AI Wireless Noise Prediction for Platform EMI Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As computing devices become smaller and operate at higher frequencies, they experience increased platform noise from electromagnetic interference, which negatively impacts wireless network throughput and performance, and existing design-time mitigation methods are insufficient.
Innovation Solution
A computing device that monitors platform activity and uses a trained machine learning model to predict radio frequency noise, allowing for dynamic noise mitigation through techniques such as noise whitening, notch filtering, and adaptive scheduling to improve wireless performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If device form factors are made smaller and components operate at higher frequency, then device compactness and processing speed are improved, but electromagnetic interference and platform noise increase
Solution Approach 1:
The system dynamically adjusts platform activity based on real-time noise predictions. The machine learning model continuously monitors platform noise levels and adjusts component operation accordingly, transitioning from static design-time mitigation to dynamic runtime adaptation. This allows the system to optimize performance while managing electromagnetic interference adaptively.
Solution Approach 2:
The system changes operational parameters of platform components based on predicted noise levels. When high noise is predicted in certain frequency bands, the system adjusts operating frequencies, timing, or power levels of components like memory devices and processors to mitigate interference while maintaining overall system performance.
2Reliability
If design-time mitigation methods are used, then initial noise reduction is achieved, but dynamic noise adaptation and wireless performance are insufficient
Solution Approach 1:
The system implements a feedback loop where the machine learning model continuously monitors actual platform noise levels, compares them with predictions, and uses this information to refine future noise predictions and mitigation strategies. This closed-loop approach enables continuous optimization of wireless performance based on real-world conditions.
Solution Approach 2:
The machine learning model performs preliminary noise prediction before wireless transmissions occur. By predicting platform noise levels in advance, the system can proactively adjust component operation to minimize interference, rather than reacting to interference after it has degraded wireless performance.
3Object-affected harmful factors
If traditional noise mitigation through antenna placement and shielding is used, then physical noise reduction is achieved, but adaptability to changing platform activity is limited
Solution Approach 1:
The system uses its own platform activity data and noise measurements to train and improve its machine learning model. The device serves itself by collecting operational data, predicting its own noise emissions, and autonomously adjusting its operation to mitigate interference, eliminating the need for external calibration or manual configuration.
Solution Approach 2:
The system replaces physical/mechanical noise mitigation approaches (antenna placement, shielding) with an information-based approach using machine learning predictions. Instead of relying solely on physical barriers and fixed designs, the system uses software-based noise prediction and adaptive scheduling to dynamically manage electromagnetic interference.
Data Source
AI summary
Technologies for dynamic wireless noise mitigation include a computing device having a wireless modem and one or more antennas. The computing device activates one or more components of the computing device, monitors platform activity, and measures wireless noise received by the antennas. The computing device trains a noise prediction model based on the platform activity and the measured noise. The computing device may monitor platform activity and predict a noise prediction with the noise prediction model based on the monitored activity. The computing device may mitigate wireless noise received by the wireless antennas based on the noise prediction. The computing device may provide the noise prediction to the wireless modem. Other embodiments are described and claimed.