Software based approaches to improve the EMC performance of a microcontroller using in situ EMI monitoring

This dissertation discusses immunity measurement and improvement of intelligent components in the automotive field. The wide frequency band and the failure criterion to represent the behavior of components under pressure of interference are the difficulties in the immunity measurement. The other main task in this dissertation is to improve the immunity of components without adding too much cost. The dissertation is divided into three chapters. The first chapter concentrates on immunity measurement. The injection circuit is important to the in situ test bench because it limits the measurement bandwidth and the accuracy of the measurement. An on board on board bias tee extends the measurement bandwidth from 15 kHz to 1 GHz. A measurement algorithm, that avoids the non-uniform comparison between the sampling data and the reference signal, improves the accuracy of the immunity result. Two failure ceriteria can identify the linear and nonlinear distortion of an analogue digital converter (ADC). The second chapter presents interference signal detection and compensation technique. A simple technique is used to detect the continuous wave interference (CWI) on the power supply and at the analogue input port of the microcontroller without adding too many components. The simulation of the schematic circuit and the measurement can validate the effectiveness of the method. Finally, based on the detection result; a compensation technique compensates both the interference signal on the power supply and at the input port of the microcontroller. The third chapter is composed of three sections. The first section introduces the defensive software and hardware methods to protect the microcontroller. Then the second section presents the application of the interference detection technique to improve the immunity of components in the automotive field. Finally the third section shows the influence of interference signal on a proportional integral derivative (PID) control system; the proportional integral derivative control system is modelled in Matlab. The interference detection is used in the fault tolerant control field to avoid the false fault detection.