Heteroscedastic Neural Networks for Path Loss Prediction with Link-Specific Uncertainty
By: Jonathan Ethier
Potential Business Impact:
Helps phones predict signal strength better.
Traditional and modern machine learning-based path loss models typically assume a constant prediction variance. We propose a neural network that jointly predicts the mean and link-specific variance by minimizing a Gaussian negative log-likelihood, enabling heteroscedastic uncertainty estimates. We compare shared, partially shared, and independent-parameter architectures using accuracy, calibration, and sharpness metrics on blind test sets from large public RF drive-test datasets. The shared-parameter architecture performs best, achieving an RMSE of 7.4 dB, 95.1 percent coverage for 95 percent prediction intervals, and a mean interval width of 29.6 dB. These uncertainty estimates further support link-specific coverage margins, improve RF planning and interference analyses, and provide effective self-diagnostics of model weaknesses.
Similar Papers
Multidimensional Distributional Neural Network Output Demonstrated in Super-Resolution of Surface Wind Speed
Machine Learning (CS)
Helps computers guess better by showing how sure they are.
Reciprocity-Aware Convolutional Neural Networks for Map-Based Path Loss Prediction
Machine Learning (CS)
Improves phone signal prediction for all directions.
Uncertainty Estimation using Variance-Gated Distributions
Machine Learning (CS)
Makes AI more sure about its answers.