Distracted driving behavior is known as a leading factor in road traffic injuries and deaths. Fortunately, rapidly developing deep learning technology has shown its potential in distracted driving detection. Nevertheless, deep learning-based solutions need to collect large amounts of driving data captured by camera sensors in the vehicle, which will cause serious privacy concerns. As a privacy-preserving distributed learning paradigm, federated learning (FL) has achieved competitive performance in many applications recently. Inspired by this, we introduce FL into distracted driving detection tasks. However, we observe that the heterogeneous data distribution across drivers leads to significant performance degradation of the model learned in FL. To address this challenge, we propose a simple and effective federated bidirectional knowledge distillation framework, FedBiKD. Specifically, FedBiKD utilizes the knowledge from the global model in guiding local training to mitigate the issue of local deviation. Meanwhile, the consensus from the ensemble of local models is also employed to fine-tune the aggregated global model for less volatility in training. Our extensive experiments demonstrate the effectiveness of FedBiKD in distracted driving detection. The results show that FedBiKD significantly outperforms other FL algorithms in terms of accuracy, communication efficiency, convergence rate, and stability.

Unmanned aerial vehicles (UAVs) are beginning to make a splash in emergency disaster scenarios owing to its excellent air mobility and flexibility. Considering that large base stations often cannot be deployed to disaster areas in the first place and the variation of communication links between UAVs, we formulate the task scheduling problem for disaster scenarios as a two-stage Lyapunov optimization problem and propose a dispersed computing network consisting of UAVs and ground mobile devices, which is used for collaborative computing. We decouple the long-term stability of the task queues of the nodes in the system in terms of time slots as a deterministic optimization problem by Lyapunov techniques. By jointly optimizing the task size transmitted from the control center to the UAVs, the task size computed locally and offloaded by the UAVs and mobile devices, the energy consumption of the dispersed computation system is minimized while ensuring the stability of the computation queues. The simulation results verify that our proposed algorithm is close to the optimal case in terms of queue stability, and our algorithm is able to reduce the system energy consumption by more than 50% compared to the local computation of UAVs.

Recent breakthroughs in deep neural networks (DNNs) have fueled a tremendously growing demand for bringing DNN-powered intelligence into mobile platforms. While the potential of deploying DNNs on resource-constrained platforms has been demonstrated by DNN compression techniques, the current practice suffers from two limitations: 1) merely stand-alone compression schemes are investigated even though each compression technique only suit for certain types of DNN layers; and 2) mostly compression techniques are optimized for DNNs’ inference accuracy, without explicitly considering other application-driven system performance (e.g., latency and energy cost) and the varying resource availability across platforms (e.g., storage and processing capability). To this end, we propose AdaDeep, a usage-driven, automated DNN compression framework for systematically exploring the desired trade-off between performance and resource constraints, from a holistic system level. Specifically, in a layer-wise manner, AdaDeep automatically selects the most suitable combination of compression techniques and the corresponding compression hyperparameters for a given DNN. Thorough evaluations on six datasets and across twelve devices demonstrate that ${\sf AdaDeep}$ can achieve up to $18.6\times$ latency reduction, $9.8\times$ energy-efficiency improvement, and $37.3\times$ storage reduction in DNNs while incurring negligible accuracy loss. Furthermore, ${\sf AdaDeep}$ also uncovers multiple novel combinations of compression techniques.

Mobile edge computing introduces a novel computing paradigm for mobile devices, reducing execution latency and energy consumption by offloading tasks to edge servers or other idle mobile devices. In this paper, we consider the utility optimization problem of two typical computing tasks, latency-sensitive tasks and latency-tolerant tasks, among multiple mobile devices and base stations. Mobile devices can choose three computing modes to optimize utility: local computing, task allocation to base stations, and task allocation to other mobile devices through device-to-device communication. To address this problem, we formalize it as a potential game for multi-mobile device multi-base station task offloading. Furthermore, we prove the existence of a Nash equilibrium for the modeled potential game and propose a task allocation scheme for hybrid tasks. This scheme maximizes both energy consumption utility and task execution utility by optimizing task offloading mode selection and task execution order scheduling. Simulation results show that our proposed scheme can substantially enhance user utility and has good scalability with the increase of mobile devices.

Message authentication is one of the most effective ways to thwart unauthorized messages from being transmitted in wireless ad-hoc networks such as wireless sensor networks (WSNs) and vehicular ad-hoc networks (VANETs). For this reason, many message authentication schemes have been developed, based on either symmetric-key cryptosystems or public-key cryptosystems. Public-key cryptosystems have simpler key management and are easier to scale, thus are more suitable for message authentication. Among all kinds of public-key cryptosystems, identity based cryptosystem is the most attractive since the public keys can be derived locally. In this paper, we propose an efficient message authentication with enhanced privacy (IMAEP) scheme based on ring signature and identity based signature. We prove that besides unconditional privacy, the proposed scheme can achieve enhanced privacy which can defend against full key exposure attack. We also prove that the proposed scheme can achieve unforgeability against adaptive chosen-message-and-identity attack. Through comparing the proposed scheme with the scheme proposed in [1] with the same level of enhanced privacy, we demonstrate that our scheme has the following advantages: 1) our scheme has much lower computational overhead, 2) our scheme can retain unconditional privacy.