving multipath components. We utilize artificial neural network(ANN) and compare the localization error of each scenario with received signal strength-based fingerprinting.