Last mile delivery by drones is a core component of innovative logistics systems, relying heavily on AI models for essential operations such as path planning and object recognition. However, models trained on region specific data often experience significant performance degradation when deployed in unfamiliar environments due to geographic domain shifts. This limitation impedes the rapid deployment of logistics networks and hinders model adaptation. Federated Learning (FL), as a distributed machine learning paradigm, enables multiple clients with diverse data to collaborate in training a global model. Nevertheless, within Mobile Edge Computing (MEC) environments, FL faces critical challenges, including data bias, high drone mobility, and intermittent communication windows between drones and edge servers. This paper proposes FedMO, a mobility aware FL framework for drone based last mile delivery with edge cloud collaboration. FedMO introduces a novel algorithmic insight by treating the drone’s flight path as a unified proxy for both communication reliability and data distribution heterogeneity. The framework implements a synergistic three stage selection policy that jointly optimizes connectivity success, data value, and resource efficiency. This effectively transforms mobility from a disruption risk into a diversity enhancing asset. Experimental results using real world drone video datasets demonstrate that FedMO improves convergence speed by approximately 15% compared to baseline methods with only 30%-40% client selection. With equivalent client participation, FedMO achieves a 25% improvement in convergence speed over the FedAvg algorithm.

A primary challenge in network defense is to mine potential attack campaigns from massive, continuously arriving alerts and telemetry data in real time. This paper presents HoSig-Align I, an edge side unsupervised method designed for network streams to discover homologous blocks. Our approach innovatively fuses heterogeneous features like Internet Protocol (IP) and Payload into a unified representation while strictly excluding temporal information from it, only incorporating time via a dual time scale decay model during graph construction to capture temporal proximity. A density robust similarity is computed using an isolation style random partition forest, leading to a sparse k-Nearest Neighbors (k-NN) graph. The stream is then accurately segmented into internally cohesive and mutually isolated homologous blocks through spectral ordering and contrastive change point detection. Each block is encoded into a lightweight HoSig signature, forming the basis for cross organizational collaboration. Experiments on real network streams show that HoSig-Align I identifies coherent attack campaign blocks and improves separation and boundary clarity over baselines, while meeting low-latency and low-overhead requirements for edge processing.

Subspace clustering is one of the most popular clustering methods due to its effectiveness. Although subspace clustering methods have been demonstrated to achieve promising performance, they still lack interpretability, especially when handling high-dimensional complicated data. To bridge this gap, this paper focuses on the interpretability of subspace clustering and proposes a novel interpretable subspace clustering method. Our goal is to answer two key questions about the interpretability in subspace clustering: 1) when handling an individual sample, which features should work for this sample? 2) Which cluster or subspace will the features that work put this sample into? To answer these two questions, we design two new interpretability regularized terms and plug them into the subspace clustering. In this way, we show that interpretability can be used to improve the clustering performance in turn. Extensive experiments on benchmark data sets demonstrate the effectiveness of our method in terms of clustering performance and interpretability.

Unmanned Aerial Vehicles (UAVs), play a significant role in the advancement of pervasive systems by providing efficient, scalable, and innovative solutions in various sectors, such as smart cities or location-based services. However, the current UAV delivery scenario presents various challenges for recipients, including lengthy identity verification processes, privacy concerns, and risks of fraud and theft. In response to these issues, this paper proposes an innovative system that leverages Blockchain technology and Adversarial Machine Learning (AML) to tackle these problems effectively. The proposed system streamlines the verification process, enhances privacy safeguards, and reduces fraud risks. The integration of AML is crucial as it enables users to have greater control over their personal data, boosting privacy and security. AML also plays a critical role in this system by creating test scenarios that reinforce the machine learning model against adversarial threats, ensuring its precision and dependability in the face of malicious manipulations. The paper also provides details on the practical implementation and evaluation of this system in real-life adversarial situations. The evaluation results demonstrate superior performance on selected metrics, highlighting the potential of this system as an effective solution for verifying recipients in UAV delivery.

FedShufde: A privacy preserving framework of federated learning for edge-based smart UAV delivery system

Artificial Intelligence of Things (AIoT) systems are playing an important role in scenarios such as smart factories, smart healthcare, and smart logistics. Edge Computing reduces the network latency by pushing compute and storage resources near the IoT devices. However, the massive of data and task requests from IoT devices and datacenters raise the optimization requirement of schedule plans. Existing studies consider either task scheduling or data placement problems. They ignore the complex relationship between data and tasks leading to an increase the task completion time and energy consumption. Therefore, this paper first formalizes the joint task scheduling and data placement problem as a constrained multi-objective optimization model. Then, a Learns to Improve (L2I) algorithm is proposed, which is a reinforcement learning-based algorithm for task scheduling and data placement to minimize the task completion time and transmission energy consumption of IoT devices. In the L2I algorithm, we design a set of low-level improvement operators to generate new schedule plans to speed up the selection process of the optimal schedule plan. The simulation experiments show that the proposed algorithm effectively outperforms traditional strategies in solving task scheduling and data placement problems.

Biometric technology has driven the rise of Biometrics as a Service (BaaS) due to its unique security and convenience. However, in the traditional cloud-based BaaS systems, raw biometric data leakage and biometric efficiency issues are still concerns, which may reduce user trust and engagement in biometric services. In this paper, we propose an innovative BaaS system named Bio-CEC in a Cloud-Edge Cooperative environment. Bio-CEC features a novel biometric template transformation scheme, rooted in multivariate polynomial transformation and random virtual feature replacement. This innovative scheme enables the revocation and regeneration of biometric templates, enhancing security in case of system breaches. For the biometric template protect scheme, a template matching algorithm using filtering operations is proposed, aiming to facilitate secure and accurate authentication within the transformation domain. Our comprehensive experiments and in-depth safety analysis verify the superiority of Bio-CEC. The results clearly demonstrate that Bio-CEC outperforms traditional cloud-based biometric system and provides a safer and more efficient solution for practical biometric system applications.

With the rise of smart city applications, the accessibility of users’ location data by smart devices has increased significantly. However, this poses a privacy concern as attackers can deduce personal information from the raw location data. In this paper, we propose a framework to collect user location data while ensuring local differential privacy (LDP) in the last-mile delivery system of Unmanned Aerial Vehicles (UAVs) within an edge computing environment. Firstly, we obtain the user location distribution Quad-tree by employing a region partitioning method based on Quad-tree retrieval in the specified data collection area. Next, the user location matrix is retrieved from the obtained Quad-tree, and we perturb the user location data using an LDP perturbation scheme on the location matrix. Finally, the collected data is aggregated using blockchain to evaluate the utility of the dataset from various regions. Furthermore, to validate the effectiveness of our framework in a real-world scenario, we conduct extensive simulations using datasets from multiple cities with varying urban densities and mobility patterns. These simulations not only demonstrate the scalability of our approach but also showcase its adaptability to different urban environments and delivery demands. Finally, our research opens new avenues for future work, including the exploration of more sophisticated LDP mechanisms that can offer higher levels of privacy without significantly compromising the quality of service. Additionally, the integration of emerging technologies such as 5G and beyond in the edge computing environment could further enhance the efficiency and reliability of UAV-based delivery systems, while also offering new challenges and opportunities for privacy-preserving data collection and analysis.

Federated learning is a new distributed privacy-preserving learning paradigm which perfectly meets the requirements of many large service systems such as banking, healthcare, and smart city. Meanwhile, person re-identification, as a technology to associate the images of the same person from different data sources, has been widely used in many smart services such as smart logistics, smart surveillance, and many public searching and rescue missions. Therefore, it is a promising solution to use federated learning for person re-identification to improve the model accuracy while protecting the data privacy. However, the common problem of non-independent and identically distributed (Non-IID) data with heterogeneous clients in federated learning often causes undesirable model accuracy. To address such a problem, in this paper, we propose a novel strategy named federated learning with data augmentation for person re-identification (Fed4ReID). Specifically, to alleviate the impact of Non-IID data, we utilise a pre-trained DCGAN (Deep Convolutional Generative Adversarial Network) model for data augmentation at each edge servers. Experiments on public datasets show that our proposed strategy can outperform baseline method in general accuracy.

AIoT (Artificial Intelligence of Things) which integrates AI and IoT has received rapidly growing interest from the software engineering community in recent years. It is crucial to design scalable, efficient, and reliable software solutions for large-scale AIoT systems in edge computing environments. However, the lack of effective service management including the support for service collaboration, AI application, and data security in the edge, has seriously limited the development of AIoT systems. To seal this gap, we propose EXPRESS 2.0 which is an intelligent service management framework for AI oT in the edge. Specifically, on top of the existing EXPRESS platform, EXPRESS 2.0 includes the intelligent service collaboration management module, AI application management module, and data security management module. To demonstrate the effectiveness of the framework, we design and implement a last-mile delivery system using both UAVs (Unmanned Aerial Vehicles) and UGVs (Unmanned Ground Vehicles). The EXPRESS 2.0 is open-sourced at https://github.com/ISEC-AHU/EXPRESS2.0. A video demonstration of EXPRESS 2.0 is at https://youtu.be/GHKD_VvJD88.

Edge computing is becoming more and more popular in both academics and industries. With the booming of edge computing technology, the Unmanned Aerial Vehicle (UAV) based delivery system is expected to achieve higher efficiency and low latency. However, the UAV often collects user-specific data during the delivery process, the data-leakage or security/privacy breaching could occur during the data-sharing process between the edge nodes and UAV devices. Privacy-preserving issues are further refraining from the popularity of the UAV-based logistic systems. It is believed that the UAV tracking and identity verification system can provide imminent access-level security and privacy protection, which is urgently required to eliminate the practical concerns under the edge computing-based environment. To the best knowledge of authors, for the first time, this paper proposes a Self-Sovereign Identity (SSI) integrated framework with the latest Blockchain technology for UAV-based delivery system. It is expected to protect the edge computing-based UAV delivery system against security flaws and privacy concerns. In this work, the benefits of using SSI with Blockchain technology are analyzed, the efficiency of identifying and authenticating UAVs and their respective users is further experimented and discussed. The experimental results show that the integrated SSI framework with Blockchain can effectively improve the efficiency of the user identity management system as well as the identity verification process in the delivery process.

As the latest computing paradigm, edge computing has attracted increasing attention from both academia and industry in recent years. It significantly affects the design and development of many smart systems and challenges conventional security solutions. Therefore, a new security framework targeted for edge computing-based smart systems is urgently required. In this paper, we investigate various security issues in the edge computing-based Unmanned Aerial Vehicle (UAV) delivery system which is a typical system in smart logistics. Specifically, we focus on security issues related to abnormal intrusion detection, various security attacks, and authentication/access control. To address these security issues, we propose A2DSEC which is a novel security framework characterized by the capabilities such as detection, defense, and authentication. With A2DSEC, we can guarantee the security of user identity authentication, and provide effective early warnings so that corresponding security measures can be taken in a timely fashion. To verify the effectiveness of A2DSEC, we test and implement the core part of the framework on a real-world edge computing-based UAV delivery system. The experiment results show that the A2DSEC can effectively ensure the security of the UAV delivery system via the timely detection of a variety of attacks as well as unknown potential attacks.