With the booming integration of IoT technology in our daily life applications such as smart industrial, smart city, smart home, smart grid, and healthcare, it is essential to ensure the security and privacy challenges of these systems. Furthermore, time‐critical IoT applications in healthcare require access from external parties (users) to their real‐time private information via wireless communication devices. Therefore, challenges such as user authentication must be addressed in IoT wireless sensor networks (WSNs). In this paper, we propose a secure and lightweight three‐factor (3FA) user authentication protocol based on feature extraction of user biometrics for future IoT WSN applications. The proposed protocol is based on the hash and XOR operations, including (i) a 3‐factor authentication (i.e., smart device, biometrics, and user password); (ii) shared session key; (iii) mutual authentication; and (iv) key freshness. We demonstrate the proposed protocol’s security using the widely accepted Burrows–Abadi–Needham (BAN) logic, Automated Validation of Internet Security Protocols and Applications (AVISPA) simulation tool, and the informal security analysis that demonstrates its other features. In addition, our simulations prove that the proposed protocol is superior to the existing related authentication protocols, in terms of security and functionality features, along with communication and computation overheads. Moreover, the proposed protocol can be utilized efficiently in most of IoT’s WSN applications, such as wireless healthcare sensor networks.
27 June 2022 in Peer-to-Peer Networking and Applications
Although deep neural networks have significantly improved object detection accuracy, manually designing efficient architectures is costly, and NAS networks incur high computational overhead, limiting their applicability in resource-constrained or real-time scenarios. This study aims to develop an efficient neural architecture search method capable of rapidly generating low-latency, high-performance detection networks on edge devices. We propose Knowledge-Guided Neural Architecture Search (KGNAS). First, explicit knowledge from a hardware universal deployment framework is used to enrich the search space; second, implicit knowledge characterizes the impact of memory interactions in hierarchical operator blocks on latency; finally, a zero-cost proxy metric evaluates the feature extraction capability of the backbone without training. Experiments show that KGNAS can design efficient networks for various edge devices within one hour, reducing inference latency by 88% compared to DetNAS with only a 3% accuracy loss.
Deep neural network-based computer vision applications have exploded and are widely used in intelligent services for IoT devices. Due to the computationally intensive nature of DNNs, the deployment and execution of intelligent applications in smart scenarios face the challenge of limited device resources. Existing job scheduling strategies are single-focused and have limited support for large-scale end-device scenarios. In this paper, we present ADDP, an adaptive distributed DNN partition method that supports video analysis on large-scale smart cameras. ADDP applies to the commonly used DNN models for computer vision and contains a feature-map layer partition module (FLP) supporting edge-to-end collaborative model partition and a feature-map size partition (FSP) module supporting multidevice parallel inference. Based on the inference delay minimization objective, FLP and FSP achieve a tradeoff between the arithmetic and communication resources of different devices. We validate ADDP on heterogeneous devices and show that both the FLP module and the FSP module outperform existing approaches and reduce single-frame response latency by 10–25% compared to the pure on-device processing.
This paper proposes a new SynFlood attack target location method based on the Abnormal TCP Connection Graph (ATCG). The method build an Abnormal TCP Connection Graph based on the status of TCP connections. Then the method calculates the Abnormal Source Number (ASN) and Attack Intensity (AI). If such two values exceed the threshold defined in advance, the node which the IP address indicated can be determined as attack target. The simulation results indicate that the method has favorable accuracy and higher data packet processing capability. It can be deployed at the backbone router in a large or medium-sized network.