Accurate Noise Source Separation in Pulse-Frequency Modulation-Controlled Power Converters Using Asynchronous Near-Field and Far-Field Measurement Analysis

Aperiodic spike noise signals generated by pulse-frequency modulation (PFM)-controlled direct-current (DC)–DC converters introduce random jitter into the switching timing, rendering it challenging to isolate individual power supply noise sources. Conventional methods based on cross-correlation analysis suffer from inaccuracies owing to the random jitter and aperiodic characteristics inherent in PFM converters. This study proposes a method for accurately separating radiated electromagnetic noise sources from multiple PFM-controlled DC–DC converters. An asynchronous near-field and far-field measurement scheme is adopted owing to its considerably lower implementation cost than that of its synchronous counterpart. The proposed method extracts individual spike noise, and then identifies and associates each with its corresponding source based on the spatial and temporal characteristics captured by the asynchronous measurements. The experimental results demonstrate improved estimation accuracy across a broad bandwidth, thereby enabling the identification of dominant noise sources. This enhanced accuracy supports more efficient mitigation of electromagnetic interference and reduces the complexity and cost of implementing noise suppression. As a result, the time-to-market can be shortened, facilitating the development of cost-effective electronic products.