GRACE: Gradient-based XAI Scheme for Channel Estimation in Wireless Communications

The support of artificial intelligence (AI) is a key element in future 6G networks, where the concept of native AI will be introduced. Moreover, AI is widely employed in different critical applications such as autonomous driving and medical diagnosis. In such applications, using AI as black-box models is risky and challenging since the logic behind the decision-making methodology of the AI model is unclear. Consequently, it is crucial to understand and trust the decisions taken by these models to ensure their safe and efficient deployment. Tackling this issue can be achieved by developing explainable AI (XAI) schemes that aim to explain the logic behind the black-box model behavior. One solution to achieve this is to identify the relevant model inputs that are contributing the most towards its decisions. In this context, this paper proposes a gradient-based XAI scheme for channel estimation denoted as GRACE that aims to classify the model inputs using gradient backpropagation. Unlike perturbation-based input filtering, the proposed GRACE filtering strategy is based on the internal model architecture. Hence it provides a more efficient and trustworthy input classification. Performance evaluation in terms of bit error rate (BER) shows that identifying the relevant model inputs by the proposed GRACE leads to a better BER performance in comparison to the recently proposed perturbation-based scheme.