Joint Optimization of Task Offloading and Resource Allocation for Edge Video Analytics

The rapid growth of Industrial Internet of Things (IIoT) terminals has resulted in tasks exhibiting increased concurrency, heterogeneous resource demands, and dynamic priorities, significantly increasing the complexity of task scheduling in edge computing. Cloud-edge-end collaborative computing leverages cross-layer task offloading to alleviate edge node resource contention and improve task scheduling efficiency. However, existing methods generally neglect the joint optimization of task offloading, resource allocation, and priority adaptation, making it difficult to balance task execution and resource utilization under resource-constrained and competitive conditions. To address this, this paper proposes a two-stage dynamic-priority-aware joint task offloading and resource allocation method (DPTORA). In the first stage, an improved Multi-Agent Proximal Policy Optimization (MAPPO) algorithm integrated with a Priority-Gated Attention Module (PGAM) enhances the robustness and accuracy of offloading strategies under dynamic priorities; in the second stage, the resource allocation problem is formulated as a single-objective convex optimization task and solved globally using the Lagrangian dual method. Simulation results show that DPTORA significantly outperforms existing multi-agent reinforcement learning baselines in terms of task latency, energy consumption, and the task completion rate.

Vehicle Edge Computing (VEC) leverages compact cloud computing at the mobile network edge to meet the processing and latency needs of vehicles. By bringing computation closer to the vehicles, VEC reduces data transmission, minimizes latency, and boosts performance for compute-intensive applications. However, during peak hours of urban road traffic, the scarce computational resources available at edge servers could pose challenges in fulfilling the processing needs of vehicles. Introducing Unmanned Aerial Vehicles (UAVs) as supplementary edge computing nodes could significantly mitigate the aforementioned issue. In this paper, we propose a flexible edge computing framework in which a fleet of UAVs function as mobile computational service providers, offering computation offloading services to multiple vehicles. We design and optimize a computation offloading model for the UAV-enabled vehicle edge computing environment. The proposed model tackles the task offloading challenge, aiming to optimize UAV revenue and task processing efficiency while considering the constraints of UAVs’ restricted computational power and energy resources. Towards this end, our model jointly considers two key factors: task partitioning and computational resource allocation. To tackle the challenges posed by the aforementioned non-convex optimization problem, we construct a Markov Decision Process (MDP) model for the multi-UAV-enabled mobile edge computing system and introduce an innovative Multi-Agent Deep Reinforcement Learning (MADRL) framework addressing the decision-making challenge represented by MDP model. Comprehensive simulation outcomes illustrate that our devised task offloading technique outperforms other optimization methods.

With the advent of the 5G era, the demands for features such as low latency and high concurrency are becoming increasingly significant. These sophisticated new network applications and services require huge gaps in network transmission bandwidth, network transmission latency, and user experience, making cloud computing face many technical challenges in terms of applicability. In response to cloud computing's shortcomings, edge computing has come into its own. However, many factors affect task offloading and resource allocation in the edge computing environment, such as the task offload latency, energy consumption, smart device mobility, end-user power, and other issues. This paper proposes a dynamic multi-winner game model based on incomplete information to solve multi-end users' task offloading and edge resource allocation. First, based on the history of end-users storage in edge data centers, a hidden Markov model can predict other end-users' bid prices at time t. Based on these predicted auction prices, the model determines their bids. A dynamic multi-winner game model is used to solve the offload strategy that minimizes latency, energy consumption, cost, and to maximizes end-user satisfaction at the edge data center. Finally, the authors designed a resource allocation algorithm based on different priorities and task types to implement resource allocation in edge data centers. To ensure the prediction model's accuracy, the authors also use the expectation-maximization algorithm to learn the model parameters. Comparative experimental results show that the proposed model can better results in time delay, energy consumption, and cost.

While service-oriented low Earth orbit (LEO) satellite edge computing (LSEC) frameworks enable diverse edge services for user tasks, the heterogeneous distribution of terrestrial users causes substantial imbalance in computational task loads across satellites. This asymmetric workload leads to inefficient resource utilization and degraded edge computing performance. To address these challenges, we propose a joint optimization framework that integrates edge service deployment and task offloading, supported by service popularity analysis and spatiotemporal user-task modeling. The framework employs a two-timescale design: at the large timescale, an improved atomic orbital search (iAOS) heuristic dynamically optimizes service placement, configuration, and resource allocation; at the small timescale, a direction-selective multi-agent double deep Q-network (DS-MDDQN) leverages deep reinforcement learning to route tasks to the most suitable processing nodes. Extensive simulations show that our approach significantly outperforms six representative baselines in both user-perceived performance and system-level efficiency. Replacement studies further verify the effectiveness of each component: iAOS enhances resource utilization and reduces task failure through optimized service deployment, while DS-MDDQN mitigates network dynamics and lowers task completion latency via adaptive task offloading.

Digital Twin (DT) and mobile edge computing are two promising solutions for achieving latency-sensitive and computing-intensive applications in Industrial Internet of Things (IIoT). However, existing task offloading schemes with DT empowerment are faced with challenges, such as the spatio-temporal heterogeneity of edge server (ES) resources, resource-constrained ESs, and the explosive growth of data in emerging applications. This paper investigates the issues of task offloading and resource allocation under the assistance of DT and multiple ESs parallel collaboration. One novel scheme, abbreviated as Mes-PCORA, is proposed. With comprehensive information within the digital space, the Mes-PCORA scheme dynamically adjusts task allocation ratios across multiple ESs to achieve collaborative task offloading. The offloading request is formulated as a non-convex problem. To make the non-convex problem solvable in polynomial time, the original problem transformed into a bilevel optimization problem. Then, a bilevel iterative optimization approach is proposed. Specifically, the upper-level optimization problem is formulated as a multi-agent Markov Decision Process, and a deep reinforcement learning-based resource allocation algorithm is designed to solve it. Subsequently, for the lower-level optimization problem, it is solved by the interior point method. Numerical results demonstrate that the proposed scheme reduces the average task completion latency by 27.16%–63.44% and decreases the task offloading failure rate by 35.83%–73.95%, compared to state-of-the-art baselines.

The purposes are to improve the server deployment capability under Mobile Edge Computing (MEC), reduce the time delay and energy consumption of terminals during task execution, and improve user service quality. After the server deployment problems under traditional edge computing are analyzed and researched, a task resource allocation model based on multi-stage is proposed to solve the communication problem between different supporting devices. This model establishes a combined task resource allocation and task offloading method and optimizes server execution by utilizing the time delay and energy consumption required for task execution and comprehensively considering the restriction processes of task offloading, partition, and transmission. For the MEC process that supports dense networks, a multi-hybrid intelligent algorithm based on energy consumption optimization is proposed. The algorithm converts the original problem into a power allocation problem via a heuristic model. Simultaneously, it determines the appropriate allocation strategy through distributed planning, duality, and upper bound replacement. Results demonstrate that the proposed multi-stage combination-based service deployment optimization model can solve the problem of minimizing the maximum task execution energy consumption combined with task offloading and resource allocation effectively. The algorithm has good performance in handling user fairness and the worst-case task execution energy consumption. The proposed hybrid intelligent algorithm can partition tasks into task offloading sub-problems and resource allocation sub-problems, meeting the user’s task execution needs. A comparison with the latest algorithm also verifies the model’s performance and effectiveness. The above results can provide a theoretical basis and some practical ideas for server deployment and applications under MEC.

Taking the mobile edge computing paradigm as an effective supplement to the vehicular networks can enable vehicles to obtain network resources and computing capability nearby, and meet the current large-scale increase in vehicular service requirements. However, the congestion of wireless networks and insufficient computing resources of edge servers caused by the strong mobility of vehicles and the offloading of a large number of tasks make it difficult to provide users with good quality of service. In existing work, the influence of network access point selection on task execution latency was often not considered. In this paper, a pre-allocation algorithm for vehicle tasks is proposed to solve the problem of service interruption caused by vehicle movement and the limited edge coverage. Then, a system model is utilized to comprehensively consider the vehicle movement characteristics, access point resource utilization, and edge server workloads, so as to characterize the overall latency of vehicle task offloading execution. Furthermore, an adaptive task offloading strategy for automatic and efficient network selection, task offloading decisions in vehicular edge computing is implemented. Experimental results show that the proposed method significantly improves the overall task execution performance and reduces the time overhead of task offloading.

Mobile edge computing (MEC) is an emerging computing and communication model that brings extremely resourceful servers at a network's edge. This reduces the latency associated with cloud computing as applications executing on mobile devices do not always need to communicate with a cloud server because most of the functionalities provided by the cloud server are available on an edge server. Moreover, with the emergence of 5G really high amount of bandwidth is available to a mobile device, hence communication latency between the edge server and a mobile device is low. The combination of 5G and edge computing has a potential to better enable many interesting time-critical and/or computationally expensive computing and communication use cases, for example, connected cars. The stated combination enables development of computationally expensive applications for mobile devices and sensors nodes as these devices can offload their computational tasks to the edge server. Task offloading is a promising feature of edge computing, therefore here priority-class-based methods are presented to allocated CPU cycles to an offloaded task at the edge server. In one of the presented methods, a mobile device or a senor node assign their task to one of the available priority classes, and then offload the task to the edge server. Afterwards, the edge server assigns CPU cycles to such a task depending upon its priority class. Similarly, another method is presented that assigns priority to each device based on the type of tasks the device is offloading. While assigning CPU cycles to different offloaded tasks, the edge server uses their device priority. The simulation results demonstrate that the presented methods are capable of providing service differentiation.

Dependent task offloading mechanism for cloud–edge-device collaboration

In today's society, the rapid development of the Internet makes People's Daily life become more intelligent and diversified. Today's society has entered a multifaceted era where everything is interconnected. Artificial intelligence technology is gradually replacing some traditional human services, such as intelligent robot customer service instead of traditional human customer service, intelligent face scanning security check in railway stations instead of traditional manual ticket checking, unmanned supermarket automatic checkout has liberated some social labor costs. All these changes are the result of the development of artificial intelligence technology in today's society. In recent years, unicorn startups focused on biometrics have sprung up all around us, such as BTU and its MEG VII (Face ++). Thanks to the development of Internet and artificial intelligence technology, in many application fields, the traditional access control and identity authentication technology based on password verification is gradually transforming to the scheme based on biometric identification verification. Secure identity authentication is very important to the application of Internet. Face recognition is the most popular technology among all biometric identification technologies. In the field of biometric identification technology, it has become the most widely used technology in the field of identity authentication because of its unique non-invasive, support for infrared and visible light, no need for user cooperation and many other advantages. In the field of education, examinee identification, pedestrian identification detection at the entrance of railway stations, face electronic payment, intelligent video surveillance system, intelligent attendance and access control system, intelligent unmanned supermarkets and customs clearance ports become the pioneer fields of face recognition applications. It can be seen that the era of “national face brushing” has arrived, and the application of face recognition technology will only be more and more widespread in the current era and in the future. However, due to the sensitivity of biometric data and the heterogeneity and openness of network environment, the privacy leakage of biometric data is difficult to avoid. At present, fog computing and edge computing have been paid more and more attention in many fields. In the case that cloud service providers are unable to provide sufficient security, edge computing shows its advantages. In this paper, mobile edge computing is introduced for the first time into the face privacy protection identity authentication system based on cloud server outsourcing computing. It can not only greatly reduce the interaction frequency between users and cloud server, improve the availability and fault tolerance of the system, but also contribute to the implementation of privacy protection scheme. A deep constitutional neural network for face feature extraction is trained using deep learning framework Cafe. Cosine similarity is used to complete face verification. A privacy protection scheme based on the secure nearest neighbor algorithm is proposed, which can not only protect the security of the face feature data at the edge computing node, but also allow the edge computing node to complete the face recognition operation against the encrypted face feature data. In addition, the encryption scheme does not require large computing resources, and the accuracy of face recognition in cipher text is exactly the same as that in explain. At present, most of the solutions either have high computational complexity or poor security performance. How to reduce the computational complexity and improve the real-time performance of the system while ensuring the high security of the private data has important research significance and value. Therefore, in the cloud server outsourcing computing environment, how to complete biometric identification on the premise of protecting the privacy of biological data has become a research hot spot.

Federated deep reinforcement learning for task offloading and resource allocation in mobile edge computing-assisted vehicular networks

Task offloading and resource allocation are the major elements of edge computing. A reasonable task offloading strategy and resource allocation scheme can reduce task processing time and save system energy consumption. Most of the current studies on the task migration of edge computing only consider the resource allocation between terminals and edge servers, ignoring the huge computing resources in the cloud center. In order to sufficiently utilize the cloud and edge server resources, we propose a coarse-grained task offloading strategy and intelligent resource matching scheme under Cloud-Edge collaboration. We consider the heterogeneity of mobile devices and inter-channel interference, and we establish the task offloading decision of multiple end-users as a game-theory-based task migration model with the objective of maximizing system utility. In addition, we propose an improved game-theory-based particle swarm optimization algorithm to obtain task offloading strategies. Experimental results show that the proposed scheme outperforms other schemes with respect to latency and energy consumption, and it scales well with increases in the number of mobile devices.

The task offloading and resource allocation strategies aim to provide efficient edge computing services for ground mobile devices in mobile edge computing (MEC) environments. In this paper, we propose a UAV-assisted dynamic deployment and decision-making framework (dDDM) to address the issue of insufficient base station coverage in remote areas or emergency scenarios, which leverages the flexibility and computational capabilities of unmanned aerial vehicle (UAV) to supplement terrestrial cellular networks. We focus on optimizing the UAV trajectory to support and enhance the effectiveness of task offloading and resource allocation strategies under conditions of user mobility and stochastic task arrivals. Therefore, a joint optimization problem is formulated with the objective of minimizing the weighted sum of the total latency of ground mobile devices and the overall energy consumption of UAV. To solve this problem, the original optimization problem is decomposed into two subproblems: UAV trajectory planning, and decision-making for task offloading & computing resource allocation. For the UAV trajectory planning subproblem, we propose an improved particle swarm optimization(PSO) algorithm integrated with <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i>-means clustering, i.e., KMPSO algorithm, enabling dynamic UAV deployment in response to user mobility. For the task offloading and resource allocation decision-making, an improved black-winged kite algorithm (IBKA) is developed to determine optimal offloading and allocation strategies. Simulation results demonstrate that the proposed approach outperforms existing benchmark algorithms in terms of both latency and energy consumption.

Task offloading in edge computing is important for the Industrial Internet of Things (IIoT) to implement computation-intensive applications in real time. However, achieving efficient task offloading in IIoT is very challenging due to the limited computing resources of IIoT devices, the coupling of computing and communication resources, and the unreliability in multihop wireless transmission. In this article, we construct a link model by considering the influence of unreliable links in multihop transmission to reveal the relationship between reliability and transmission delay. Then, a nonconvex optimization problem that minimizes task processing delay is formulated, and task offloading is decided by considering transmission path selection, bandwidth allocation, and computational resource allocation. To solve this problem, an algorithm based on the alternating direction method of multipliers (ADMM) is designed using auxiliary variables and reformulation linearization technology (RLT). The simulation results show that our proposed algorithm can fully utilize the computing power of the edge server and reduce the task processing delay. Compared with the centralized algorithms, the performance of the proposed scheme is only 1% worse, but the calculation time can be reduced by 40%.

Joint task offloading and data caching in mobile edge computing networks