LoRa technology has been extensively implemented in various IoT applications, offering widespread low-power connectivity for millions of nodes across thousands of logical channels. However, current LoRa networks lack an efficient mechanism for monitoring channel activity across these numerous channels, which prevents network operators from effectively detecting physical layer activities and implementing additional functionalities (e.g., channel access control). Existing solutions either involve complex iterations over each logical channel or fail to detect extremely weak packets in low SNR conditions. These limitations affect their scalability and robustness in monitoring the vast number of logical channels available in the LoRa spectrum. To address this issue, this paper introduces SlideLoRa, an innovative packet detection method that enables detection across all logical channels under various channel conditions. SlideLoRa consolidates the complete energy of LoRa symbols using an expanded demodulation window combined with a fine-grained sliding window, effectively reconstructing the distorted frequency-domain information of LoRa packets. To achieve this, SlideLoRa incorporates a series of novel solutions, including peak tracking in low SNR, peak sequence matching, peak extraction, and packet parameter retrieval. Experimental results demonstrate that SlideLoRa enhances packet detection capability by 1.7× compared to the state-of-the-art.

Federated learning (FL) has become a prominent paradigm for collaborative model training while ensuring data privacy. However, in resource-constrained environments, such as the Internet of Things (IoT), FL faces a distinct challenge from Lazybone attackers, who compromise system performance by providing low-quality data or conducting minimal local training to reduce their computational burden. In this article, we propose Fedeval, a novel multi-dimensional evaluation framework designed to defend against Lazybone attacks. Fedeval leverages a server-side base validation dataset and a base model to assess the quality and relevance of client contributions through gradient inversion, and it compares client-uploaded gradients with an honest baseline to detect training inconsistencies. By assigning adaptive importance scores based on client contributions, Fedeval enhances the robustness of FL by mitigating the impact of non-contributing participants. We also provide a theoretical analysis of Fedeval’s convergence properties and validate its effectiveness through extensive experiments on four datasets and two attack scenarios. Our results demonstrate that Fedeval significantly accelerates convergence and improves accuracy by up to 13% compared to traditional methods.