Software-defined network (SDN) and network function virtualization (NFV) are acknowledged as the most promising technologies to effectively allocate resource for network service. A service function chain (SFC), which can deploy virtualized network functions (VNFs) and chain them with associated flows allocation, can be used to represent each network service owing to the introduction of the SDN/NFV technology. Co-hosted applications on multiple Internet of Things terminals have dynamic and time-varying service requirements, in order to allocate network resources optimally and meet the end-to-end delay requirements of services, sufficient strategies are required to satisfy the continuously changing service demands. In this article, a deep learning model that combines the multitask regression layer above the graph neural networks is first presented to predict the future resource requirements of each VNF instance. The SFC deployment problem is then solved using the integer nonlinear programming (INLP) approach, and a novel prediction-assisted Viterbi algorithm is presented to overcome the scalability problem of the INLP approach. According to the simulation findings, the proposed deep model provides a minimum of a 6.2% improvement in prediction accuracy over baseline prediction models, and the proposed SFC deployment strategy has been demonstrated to deliver better performance in terms of acceptance ratio and revenue, compared to the current passive deployment algorithms.

With the emerge of the network function virtualization (NFV) and software-defined network (SDN), the SDN/NFV-enabled network has been recognized as one of the most promising technologies to efficiently achieve resource allocation for network service. By introducing the SDN/NFV technology, each service can be represented by a service function chain (SFC), which can deploy the virtualized network functions (VNFs) and chain them with corresponding flows allocation. Considering the dynamic and complex nature of mobile terminals in cloud networks, how to efficiently embedding SFCs remains as a challenging problem. However, the traditional methods (e.g., exact, heuristic, meta-heuristic, and game, etc.) are subjected to the complexity of cloud network scenarios with dynamic network states, high-speed computational requirements, and enormous service requests. Recent studies have shown that deep reinforcement learning (DRL) is a promising way to deal with the limitations of the traditional methods. However, DRL agent training easily suffers from the problem of slow convergence performance. In order to overcome this narrow, in this paper, we design a novel DRL framework based on the enhanced deep deterministic policy gradient (E-DDPG) for the efficient SFC embedding in the dynamic and complex cloud network scenarios. Simulation results validate the high efficiency of the proposed DRL framework as it not only converges faster than currently baseline algorithms, but also reduces the end-to-end delay down to at least 28.3% compared to the benchmarks. All our proposed algorithms and code are available at https://github.com/ jn-z/.

Multitask joint learning technology continues gaining more attention as a paradigm shift and has shown promising performance in many applications. Depth estimation and semantic understanding from monocular images emerge as a challenging problem in computer vision. While the other joint learning frameworks establish the relationship between the semantics and depth from stereo pairs, the lack of learning camera motion renders the frameworks that fail to model the geometric structure of the image scene. We make a further step in this article by proposing a multitask learning method, namely DPSNet, which can jointly perform depth and camera pose estimation and semantic scene segmentation. Our core idea for depth and camera pose prediction is that we present the rigid semantic consistency loss to overcome the limitation of moving pixels from image reconstruction technology and further infer the segmentation of moving instances based on them. In addition, our proposed model performs semantic segmentation by reasoning the geometric correspondences between the pixel semantic outputs and the semantic labels at multiscale resolutions. Experiments on open-source datasets and a video dataset captured on a micro-smart car show the effectiveness of each component of DPSNet, and DPSNet achieves state-of-the-art results in all three tasks compared with the best popular methods. All our models and code are available at https://github.com/jn-z/DPSNet: Multitask Learning Using Geometry Reasoning for Scene Depth and semantics.

With the rapid development of emitter individual identification technology in cognitive radio networks, electromagnetic emitter individual target identification based on deep learning has received much attention. However, the confusion of unintentional features (i.e., individual fingerprint features) and modulation features resulting from the received signal might lead to low identification accuracy. In order to address this narrow, we propose an emitter individual identification network based on the competitive collaboration framework, called Specific Emitter Identification with Adaptive Decomposition and Extraction of individual fingerprint features (SEI-ADE), which can adaptively decompose and extract individual fingerprint features. Firstly, a signal adaptive decomposition network is proposed to distinguish the emitter signal and the interference signal by adopting the gradient inversion layer and the non-sequential characteristics of the signal. Then, in order to distinguish and extract corresponding features, the feature extractor and the training loss constraints are constructed for individual fingerprint feature signals, modulation signals, and external emitter interference signals, respectively. The proposed framework can continuously adjust the gradient loss, classification loss, and timing coding contrast loss, thus minimizing the entire training loss. For the separation of the modulation signal and individual fingerprint feature signal, the signal is transformed into the feature domain, and a mask prediction network is proposed to locate the domain of the individual fingerprint feature. The obtained experimental results show the outstanding performance of our proposal, compared with the current benchmarks. All our models and code are available at https://github.com/jn-z/SEI-ADE.

Point Cloud Registration (PCR) has been viewed as an essential part of photogrammetry, remote sensing, and autonomous robot mapping. Existing methods are either sensitive to rotation transformations, or rely on feature learning networks with poor generalization. We propose a novel outdoor point cloud registration algorithm, including preprocessing, yaw angle estimation, coarse registration, and fine registration (in short, PYRF-PCR). Specifically, the preprocessing effectively eliminates the interference of ground point clouds to PCR. The proposed yaw angle estimator solves large yaw-angle matching via a cross-correlation function that converts the focus from the yaw angle estimation to the LiDAR horizontal angular resolution analyses. Then, by using frequency distribution histograms, we improve the fast point feature histogram algorithm to filter the point clouds with a more stable density. For the fine registration, an improved iterative closest point based on target centroid distance is proposed, which reduces the running time and the search range between two point clouds. To validate the widespread applicability of PYRF-PCR, we experimented on both the open-source dataset (KITTI) and the local campus scene dataset. On the KITTI dataset, experimental results illustrate that the PYRF-PCR can achieve state-of-the-art results compared with the existing best methods. On the local scene datasets, the higher quality matching in different types of target point clouds reflects the generalization ability of PYRF-PCR. All our models and code are available at https://github.com/jn-z/PYRF-PCR/tree/master

Specific emitter identification (SEI) separates the radio frequency fingerprint (RFF) from signals, which is of great significance in solving Internet of Things (IoT) security problems. However, the scarcity of high-quality, diverse, and labeled data in real-world scenarios limits the application of SEI. Under such conditions, the SEI is referred to as few-shot SEI (FS-SEI). To surmount this challenge, we propose a diffusion model-based data augmentation method capable of generating a substantial volume of diverse, high-quality data. Specifically, we develop a multi-scale convolutional block attention module denoising diffusion probabilistic model (MSCBAM-DDPM), which enhances feature capture capabilities, laying the foundation for the generation of diverse data. Furthermore, we propose an adaptive two-stage multi-domain loss function that guides the model to learn the characteristics of the original data and further derive other similar features, thereby achieving the goal of generating diverse and high-quality data. Finally, we theoretically derive the feasibility of the proposed loss function and further demonstrate the excellent diversity and quality of the data generated by our method, as well as its considerable gain for FS-SEI, through extensive experiments on real-world signal datasets.