Hardware Implementation of Selected Statistical Quantities for Applications in Automotive V2I Communication System

In this work we propose a concept of a transistor level implementation of a simplified iterative methods for computing several basic statistical quantities, such as the mean and the variance. The motivation behind the presented work is the realization of a calibration algorithm for determining the positions of the V2I (vehicle-to-infrastructure) communication devices in novel automotive applications. Such devices, mounted in fixed points of the road and urban infrastructure (RSU - road side equipment) will be used to support autonomous vehicles moving in urban and suburban environments. The role of the calibration procedure is to determine the positions of the RSU devices in global coordinate system (GCS) and save it in their internal memory blocks. To facilitate the hardware implementation, we introduced some modifications to existing (conventional) iterative algorithms used for the computation of the statistical quantities. For this purpose, we eliminated division operations, substituting them with bit shift operations. Shifting the bits may be easily realized fully asynchronously in hardware, using only a passive commutation field.