3D perception is an essential capability of autonomous vehicles. Most state-of-the-art stereo matching networks pursue higher prediction accuracy at the cost of the inference speed. However, high demand on computational resource pushes the cost of hardware, hindering practical applications of stereo matching. In this paper, we propose RMCNet, a novel re-parameterized coarse-to-fine network for stereo matching. RMCNet achieves real time on edge devices while outputs accurate disparity maps. To reduce the computing complexity of 3D convolution in cost aggregation, we propose Mixed Cost Volumes that consist of 4D and 3D cost volumes, enabling efficient optimization of the initial disparity map for real-time inference. An Efficient Feature Matching module is proposed to provide absolute disparity candidates for 3D cost volume. To increase the reliability of 4D and 3D cost volumes, we propose a simple but efficient 3D cost aggregation module and a pyramid 2D cost aggregation module for the aggregation of Mixed Cost Volumes. Moreover, we propose a Re-parameterized Channel Sensing module and a Re-parameterized Disparity-Aware module to replace 2D and 3D convolutional blocks with residual connections. Being evaluated on KITTI 2012 and KITTI 2015, the comprehensive performance of the proposed network is improved by the re-parameterization approach. Outstanding trade-off between accuracy and speed on edge devices over existing SOTA works is achieved. With fast average inference speed on NVIDIA Jetson Nano of 33.74 FPS, the accuracy in terms of 3px-all on KITTI 2012 test set and D1-all on KITTI 2015 test set are 3.01% and 3.41%, respectively.

Unmanned aerial vehicles (UAVs) that are widely utilized for video capturing, processing and transmission have to address jamming attacks with dynamic topology and limited energy. In this paper, we propose a reinforcement learning (RL)-based UAV anti-jamming video transmission scheme to choose the video compression quantization parameter, the channel coding rate, the modulation and power control strategies against jamming attacks. More specifically, this scheme applies RL to choose the UAV video compression and transmission policy based on the observed video task priority, the UAV-controller channel state and the received jamming power. This scheme enables the UAV to guarantee the video quality-of-experience (QoE) and reduce the energy consumption without relying on the jamming model or the video service model. A safe RL-based approach is further proposed, which uses deep learning to accelerate the UAV learning process and reduce the video transmission outage probability. The computational complexity is provided and the optimal utility of the UAV is derived and verified via simulations. Simulation results show that the proposed schemes significantly improve the video quality and reduce the transmission latency and energy consumption of the UAV compared with existing schemes.

Mobile edge computing (MEC) raises the issue of resisting selfish edge attackers that use less computation resources than promised to process offloading tasks or provide faked computation results. In this paper, we present a blockchain based trust mechanism to help MEC address selfish edge attacks and faked service record attacks. This mechanism evaluates the computational performance of the edge devices and broadcasts such information to the neighboring edge devices and mobile devices. By building a reputation assignment method for the edge devices, the edge reputation system chooses the miner of the blockchain, which applies the joint Proof-of-Work and Proof-of-Stake consensus protocol to append a block recording the new service reputations onto the MEC blockchain. We propose a reinforcement learning (RL) based edge central processing unit (CPU) allocation algorithm without knowing the mobile service generation model and the network model in the dynamic edge computing process and a deep RL version to further improve the computational performance. The security performance is analyzed and the performance bound of the edge utility is provided. Experimental results show that this framework suppresses the selfish edge attacks, decreases the response latency and saves the energy compared with a benchmark MEC scheme.

Unmanned aerial vehicles (UAVs) are widely utilized to capture and compress videos of the target area and then transmit the processed videos to the control station (CS) on the ground. The media transmissions in the UAV-aided network face many challenges due to the highly dynamic network topology and limited resources such as bandwidth and energy. This paper introduces a media transmission scheme in the UAV-aided network utilizing reinforcement learning algorithms to efficiently process and transmit the captured video, which is able to improve the quality-of-experience (QoE) and reduce the energy consumption. Exploiting the proposed reinforcement learning algorithm, the UAV dynamically selects the quantization parameter in the source coding process and determines the transmit power without knowing the video transmission model. Simulation results demonstrate that the proposed scheme is capable of achieving a higher video quality and utility with lower energy consumption compared with the state-of-the-art schemes.