Search

In this paper, we study the deadline-aware task data offloading in edge-cloud computing systems. The hard-deadline tasks strictly demand to be processed within their delay deadline, whereas the deadline can be relaxed for the soft-deadline tasks. Generally, edge computing aims to shorten the transmission delay between the remote cloud and the end-user, however, at the cost of limited computing capability. Therefore, it is challenging to decide where to offload the hard- and soft-deadline tasks based on the average delay and the service price set by the edge and cloud servers. Both edge and cloud servers aim to maximize their revenue by selling the computational resources at the optimal price. Interestingly, a Wardrop equilibrium is reached, considering that each task is considered independently to be offloaded to a suitable location. The numerical results demonstrate that the proposed price- and deadline-sensitive task offloading policy reaches the equilibrium and finds the optimal location for processing while maximizing the revenue of both edge and cloud servers. [ABSTRACT FROM AUTHOR]

Salient object detection (SOD) from raw sensor images in the edge networks can effectively speed up the decision-making process in the complex environments, because it simulates the mechanism of human attention to identify salient objects from images. The success of supervised deep learning approaches have been widely proved SOD field. However, the imbalanced and limited training data at each edge device pose a huge challenge for us to deploy deep learning methods in the edge computing environments. In this article, we propose a cloud-edge distributed augmented semi-supervised learning architecture for SOD over the edge networks. The framework consists of two components: the base classification networks are employed in different edge nodes, and the reverse augmented network is employed in cloud. First, the base classification networks are trained with data from edge nodes while the reverse augmented network is trained with the whole data. Then, we concatenate each base classification network with reverse augmented network, thus the latter network can help the training of former network. Finally, we integrate the outputs of all base classification network to generate the pseudo-labels, which are used for semi-supervised learning of the augment network. We demonstrated a convincing performance of our semi-supervised learning framework on four bench-marked data-sets. These results show that our augmented semi-supervised learning framework can outperform other optimization strategies on deep learning for the edge computing. [ABSTRACT FROM AUTHOR]

The hardware realization of Universal Filtered Multi Carrier (UFMC) architecture has attracted significant attention in fifth generation (5G) and beyond. In addition to the flexibility in fast Fourier transform (FFT)-length, a flexible prototype filter in combination with multiplicative complex spectrum shifting co-efficients is required for realizing flexible UFMC architecture. The existing architectures of UFMC transmitter commonly adopted fixed-size FFT-length, number of subbands, subband size, and filter-length. Moreover, the lack of flexible prototype filter and spectrum localization of filter co-efficients to individual subbands limits the flexible UFMC system design. In this paper, we propose VLSI architecture for a flexible length prototype filter that can generate spectrally shifted filter co-efficients to individual subbands in tune with the changing value of FFT-length, number of subbands, subband size, and filter-length. For 16-bit word size architecture, our proposed design produces filter co-efficients and spectrum shifting co-efficients upto length, $2^{15}$. Thus, any desired combination of FFT-length, number of subbands, subband size and filter-length is selected to generate the filter co-efficients for the individual subbands. Moreover, complex multiplication and addition operations are reduced in proposed architecture, quantitatively, about 58.81% reduction in filtering unit is achieved over the state-of-the-art architecture. Finally, hardware implementation output and XILINX post route simulation result matches perfectly with MATLAB simulations. [ABSTRACT FROM AUTHOR]

Radio-frequency-energy-powered cognitive radio network (RF-CRN) is being taken seriously in Connected Vehicles, especially in 5G network, which can better address the challenges of energy limitation and spectrum scarcity. However, the energy efficiency (EE) of the RF-CRN wherein multiple secondary users (SUs) share the same channel is rarely presented. In this article, we consider a RF-CRN in which SUs first harvest energy from RF signals originating from a primary network (PN) and then utilize the available energy in the battery to transmit data. Since all SUs can access the authorized spectrum for transmission simultaneously, co-frequency interference (Co-FI) occurs among SUs. Given the quality of service (QoS) requirement, our goal is to achieve the maximum EE of the RF-CRN by jointly optimizing transmission time and power control. To this end, a resource allocation scheme referred to as approximate convex policy for co-frequency interference (CO-ACP) is proposed. Specifically, the EE problem is firstly converted into a convex one by CO-ACP. Then, we utilize Frank-Wolfe (FW) and one-dimensional linear programming to obtain the optimal solution. Simulation results demonstrate that a tight lower-bound optimum solution for the non-convex EE maximization can be achieved by CO-ACP. Moreover, the CO-ACP provides meaningful system features, such as the number of SUs, energy harvesting efficiency, and the battery energy state of the SUs under different RF-CRN scenarios, providing a clear reference for future deployment of RF-CRN. [ABSTRACT FROM AUTHOR]

Intelligent reflecting surface (IRS), which is composed of a large number of low-cost passive elements, has the ability to reflect the incident signal independently with an adjustable phase and amplitude shifts, has been regarded as a key and emerging technology for achieving the cost-effectively spectrum and addressing energy issues in the next generation of wireless networks. In this paper, we propose a general wideband non-stationary channel model for IRS-assisted multiple-input multiple-output (MIMO) communication scenarios, which aims at capturing the underlying propagation characteristics of IRS-assisted communication systems. By properly adjusting the key system parameters, the proposed channel model can be used to describe various IRS-assisted communication scenarios. Furthermore, we separate the channel between the mobile transmitter (MT) and mobile receiver (MR) into the subchannel between the MT and IRS, the subchannel between the IRS and MR, and the subchannel between the MT and MR. The physical properties of each subchannel are investigated accordingly; then, we develop an equivalent channel model to study the key characteristics of the proposed IRS-assisted channel model, such as the time-varying spatial-temporal (ST) cross-correlation functions (CCFs), temporal auto-correlation functions (ACFs), and frequency CCFs. Finally, numerical results demonstrate that the proposed channel model is practical for describing the IRS-assisted MIMO wireless communication scenarios. [ABSTRACT FROM AUTHOR]

A cooperative relaying system based on non-orthogonal multiple access (NOMA) is considered a promising technique in next-generation wireless communications. In this letter, we investigate the performance of a dual-hop decode-and-forward based on the NOMA relaying system in terms of achievable rate and power allocation coefficient over imperfect estimation of κ-μ shadowed fading channels. Moreover, we consider imperfect successive interference cancellation at the relay in the first phase of transmission. Specifically, we derive the closed expressions for the average achievable rate and power allocation coefficient in terms of hypergeometric function. Finally, the numerical results confirm that our derived analytical results fit well the Monte Carlo simulations. [ABSTRACT FROM AUTHOR]

Next-generation wireless networks are witnessing an increasing number of clustering applications, and produce a large amount of non-linear and unlabeled data. In some degree, single kernel methods face the challenging problem of kernel choice. To overcome this problem for non-linear data clustering, multiple kernel graph-based clustering (MKGC) has attracted intense attention in recent years. However, existing MKGC methods suffer from two common problems: (1) they mainly aim to learn a consensus kernel from multiple candidate kernels, slight affinity graph learning, such that cannot fully exploit the underlying graph structure of non-linear data; (2) they disregard the high-order correlations between all base kernels, which cannot fully capture the consistent and complementary information of all kernels. In this paper, we propose a novel non-negative matrix factorization (NMF) tailored graph tensor MKGC method for non-linear data clustering, namely TMKGC. Specifically, TMKGC integrates NMF and graph learning together in kernel space so as to learn multiple candidate affinity graphs. Afterwards, the high-order structure information of all candidate graphs is captured in a 3-order tensor kernel space by introducing tensor singular value decomposition based tensor nuclear norm, such that an optimal affinity graph can be obtained subsequently. Based on the alternating direction method of multipliers, the effective local and distributed solvers are elaborated to solve the proposed objective function. Extensive experiments have demonstrated the superiority of TMKGC compared to the state-of-the-art MKGC methods. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) technology has been widely introduced to the existing medical system. An eHealth system based on IoT devices has gained widespread popularity. In this article, we propose an IoT eHealth framework to provide an autonomous solution for patients with interventional cardiovascular diseases. In this framework, wearable sensors are used to collect a patient's health data, which is daily monitored by a remote doctor. When the monitoring data is abnormal, the remote doctor will ask for image acquisition of the patient's cardiovascular internal conditions. We leverage edge computing to classify these training images by the local base classifier; thereafter, pseudo-labels are generated according to its output. Moreover, a deep segmentation network is leveraged for the segmentation of stent structs in intravascular optical coherence tomography and intravenous ultrasound images of patients. The experimental results demonstrate that remote and local doctors perform real-time visual communication to complete telesurgery. In the experiments, we adopt the U-net backbone with a pretrained SeResNet34 as the encoder to segment the stent structs. Meanwhile, a series of comparative experiments have been conducted to demonstrate the effectiveness of our method based on accuracy, sensitivity, Jaccard, and dice. [ABSTRACT FROM AUTHOR]

Multiple pedestrian tracking (MPT) has gained significant attention due to its huge potential in a commercial application. It aims to predict multiple pedestrian trajectories and maintain their identities, given a video sequence. In the past decade, due to the advancement in pedestrian detection algorithms, Tracking-by-Detection (TBD) based algorithms have achieved tremendous successes. TBD has become the most popular MPT framework, and it has been actively studied in the past decade. In this paper, we give a comprehensive survey of recent advances in TBD-based MPT algorithms. We systematically analyze the existing TBD-based algorithms and organize the survey into four major parts. At first, this survey draws a timeline to introduce the milestones of TBD-based works which briefly reviews the development of the existing TBD-based methods. Second, the main procedures of the TBD framework are summarized, and each stage in the procedure is described in detail. Afterward, this survey analyzes the performance of existing TBD-based algorithms on MOT challenge datasets and discusses the factors that affect tracking performance. Finally, open issues and future directions in the TBD framework are discussed. [ABSTRACT FROM AUTHOR]

In the cognitive computing of intelligent industrial Internet of Things, clustering is a fundamental machine learning problem to exploit the latent data relationships. To overcome the challenge of kernel choice for nonlinear clustering tasks, multiple kernel clustering (MKC) has attracted intensive attention. However, existing graph-based MKC methods mainly aim to learn a consensus kernel as well as an affinity graph from multiple candidate kernels, which cannot fully exploit the latent graph information. In this article, we propose a novel pure graph-based MKC method. Specifically, a new graph model is proposed to preserve the local manifold structure of the data in kernel space so as to learn multiple candidate graphs. Afterward, the latent consistency and selfishness of these candidate graphs are fully considered. Furthermore, a graph connectivity constraint is introduced to avoid requiring any postprocessing clustering step. Comprehensive experimental results demonstrate the superiority of our method. [ABSTRACT FROM AUTHOR]

Classical automated schemes in the industrial Internet of Things (IIoT) are challenged by the problems related to huge record storage and the way they respond. To properly manage the manufacturing settings, cognitive systems aim to find a way to efficiently adapt their actions based on uncertainty management and sensory data. However, due to the lack of existing IT integration, cognitive systems are not fully exploited by organizations. In this article, we provide a novel decision-making process in industrial informatics during information transmission, manufacturing, and storing records through the simple additive weighting and analytic hierarchy process. The proposed mechanism is analyzed and validated rigorously using various sensing and decision-making parameters against a baseline solution in industrial parameter settings. The simulation results suggest that the proposed mechanism leads to 87% efficiency in terms of better detection of the sensor node, decision-making, and alteration of transmitted data during analyses of product manufacturing in the IIoT. [ABSTRACT FROM AUTHOR]

The Industrial Internet of Things (IoT) typically consists of several thousands of heterogeneous devices, such as sensors, actuators, access points, machinery, end-users' handheld equipment, and supply chain. In such an industrial environment, a multitude of data is generated from massive IoT devices, e.g., sensors for monitoring the environment, reading temperature, and gauging pressure. Most of the data are from delay-sensitive and computation-intensive applications, such as real-time manufacturing and automated diagnostics, which require big data analytics with low latency. Machine learning (ML) has been witnessed as an efficient solution for big data analytics. The majority of such ML algorithms are centralized methods, meaning that they first gather data from different users for use as a training dataset, which is placed on the ML server, and then build a model to classify the new data samples by applying the ML algorithms to this training dataset. However, the access to these datasets in the centralized ML methods raises concerns about data privacy for users. Federated learning (FL) was designed to protect data privacy to address a part of these issues. In FL, each participant uses a global training model without uploading their private data to a third-party server. Compared with the conventional ML, FL can preserve data security, especially in terms of participant data during the learning process. In particular, FL can also help in updating server-side data for the global model, and the participant is not required to provide their data. However, in FL, individual computing units may show abnormal actions, such as faulty software, hardware invasions, unreliable communication channels, and malicious samples deliberately crafting the model. To mitigate these challenges, we require robust policies to control the learning phases in FL. Motivated by the abovementioned issues, this special section solicits original research and practical contributions that advance the security and privacy of the FL solutions for industrial IoT applications as follows. [ABSTRACT FROM AUTHOR]

Universal-filtered multicarrier (UFMC) is one of the potential candidates for 5G multicarrier waveforms due to its several attractive features such as suppressed out-of-band radiation to the nearby sub-band. However, the hardware realization of UFMC systems is limited by a large number of arithmetic units for inverse fast Fourier transform (IFFT) and pulse shaping filters. In this letter, we propose an architecture that presents a refreshing approach toward designing a low-complexity architecture for the baseband UFMC transmitter with Dolph–Chebyshev filter. Compared to the read-only-memory (ROM)-based state-of-the-art, the proposed architecture requires less number of ROM locations and has the flexibility to externally select the inverse discrete Fourier transform (IDFT)-size, number of sub-bands, and number of subcarriers in a sub-band. Moreover, we implement the proposed architecture on a commercially available Virtex-5 field-programmable gate array (FPGA) device for testing and analyzing the baseband UFMC signal. Finally, the XILINX post-route results are found comparable with MATLAB simulations. [ABSTRACT FROM AUTHOR]

Edge computing has already shown its potential benefits to support the delay-sensitive and computation-intensive service provisioning in the Internet-of-Things-based environments. Simultaneously, artificial intelligence is expected to enhance the cognizance and intelligence of edge computing, resulting in a new paradigm, intelligent edge computing. However, as is evident, several new security and privacy-related issues will arise due to the distinct characteristics and working principles of intelligent edge computing. In this article, we focus on malicious attacks targeting the intelligent engines of edge computing. These attacks counteract the gains from intelligent systems and also compromise the edge computing system. We provide the main factors that are responsible for the security and privacy-related challenges in intelligent edge computing. Finally, the potential research directions in privacy and security-related issues are outlined for intelligent edge computing. [ABSTRACT FROM AUTHOR]

With the recent advancement in wireless communication and networks, we are at the doorstep of the Tactile Internet. The Tactile Internet aims to enable the skill delivery and thereafter democratize the specialized skills for many emerging applications (e.g., remote medical, industrial machinery, remote robotics, autonomous driving). In this article, we start with the motivation of applying intelligent edge computing for computation offloading in the Tactile Internet. Afterward, we outline the main research challenges to leverage edge intelligence at the master, network, and controlled domain of the Tactile Internet. The key research challenges in the Tactile Internet lie in its stringent requirements such as ultra-low latency, ultra-high reliability, and almost zero service outage. We also discuss major entities in intelligent edge computing and their role in the Tactile Internet. Finally, several potential research challenges in edge intelligence for the Tactile Internet are highlighted. [ABSTRACT FROM AUTHOR]

Fog computing leverages the computational resources at the network edge to meet the increasing demand for latency-sensitive applications in large-scale industries. In this article, we study the computation offloading in a fog computing network, where the end users, most of the time, offload part of their tasks to a fog node. Nevertheless, limited by the computational and storage resources, the fog node further simultaneously offloads the task data to the neighboring fog nodes and/or the remote cloud server to obtain the additional computing resources. However, meanwhile, the offloaded tasks from the neighboring node incur burden to the fog node. Moreover, the task offloading to the remote cloud server can suffer from limited communication resources. Thus, to jointly optimize the amount of tasks offloaded to the neighboring fog nodes and communication resource allocation for the offloaded tasks to the remote cloud, we formulate a latency-driven task data offloading problem considering the transmission delay from fog to the cloud and service rate that includes the local processing time and waiting time at each fog node. The optimization problem is formulated as a quadratically constraint quadratic programming. We solve the problem by semidefinite relaxation. The simulation results demonstrate that the proposed strategy is effective and scalable under various simulation settings. [ABSTRACT FROM AUTHOR]

In this letter, we study the computational offloading scheme for the delay-aware tasks of the end-users in the fog computing network. We consider a fog federation of different service providers where an individual fog node allocates its computing resources to the end-user in its proximity, while a fog manager coordinates the load balancing among the fog nodes over the entire network. At first, an individual fog node aims to maximize its revenue by selling the computational resources to the end-user in a distributed manner without any global knowledge of the network. To further maximize the overall revenue considering all fog nodes in the fog federation, the fog manager utilizes the remaining computing resources of the underloaded fog nodes. The extensive simulation results show the revenue improvement leveraging fog federation over entire network while maintaining the same and even better delay-performance for the end-users. [ABSTRACT FROM AUTHOR]

A universal-filtered multicarrier (UFMC) system that is a generalization of filtered orthogonal frequency-division multiplexing (OFDM) and filter-bank-based multicarrier is being considered as a potential candidate for fifth-generation due to its robustness against intercarrier interference as in cyclic-prefix-based OFDM systems. However, real-time hardware realization of multicarrier systems is limited by a large number of arithmetic units for inverse fast Fourier transform and pulse-shaping filters. In this paper, we aim to propose a low-complexity and reconfigurable architecture for a baseband UFMC transmitter. To the best of our knowledge, the proposed architecture is the first reconfigurable architecture that has the flexibility to choose the number of subcarriers in a subband without any change in hardware resources. In addition, the proposed architecture selects the filter from a group of filters with a single selection line. Moreover, we use a commercially available field-programmable gate array device for real-time testing and analyzing the baseband UFMC signal. From the extensive experiments, we study the occupied bandwidth, main-lobe power, and sidelobe power of the baseband signal with different filters in real-time scenarios. Finally, we measure the quantization error in baseband signal generation for the proposed UFMC transmitter architecture and find comparable with the error bound. [ABSTRACT FROM AUTHOR]