Knowing the location and location of equipment, vehicles and forces in order to make the police smarter for the accurate execution of missions is a requirement and in this regard, the inertial navigation system has high accuracy and efficiency and can be used in electronic warfare. Due to the increasing use of Inertial Navigation Systems (INSs) in various fields, increasing the accuracy and efficiency of this system is significant. INS is a standalone system that works in any weather conditions and can provide continuous navigation information of position, velocity, and attitude without affecting the environment. However, due to errors in gyroscope and accelerometer sensors, the accuracy of INS sensor performance decreases over time. Therefore, the main problem in using INS for navigation systems is the unlimited errors that occur over time. In other words, over time, its performance suffers from errors that are increasing rapidly. Therefore, to achieve the desired outputs, noise removal must be performed. In this paper, according to the types of sensors, the best noise removal method has been presented. The wavelet noise removal based on the lifting scheme is utilized for de-noising. The lifting method is real-time and does not require blocking the received data. Also, we have leveraged some multi-level liftings. Further, we propose an optimization method based on the genetic algorithm to increase accuracy by combining the results of multi-level liftings. For evaluation, two types of dynamic and static data have been used. The achieved results have clarified the effectiveness of our proposed method. Therefore, a suitable method for good, reliable, and fast noise removal is presented.