Wavelet Denoising of Inertial Sensor Input Data for Gnss-ins Integration: Simulation Approach

The Global Navigation Satellite Systems (GNSS) provides autonomous geospatial positioning across the globe, in the name of GPS, GLONASS, GALILEO, Beiduo, and other regional systems like IRNSS, QZSS. The advantage of having access to multiple satellites is accuracy, redundancy, and availability at all times. GPS (Global Positioning System) was the first GNSS in the United States which offers excellent multi-constellation solutions consists of up to 32 medium Earth observation satellites in six different orbital planes of four satellites in each can be used for many purposes like commercial, military, space, agriculture and so many. The Inertial Navigation System (INS) is a self-dependent system that does not depend on external electronic signals. Using the basic laws proposed by Issac Newton, it gives the position based on the time distance relation. The basic need of this navigation system is to know the initial position and the time at which the vehicle starts navigation. It mainly consists of two devices namely accelerometers and gyroscopes which give the amount of acceleration and rotational information of the moving vehicle using inertial sensors. GPS signal is an electromagnetic signal, can be blocked by mountains, dense forests, areas with high buildings, and in bad weather conditions. Therefore, GPS is not continuous all the time where INS is in any situation. Because of this drawback integration of GPS and INS is needed to provide continuous information all the time. The INS will update the GPS measurements whenever it is required and it will also correct GPS phase cycle slips. But most of the errors in the INS are caused by sensor short-term errors that have to be rectified before giving it as an input to the IMU for integration. The wavelet signal processing technique is applied to improve navigation signals. The wavelet will decompose the signal to multiple levels to improve the performance of the inertial sensor signals by separating the signal into lower and higher frequency noise components. However, the wavelet threshold method does not distort the rapidly-changing signal but reduces signal noise. This paper has applied different threshold techniques as part of simulation to improve the navigational performance of the experiment. On performing the wavelet denoising decimated wavelet transform and un-decimated wavelet transform using level-dependent soft threshold technique and also the hybrid un- decimated transform with the least square method as threshold technique results in the reduction of noise.