Due to the reliance of task processing on specific service models, service caching has emerged as a promising technology for enhancing low-latency performance in vehicular edge computing (VEC) networks. When vehicular long-term service caching mismatches real-time task requirements, the mobility and autonomy of service vehicles enable them to fetch service models from providers actively. However, highly dynamic topology and resource constraints of VEC networks call for a more adaptable approach for jointly designing vehicular service fetching and task offloading. To this end, this paper aims at jointly optimizing active service fetching, task offloading, and communication and computing resource allocation to boost low-latency performance in two-tier caching-assisted VEC networks. In doing so, a total task completion delay minimization problem is formulated, where the social-mobility-aware network topology, vehicle classification, delay tolerance requirements, and resource constraints are taken into account. Since it is a mixed-integer non-linear programming (MINLP) problem, it is typically NP-hard. To solve it effectively, we decompose it into three subproblems, which can be efficiently solved by an iterative algorithm based on block coordinate descent (BCD) with low complexity. Extensive simulation results demonstrate that our proposed algorithm achieves fast convergence and reduces the task completion delay by 9.27%-59.77% compared to five representative baselines.

The number of categories of instances in the real world is normally huge, and each instance may contain multiple labels. To distinguish these massive labels utilizing machine learning, eXtreme Label Classification (XLC) has been established. However, as the number of categories increases, the number of parameters and nonlinear operations in the classifier also rises. This results in a Classifier Computational Overload Problem (CCOP). To address this, we propose a Multi-Head Encoding (MHE) mechanism, which replaces the vanilla classifier with a multi-head classifier. During the training process, MHE decomposes extreme labels into the product of multiple short local labels, with each head trained on these local labels. During testing, the predicted labels can be directly calculated from the local predictions of each head. This reduces the computational load geometrically. Then, according to the characteristics of different XLC tasks, e.g., single-label, multi-label, and model pretraining tasks, three MHE-based implementations, i.e., Multi-Head Product, Multi-Head Cascade, and Multi-Head Sampling, are proposed to more effectively cope with CCOP. Moreover, we theoretically demonstrate that MHE can achieve performance approximately equivalent to that of the vanilla classifier by generalizing the low-rank approximation problem from Frobenius-norm to Cross-Entropy. Experimental results show that the proposed methods achieve state-of-the-art performance while significantly streamlining the training and inference processes of XLC tasks. The source code has been made public at https://github.com/Anoise/MHE.

In this article, we focus on the problem of product defect detection (PDD) in air conditioner (AC) manufacturing. The challenges are twofold: first, the scale of the objects undergoes significant variations, thereby increasing the difficulty of detection; second, the computing power of terminals is limited, and it may be difficult to meet the quasi-real-time detection requirements. Therefore, to improve detection accuracy and speed, a lightweight object detection model tailored for deployment on the compact wireless Raspberry Pi, is proposed: a deep cascaded U-shape network is presented to effectively capture both global context and local details of objects, which can reduce the feature redundancy and the number of the model parameters. An adaptive multiscale squeeze-and-excitation is designed for feature reuse and fusion, enhancing both detection accuracy and efficiency. Then, to meet the quasi-real-time detection demands and better utilize end-edge-cloud resources in industrial internet of things (IIoTs), an actor–critic-based dynamic offloading (ACDO) algorithm is proposed to minimize the long-term cumulative time of task detection. ACDO utilizes the elapsed time as the reward to directly optimize the mixed variables of the task to achieve efficient offloading. The proposed methods are verified at an AC manufacturing line, which demonstrates accurate and quasi-real-time defect detection, achieving a 64% reduction in runtime and a 1.2% improvement in average mean average precision. In addition, we publish two PDD datasets to accelerate the related research.

Edge caching has emerged as a promising technique against latency caused by explosive growth of mobile data traffic through caching popular contents at the edge networks. However, the dynamically changing content popularity nature and limited caching capacity make it challenging to design an effective caching scheme to reduce latency. To solve this, a learning-based hierarchical edge caching (LHEC) scheme is proposed in this work. We first propose a novel deep learning architecture, namely Stacked Autoencoder-Long Short Term Memory Network (SAE-LSTMNet) to capture both the correlation of the request patterns among different content and the periodicity in time domain to improve the prediction accuracy of the content popularity. Then, to predict the popularity of these newly-added contents, a dynamic content catalog is introduced and a similarity-based content popularity prediction (SCPP) approach is proposed. Based on the content popularity prediction, a hierarchical edge caching optimization problem is formulated to minimize the average content downloading latency. Since the formulated problem is NP-hard and difficult to be solved, a low-complexity algorithm is proposed to obtain the near-optimal solutions. Simulation results show that the proposed content popularity prediction approach outperforms up to 6.36% in terms of the mean absolute error compared with the state-of-the-art methods and the proposed LHEC scheme reduces the average downloading latency at about 5.3%~7.9% compared with those existing caching schemes.

Wireless traffic prediction has drawn increasing research interests as it can provide guidance to the network optimization. With the predicted information, one can preassign the resources on demand and perform network congestion control adaptively. The network efficiency is therefore enhanced. However, the wireless traffic prediction in the context of mobile scenario, such as Internet of Vehicles (IoVs), is still a challenge issue. The mobile nature of devices, which dynamically changes the topology of network, would brings difficulties to the prediction. This paper focuses on the deep learning based wireless traffic prediction in the IoVs scenario. We first propose a novel method to match up the movement- and communication-behavior of users, by merging two independent datasets on the trajectories of vehicles and communication traffic volumes together. Then a novel STeP-UNet is proposed, in which the SpatioTemporal Partial (STeP) Convolutional Neural Network module is embedded to capture cross-domain features of the wireless traffic pattern, and the UNet structure is utilized to realize the skipping connection from front layer to back layer to fuse different resolutions. Experimental results confirms the promising performance of the proposed model, where 4%~8% performance improvement over other benchmark methods can be achieved.

Due to the booming of various devices in Internet-of-Things (IoT) networks, more data should be transmitted over the networks, which will thereby consume more transmission bandwidth and more transmit power. Compressed data aggregation (CDA) has been proposed as an effective way to reduce the amount of the collected data in IoT networks. Although adopting compressed sensing (CS), CDA can sample the source data efficiently, and sparse data reconstruction is still a big challenge. In this work, inspired by this, we propose a learning-based sparse data reconstruction scheme by jointly utilizing CS and deep learning. Our objective is to reduce the volume of data to be transmitted over IoT networks without losing reconstruction accuracy. A deep CS network is designed by adopting an end-to-end learning method to build a measurement matrix and an efficient and high-accuracy reconstruction network. To show the performance of the proposed scheme, six data sets with different structured sparse models and a real sensor data set are utilized in doing experiments. The performance of the proposed scheme in terms of mean-squared error, peak-signal-noise-ratio, and structural similarity is investigated. The results demonstrate the effectiveness of the proposed scheme in reconstruction accuracy for given compression ratio. The results also show that the proposed scheme is suitable for the process of CDA, thus can effectively reduce the amount of data to be transmitted in IoT networks.

When doing image classification, the core task of convolutional neural network (CNN)-based methods is to learn better feature representation. Our analysis has shown that a better feature representation in the layer before softmax operation (BSM-layer) means a better feature embedding location that has a larger distance to the separating hyperplane. By defining this property “Location Property” of CNN, the core task of CNN-based methods can be regarded as to find out the optimal feature embedding location in the BSM-layer. In order to achieve this, in this work, we first propose two feature embedding directions, principal embedding direction (PE-direction) and secondary embedding direction (SE-direction). And then, we further propose a loss-based optimization framework, location property loss (LP-loss), which can make feature representation move in the PE-direction and the SE-direction simultaneously during the training phase. LP-loss consists of two parts, LPPEand LPSE, where LPPEfocuses on PE-direction, and LPSEfocuses on SE-direction. Any loss function focusing on these two embedding directions can be chosen as LPPEand LPSE. Based on the analysis that softmax, L-softmax, and AM softmax can make the feature representation move in PE-direction to a different extent, any of them can be chosen as LPPE. Since there is no existing works can fulfill the purpose of LPSE, a novel loss, secondary optimal feature plane loss (S-OFP loss), is developed. S-OFP loss is designed to make feature representations belonging to the same category embed onto their corresponding S-OFP. It is proved that S-OFP loss is the optimal feature plane in the SE-direction. Experiments are done with shallow, moderate, and deep models on four benchmark data sets, including the MNIST, SVHN, CIFAR-10, and CIFAR-100, and results demonstrate that CNN models can obtain remarkable performance improvements with LPsoftmax, S-OFP and LPAM softmax, S-OFP, which verify the effectiveness of location property.

In this paper, a novel three-dimensional (3D) non-stationary geometry-based stochastic model (GBSM) for the fifth generation (5G) and beyond 5G (B5G) systems is proposed. The proposed B5G channel model (B5GCM) is designed to capture various channel characteristics in (B)5G systems such as space-time-frequency (STF) non-stationarity, spherical wavefront (SWF), high delay resolution, time-variant velocities and directions of motion of the transmitter, receiver, and scatterers, spatial consistency, etc. By combining different channel properties into a general channel model framework, the proposed B5GCM is able to be applied to multiple frequency bands and multiple scenarios, including massive multiple-input multiple-output (MIMO), vehicle-to-vehicle (V2V), high-speed train (HST), and millimeter wave-terahertz (mmWave-THz) communication scenarios. Key statistics of the proposed B5GCM are obtained and compared with those of standard 5G channel models and corresponding measurement data, showing the generalization and usefulness of the proposed model.

As one of the indispensable components in convolutional neural network (CNN), loss function assists in updating parameters of CNN models during the training phase. Generally, different loss functions can assist convolutional neural network (CNN) to learn different feature representations, and different feature representations can be treated as different knowledge learned from objects. In this paper we introduce a novel training framework, namely Loss Transferring (LT), to improve the generalization ability of CNN. LT contains multiple training phases, and each training phase uses a different loss function. Under this framework, CNN models can combine different knowledge of objects by transferring the knowledge learned via one loss function to another. LT contains two components, i.e., loss function set and training strategy. In order to build appropriate loss function set, we establish two basic guides. And according to these basic guides, we design a new loss function in the last layer of CNN models (layer before softmax operation), namely Near Classifier Hyper-Plane (N-CHP) loss, which makes the learned object features belonging to the same category have the minimum intra-class distance and be near the classifier hyper-plane. Based on the two loss function set (MSE, softmaxl and (N-CHP, softmaxl, we setup two specific training methods, LTMSE, softmax and LTN-CHP, softmax, which can be universally applied to different CNN models with low additional computation cost. Meanwhile, two training strategies, multi-phase strategy 1 and multi-phase strategy 2, are further proposed to improve the training efficiency of LT. Extensive experimental results on shallow, moderate and deep models with four benchmark datasets, including MNIST, SVHN, CIFAR-10 and CIFAR-100, demonstrate that CNN models can bring obvious performance improvements when working with LTMSE, softmax and LTN-CHP, softmax, which verifies the effectiveness of LT and the proposed two basic guides.

In cellular Internet of Things networks, massive access and highly dynamic traffic of machine type communication devices may cause network congestion, heavy energy consumption or even service unavailability. To reduce the amount of data transmission and improve the efficiency of data aggregation in sporadic machine type communications, we develops a two-stage compressed data aggregation scheme by jointly utilizing compressive sensing and sparse autoencoder. A deep compressive sensing network (DCSNet) is designed by using deep learning method to reconstruct signals from compressive measurements. Experimental results demonstrate that, our scheme can effectively reduce the data traffic, achieve higher reconstruction accuracy in different sparse conditions of signal and brake the obstacles that compressive sensing method has a large reconstruction error when the measured data is small.

Machine (deep) learning-enabled accurate traffic modeling and prediction is an indispensable part for future big data-driven intelligent cellular networks, since it can help autonomic network control and management as well as service provisioning. Along this line, this paper proposes a novel deep learning architecture, namely Spatial–Temporal Cross-domain neural Network (STCNet), to effectively capture the complex patterns hidden in cellular data. By adopting a convolutional long short-term memory network as its subcomponent, STCNet has a strong ability in modeling spatial–temporal dependencies. Besides, three kinds of cross-domain datasets are actively collected and modeled by STCNet to capture the external factors that affect traffic generation. As diversity and similarity coexist among cellular traffic from different city functional zones, a clustering algorithm is put forward to segment city areas into different groups, and consequently, a successive inter-cluster transfer learning strategy is designed to enhance knowledge reuse. In addition, the knowledge transferring among different kinds of cellular traffic is also explored with the proposed STCNet model. The effectiveness of STCNet is validated through real-world cellular traffic datasets using three kinds of evaluation metrics. The experimental results demonstrate that STCNet outperforms the state-of-the-art algorithms. In particular, the transfer learning based on STCNet brings about 4%~13% extra performance improvements.

In this paper, a three-dimensional (3D) non-stationary wideband multiple-input multiple-output (MIMO) channel model based on the WINNER+ channel model is proposed. The angular distributions of clusters in both the horizontal and vertical planes are jointly considered. The receiver and clusters can be moving, which makes the model more general. Parameters, including number of clusters, powers, delays, azimuth angles of departure (AAoDs), azimuth angles of arrival (AAoAs), elevation angles of departure (EAoDs), and elevation angles of arrival (EAoAs) are time-variant. The cluster time evolution is modeled using a birth-death process. Statistical properties, including spatial cross-correlation function (CCF), temporal autocorrelation function (ACF), Doppler power spectrum density (PSD), level-crossing rate (LCR), average fading duration (AFD), and stationary interval are investigated and analyzed. The LCR, AFD, and stationary interval of the proposed channel model are validated against the measurement data. Numerical and simulation results show that the proposed channel model has the ability to reproduce the main properties of real non-stationary channels. Furthermore, the proposed channel model can be adapted to various communication scenarios by adjusting different parameter values.

Most channel models in the literature are based on the assumption that the mobile station (MS) moves along a straight line with a constant speed. In a realistic environment, the MS may experience changes in their speeds and trajectories. In this paper, a three-dimensional (3-D) non-stationary wideband multiple-input multiple-output (MIMO) channel model allowing for velocity variations of the MS is proposed. The parameters are obtained from the WINNER+ channel model to make the simulations more realistic. Statistical properties including spatial cross-correlation function (CCF), temporal autocorrelation function (ACF), and Doppler power spectral density (PSD) are derived and analyzed. Our findings show that a variation of the velocity of the MS has a significant impact on the statistical properties of the channel model. Furthermore, the proposed channel model can be used as a basic framework for future non-stationary channel modeling.