Computing at the edge is increasingly important since a massive amount of data is generated. This poses challenges in transporting all that data to the remote data centers and cloud, where they can be processed and analyzed. On the other hand, harnessing the edge data is essential for offering data-driven and machine learning-based applications, if the challenges, such as device capabilities, connectivity, and heterogeneity can be mitigated. Machine learning applications are very compute-intensive and require processing of large amount of data. However, edge devices are often resources-constrained, in terms of compute resources, power, storage, and network connectivity. Hence, limiting their potential to run efficiently and accurately state-of-the art deep neural network (DNN) models, which are becoming larger and more complex. This paper proposes a novel offloading mechanism by leveraging installed-base on-premises (edge) computational resources. The proposed mechanism allows the edge devices to offload heavy and compute-intensive workloads to edge nodes instead of using remote cloud. Our offloading mechanism has been prototyped and tested with state-of-the art person and object detection DNN models for mobile robots and video surveillance applications. The performance shows a significant gain compared to cloud-based offloading strategies in terms of accuracy and latency.
This paper presents Systolic-CNN, an OpenCL-defined scalable, run-time-flexible FPGA accelerator architecture, optimized for accelerating the inference of various convolutional neural networks (CNNs) in multi-tenancy cloud/edge computing. The existing OpenCL-defined FPGA accelerators for CNN inference are insufficient due to limited flexibility for supporting multiple CNN models at run time and poor scalability resulting in underutilized FPGA resources and limited computational parallelism. Systolic-CNN adopts a highly pipelined and paralleled 1-D systolic array architecture, which efficiently explores both spatial and temporal parallelism for accelerating CNN inference on FPGAs. Systolic-CNN is highly scalable and parameterized, which can be easily adapted by users to achieve up to 100% utilization of the coarse-grained computation resources (i.e., DSP blocks) for a given FPGA. Systolic-CNN is also run-time-flexible in the context of multi-tenancy cloud/edge computing, which can be time-shared to accelerate a variety of CNN models at run time without the need of recompiling the FPGA kernel hardware nor reprogramming the FPGA. The experiment results based on an Intel Arria/Stratix 10 GX FPGA Development board show that the optimized single-precision implementation of Systolic-CNN can achieve an average inference latency of 7ms/2ms, 84ms/33ms, 202ms/73ms, 1615ms/873ms, and 900ms/498ms per image for accelerating AlexNet, ResNet-50, ResNet-152, RetinaNet, and Light-weight RetinaNet, respectively. Codes are available at https://github.com/PSCLab-ASU/Systolic-CNN.
Mobile Augmented Reality (MAR) applications employ computationally demanding vision algorithms on resourcelimited devices. In parallel, communication networks are becoming more ubiquitous. Offloading to distant servers can thus overcome the device limitations at the cost of network delays. Multipath networking has been proposed to overcome network limitations but it is not easily adaptable to edge computing due to the server proximity and networking differences. In this article, we extend the current mobile edge offloading models and present a model for multi-server device-to-device, edge, and cloud offloading. We then introduce a new task allocation algorithm exploiting this model for MAR offloading. Finally, we evaluate the allocation algorithm against naive multipath scheduling and single path models through both a real-life experiment and extensive simulations. In case of sub-optimal network conditions, our model allows reducing the latency compared to single-path offloading, and significantly decreases packet loss compared to random task allocation. We also display the impact of the variation of WiFi parameters on task completion. We finally demonstrate the robustness of our system in case of network instability. With only 70% WiFi availability, our system keeps the excess latency below 9 ms. We finally evaluate the capabilities of the upcoming 5G and 802.11ax
With the rise in edge-computing devices, there has been an increasing demand to deploy energy and resource-efficient models. A large body of research has been devoted to developing methods that can reduce the size of the model considerably without affecting the standard metrics such as top-1 accuracy. However, these pruning approaches tend to result in a significant mismatch in other metrics such as fairness across classes and explainability. To combat such misalignment, we propose a novel multi-part loss function inspired by the knowledge-distillation literature. Through extensive experiments, we demonstrate the effectiveness of our approach across different compression algorithms, architectures, tasks as well as datasets. In particular, we obtain up to $4.1\times$ reduction in the number of prediction mismatches between the compressed and reference models, and up to $5.7\times$ in cases where the reference model makes the correct prediction; all while making no changes to the compression algorithm, and minor modifications to the loss function. Furthermore, we demonstrate how inducing simple alignment between the predictions of the models naturally improves the alignment on other metrics including fairness and attributions. Our framework can thus serve as a simple plug-and-play component for compression algorithms in the future.
The sixth-generation (6G) wireless communication network is expected to integrate the terrestrial, aerial, and maritime communications into a robust network which would be more reliable, fast, and can support a massive number of devices with ultra-low latency requirements. The researchers around the globe are proposing cutting edge technologies such as artificial intelligence (AI)/machine learning (ML), quantum communication/quantum machine learning (QML), blockchain, tera-Hertz and millimeter waves communication, tactile Internet, non-orthogonal multiple access (NOMA), small cells communication, fog/edge computing, etc., as the key technologies in the realization of beyond 5G (B5G) and 6G communications. In this article, we provide a detailed overview of the 6G network dimensions with air interface and associated potential technologies. More specifically, we highlight the use cases and applications of the proposed 6G networks in various dimensions. Furthermore, we also discuss the key performance indicators (KPI) for the B5G/6G network, challenges, and future research opportunities in this domain.
Time-sensitive services (TSSs) have been widely envisioned for future sixth generation (6G) wireless communication networks. Due to its inherent low-latency advantage, mobile edge computing (MEC) will be an indispensable enabler for TSSs. The random characteristics of the delay experienced by users are key metrics reflecting the quality of service (QoS) of TSSs. Most existing studies on MEC have focused on the average delay. Only a few research efforts have been devoted to other random delay characteristics, such as the delay bound violation probability and the probability distribution of the delay, by decoupling the transmission and computation processes of MEC. However, if these two processes could not be decoupled, the coupling will bring new challenges to analyzing the random delay characteristics. In this paper, an MEC system with a limited computation buffer at the edge server is considered. In this system, the transmission process and computation process form a feedback loop and could not be decoupled. We formulate a discrete-time two-stage tandem queueing system. Then, by using the matrix-geometric method, we obtain the estimation methods for the random delay characteristics, including the probability distribution of the delay, the delay bound violation probability, the average delay and the delay standard deviation. The estimation methods are verified by simulations. The random delay characteristics are analyzed by numerical experiments, which unveil the coupling relationship between the transmission process and computation process for MEC. These results will largely facilitate elaborate allocation of communication and computation resources to improve the QoS of TSSs.
Multi-access edge computing (MEC) can enhance the computing capability of mobile devices, while non-orthogonal multiple access (NOMA) can provide high data rates. Combining these two strategies can effectively benefit the network with spectrum and energy efficiency. In this paper, we investigate the task delay minimization in multi-user NOMA-MEC networks, where multiple users can offload their tasks simultaneously through the same frequency band. We adopt the partial offloading policy, in which each user can partition its computation task into offloading and locally computing parts. We aim to minimize the task delay among users by optimizing their tasks partition ratios and offloading transmit power. The delay minimization problem is first formulated, and it is shown that it is a nonconvex one. By carefully investigating its structure, we transform the original problem into an equivalent quasi-convex. In this way, a bisection search iterative algorithm is proposed in order to achieve the minimum task delay. To reduce the complexity of the proposed algorithm and evaluate its optimality, we further derive closed-form expressions for the optimal task partition ratio and offloading power for the case of two-user NOMA-MEC networks. Simulations demonstrate the convergence and optimality of the proposed algorithm and the effectiveness of the closed-form analysis.
The combination of non-orthogonal multiple access (NOMA) and mobile edge computing (MEC) can significantly improve the spectrum efficiency beyond the fifth-generation network. In this paper, we mainly focus on energy-efficient resource allocation for a multi-user, multi-BS NOMA assisted MEC network with imperfect channel state information (CSI), in which each user can upload its tasks to multiple base stations (BSs) for remote executions. To minimize the energy consumption, we consider jointly optimizing the task assignment, power allocation and user association. As the main contribution, with imperfect CSI, the optimal closed-form expressions of task assignment and power allocation are analytically derived for the two-BS case. Specifically, the original formulated problem is nonconvex. We first transform the probabilistic problem into a non-probabilistic one. Subsequently, a bilevel programming method is proposed to derive the optimal solution. In addition, by incorporating the matching algorithm with the optimal task and power allocation, we propose a low complexity algorithm to efficiently optimize user association for the multi-user and multi-BS case. Simulations demonstrate that the proposed algorithm can yield much better performance than the conventional OMA scheme but also the identical results with lower complexity from the exhaustive search with the small number of BSs.
Wireless charging coupled with computation offloading in edge networks offers a promising solution for realizing power-hungry and computation intensive applications on user devices. We consider a mutil-access edge computing (MEC) system with collocated MEC servers and base-stations/access points (BS/AP) supporting multiple users requesting data computation and wireless charging. We propose an integrated solution with computation offloading to satisfy the largest proportion of requested wireless charging while keeping the energy consumption at the minimum subject to the MEC-AP transmit power and latency constraints. We propose a novel algorithm to perform data partitioning, time allocation, transmit power control and design the optimal energy beamforming for wireless charging. Our resource allocation scheme offers an energy minimizing solution compared to other schemes while also delivering higher amount of transferred charge to the users.
Wireless sensor network (WSN) based SHM systems have shown significant improvement as compared to traditional wired-SHM systems in terms of cost, accuracy, and reliability of the monitoring. However, due to the resource-constrained nature of the sensor nodes, it is a challenge to process a large amount of sensed vibration data in real-time. Existing mechanisms of data processing are centralized and use cloud or remote servers to analyze the data to characterize the state of the bridge, i.e., healthy or damaged. These methods are feasible for wired-SHM systems, however, transmitting huge data-sets in WSNs has been found to be arduous. In this paper, we propose a mechanism named as ``in-network damage detection on edge (INDDE)" which extracts the statistical features from raw acceleration measurements corresponding to the healthy condition of the bridge and use them to train a probabilistic model, i.e., estimating the probability density function (PDF) of multivariate Gaussian distribution. The trained model helps to identify the anomalous behaviour of the new data points collected from the unknown condition of the bridge in real-time. Each edge device classifies the condition of the bridge as either "healthy" or "damaged" around its deployment region depending on their respective trained model. Experimentation results showcase a promising 96-100% damage detection accuracy with the advantage of no data transmission from sensor nodes to the cloud for processing.
With the development of intelligent vehicles, security and reliability communication between vehicles has become a key problem to be solved in Internet of vehicles(IoVs). Blockchain is considered as a feasible solution due to its advantages of decentralization, unforgeability and collective maintenance. However, the computing power of nodes in IoVs is limited, while the consensus mechanism of blockchain requires that the miners in the system have strong computing power for mining calculation. It consequently cannot satisfy the requirements, which is the challenges for the application of blockchain in IoVs. In fact, the application of blockchain in IoVs can be implemented by employing edge computing. The key entity of edge computing is the edge servers(ESs). Roadside nodes(RSUs) can be deployed as ESs of edge computing in IoVs. We have studied the ES deployment scheme for covering more vehicle nodes in IoVs, and propose a randomized algorithm to calculate approximation solutions. Finally, we simulated the performance of the proposed scheme and compared it with other deployment schemes.
Driven by explosive computation demands of Internet of Things (IoT), mobile edge computing (MEC) provides a promising technique to enhance the computation capability for mobile users. In this paper, we propose a joint resource allocation and load management mechanism in an MEC system with wireless power transfer (WPT), by jointly optimizing the transmit power for WPT, the local/edge computing load, the offloading time, and the frequencies of the central processing units (CPUs) at the access point (AP) and the users. To achieve an energy-efficient and sustainable WPT-MEC system, we minimize the total energy consumption of the AP, while meeting computation latency requirements. Cooling energy which is non-negligible, is taken into account in minimizing the energy consumption of the MEC system. By rigorously orchestrating the state-of-the-art optimization techniques, we design an iterative algorithm and obtain the optimal solution in a semi-closed form. Based on the solution, interesting properties and insights are summarized. Extensive numerical tests show that the proposed algorithm can save up to 90.4% the energy of existing benchmarks.
With the development of smart vehicles, computing-intensive tasks are widely and rapidly generated. To alleviate the burden of on-board CPU, connected vehicles can offload tasks to or make request from nearby edge server thanks to the emerging Mobile Edge Computing (MEC). However, such approach may sharply increase the workload of an edge server, and cause network congestion, especially in rural and mountain areas where there are few edge servers. To this end, a UAV-assisted MEC system is proposed in this paper, and joint optimization algorithm of the deployment and resource allocation of UAVs (JOAoDR) is proposed to decide the location and balance the resource and rewards of the UAVs. We solve a long-term profit maximization problem in terms of the operator. Numerical results demonstrated that our algorithm outperforms other benchmarks algorithm, and validated our solution.
In this paper, a real-time signal processing frame-work based on a 60 GHz frequency-modulated continuous wave (FMCW) radar system to recognize gestures is proposed. In order to improve the robustness of the radar-based gesture recognition system, the proposed framework extracts a comprehensive hand profile, including range, Doppler, azimuth and elevation, over multiple measurement-cycles and encodes them into a feature cube. Rather than feeding the range-Doppler spectrum sequence into a deep convolutional neural network (CNN) connected with recurrent neural networks, the proposed framework takes the aforementioned feature cube as input of a shallow CNN for gesture recognition to reduce the computational complexity. In addition, we develop a hand activity detection (HAD) algorithm to automatize the detection of gestures in real-time case. The proposed HAD can capture the time-stamp at which a gesture finishes and feeds the hand profile of all the relevant measurement-cycles before this time-stamp into the CNN with low latency. Since the proposed framework is able to detect and classify gestures at limited computational cost, it could be deployed in an edge-computing platform for real-time applications, whose performance is notedly inferior to a state-of-the-art personal computer. The experimental results show that the proposed framework has the capability of classifying 12 gestures in real-time with a high F1-score.
Improving efficiency of healthcare systems is a top national interest worldwide. However, the need of delivering scalable healthcare services to the patients while reducing costs is a challenging issue. Among the most promising approaches for enabling smart healthcare (s-health) are edge-computing capabilities and next-generation wireless networking technologies that can provide real-time and cost-effective patient remote monitoring. In this paper, we present our vision of exploiting multi-access edge computing (MEC) for s-health applications. We envision a MEC-based architecture and discuss the benefits that it can bring to realize in-network and context-aware processing so that the s-health requirements are met. We then present two main functionalities that can be implemented leveraging such an architecture to provide efficient data delivery, namely, multimodal data compression and edge-based feature extraction for event detection. The former allows efficient and low distortion compression, while the latter ensures high-reliability and fast response in case of emergency applications. Finally, we discuss the main challenges and opportunities that edge computing could provide and possible directions for future research.
Wireless powered mobile edge computing (WP-MEC) has been recognized as a promising technique to provide both enhanced computational capability and sustainable energy supply to massive low-power wireless devices. However, its energy consumption becomes substantial, when the transmission link used for wireless energy transfer (WET) and for computation offloading is hostile. To mitigate this hindrance, we propose to employ the emerging technique of intelligent reflecting surface (IRS) in WP-MEC systems, which is capable of providing an additional link both for WET and for computation offloading. Specifically, we consider a multi-user scenario where both the WET and the computation offloading are based on orthogonal frequency-division multiplexing (OFDM) systems. Built on this model, an innovative framework is developed to minimize the energy consumption of the IRS-aided WP-MEC network, by optimizing the power allocation of the WET signals, the local computing frequencies of wireless devices, both the sub-band-device association and the power allocation used for computation offloading, as well as the IRS reflection coefficients. The major challenges of this optimization lie in the strong coupling between the settings of WET and of computing as well as the unit-modules constraint on IRS reflection coefficients. To tackle these issues, the technique of alternative optimization is invoked for decoupling the WET and computing designs, while two sets of locally optimal IRS reflection coefficients are provided for WET and for computation offloading separately relying on the successive convex approximation method. The numerical results demonstrate that our proposed scheme is capable of monumentally outperforming the conventional WP-MEC network without IRSs.
This work presents InfiniWolf, a novel multi-sensor smartwatch that can achieve self-sustainability exploiting thermal and solar energy harvesting, performing computationally high demanding tasks. The smartwatch embeds both a System-on-Chip (SoC) with an ARM Cortex-M processor and Bluetooth Low Energy (BLE) and Mr. Wolf, an open-hardware RISC-V based parallel ultra-low-power processor that boosts the processing capabilities on board by more than one order of magnitude, while also increasing energy efficiency. We demonstrate its functionality based on a sample application scenario performing stress detection with multi-layer artificial neural networks on a wearable multi-sensor bracelet. Experimental results show the benefits in terms of energy efficiency and latency of Mr. Wolf over an ARM Cortex-M4F micro-controllers and the possibility, under specific assumptions, to be self-sustainable using thermal and solar energy harvesting while performing up to 24 stress classifications per minute in indoor conditions.
By taking full advantage of Computing, Communication and Caching (3C) resources at the network edge, Mobile Edge Computing (MEC) is envisioned as one of the key enablers for the next generation networks. However, current fixed-location MEC architecture may not be able to make real-time decision in dynamic environment, especially in large-scale scenarios. To address this issue, in this paper, a Heterogeneous MEC (H-MEC) architecture is proposed, which is composed of fixed unit, i.e., Ground Stations (GSs) as well as moving nodes, i.e., Ground Vehicles (GVs) and Unmanned Aerial Vehicles (UAVs), all with 3C resource enabled. The key challenges in H-MEC, i.e., mobile edge node management, real-time decision making, user association and resource allocation along with the possible Artificial Intelligence (AI)-based solutions are discussed. In addition, the AI-based joint Resource schEduling (ARE) framework with two different AI-based mechanisms, i.e., Deep neural network (DNN)-based and deep reinforcement learning (DRL)-based architectures are proposed. DNN-based solution with online incremental learning applies the global optimizer and therefore has better performance than the DRL-based architecture with online policy updating, but requires longer training time. The simulation results are given to verify the efficiency of our proposed ARE framework.
We study a wireless edge-computing system which allows multiple users to simultaneously offload computation-intensive tasks to multiple massive-MIMO access points, each with a collocated multi-access edge computing (MEC) server. Massive-MIMO enables simultaneous uplink transmissions from all users, significantly shortening the data offloading time compared to sequential protocols, and makes the three phases of data offloading, computing, and downloading have comparable durations. Based on this three-phase structure, we formulate a novel problem to minimize a weighted sum of the energy consumption at both the users and the MEC server under a round-trip latency constraint, using a combination of data partitioning, transmit power control and CPU frequency scaling at both the user and server ends. We design a novel nested primal-dual algorithm using two different methods to solve this problem efficiently. Optimized solutions show that for larger requests, more data is offloaded to the MECs to reduce local computation time in order to meet the latency constraint, despite higher energy cost of wireless transmissions. Massive-MIMO channel estimation errors under pilot contamination also causes more data to be offloaded to the MECs. Compared to binary offloading, partial offloading with data partitioning is superior and leads to significant reduction in the overall energy consumption.
This paper presents a Gated Recurrent Unit (GRU) based recurrent neural network (RNN) accelerator called EdgeDRNN designed for portable edge computing. EdgeDRNN adopts the spiking neural network inspired delta network algorithm to exploit temporal sparsity in RNNs. It reduces off-chip memory access by a factor of up to 10x with tolerable accuracy loss. Experimental results on a 10 million parameter 2-layer GRU-RNN, with weights stored in DRAM, show that EdgeDRNN computes them in under 0.5 ms. With 2.42 W wall plug power on an entry level USB powered FPGA board, it achieves latency comparable with a 92 W Nvidia 1080 GPU. It outperforms NVIDIA Jetson Nano, Jetson TX2 and Intel Neural Compute Stick 2 in latency by 6X. For a batch size of 1, EdgeDRNN achieves a mean effective throughput of 20.2 GOp/s and a wall plug power efficiency that is over 4X higher than all other platforms.