Test-Time Adversarial Detection and Robustness for Localizing Humans Using Ultra Wide Band Channel Impulse Responses

Keyless entry systems in cars are adopting neural networks for localizing its operators. Using test-time adversarial defences equip such systems with the ability to defend against adversarial attacks without prior training on adversarial samples. We propose a test-time adversarial example detector which detects the input adversarial example through quantifying the localized intermediate responses of a pre-trained neural network and confidence scores of an auxiliary softmax layer. Furthermore, in order to make the network robust, we extenuate the non-relevant features by non-iterative input sample clipping. Using our approach, mean performance over 15 levels of adversarial perturbations is increased by 53.3% for the fast gradient sign method and 60.9% for both the basic iterative method and the projected gradient method when compared to adversarial training.