Edge-Assisted Workload-Aware Image Processing System

Edge computing, which locates edge servers with limited computing and storage resources at the edge of networks, is expected as a novel architecture for low-latency applications. In edge computing, users offload a task to edge servers due to poor computing resources and batteries of mobile devices so that the edge servers execute the offloaded task and return the result of it to users. Users can thus enjoy various applications without depending on the limitations of mobile devices. However, when the edge server’s load is too heavy, a large number of tasks will be offloaded to distant cloud servers. The long distance between users and cloud servers significantly degrades the quality of mobile applications. To prevent this problem, we propose flow splitting and aggregation schemes to improve offload delay to cloud servers in edge computing. This scheme splits TCP connections between users and cloud servers at the edge server, and then aggregates TCP connections between the edge server and cloud servers. We show the effectiveness of our scheme through simulation evaluations.

Dependent task offloading mechanism for cloud–edge-device collaboration

In many parts of the world including south-east Asia, a lot of landslides occur every year. There are a few IoT technologies exist that allow landslides' monitoring but there is a requirement of a more reliable and efficient Landslide Early Warning System (LEWS). The communication between the IoT nodes and the cloud is vulnerable to link failure due to various types of disruptions in mountainous regions. In real life scenarios, this connection loss might hinder the decision making at the cloud. So, the IoT system must not fail in any given situation. Taking reliability into consideration, this paper provides an edge computing based solution to resolve the issue. Recently, edge computing has emerged as an effective solution to decrease latency for delay sensitive IoT applications. Furthermore, it has scope to make an IoT application such as LEWS more reliable as edge server can keep on system running in case of cloud's failure or communication failure between IoT node and cloud. This paper demonstrate the implementation of reliable data processing so that even if the connection is lost between the source/coordinator node and the cloud server, the data can still be processed and feedback are obtained. During implementation, the edge server has a limited computing and storage resources, but enough to process and analyze landslide data such as rain-fall, pore pressure, moisture, and displacement, to produce meaningful results similar to the cloud. The system keeps on working even if there is a network failure between edge server and cloud server or cloud server crashes.

Mobile edge computing (MEC) provides a new distributed computing paradigm that overcomes the inability of cloud computing to offer low end-to-end latency. In a MEC environment, app vendors can deliver lower-latency services to mobile app users by placing applications on edge servers in close proximity to app users. From an app vendor's perspective, an optimal edge application placement strategy under a budget ( <inline-formula><tex-math notation="LaTeX">$k$</tex-math></inline-formula> ) constraint aims to place application instances on <inline-formula><tex-math notation="LaTeX">$k$</tex-math></inline-formula> edge servers within a specific area to maximize user coverage. However, edge servers may be subject to failure due to multiple reasons, e.g., hardware faults, software exceptions, cyber-attacks, etc. App users served by failed edge servers need to access applications from remote cloud servers if they cannot access any other edge servers. This impacts app users’ quality of experience significantly. Thus, app vendors need to consider the reliability of the edge server network when choosing edge servers for placing their application instances. We make the first attempt in this paper to tackle this problem of joint optimization of coverage and reliability for edge application placement (EAP-CR). We formulate this problem as a constrained optimization problem and prove its <inline-formula><tex-math notation="LaTeX">$\mathcal {NP}$</tex-math></inline-formula> -hardness theoretically. Then, we propose an optimal approach to find the optimal solutions with the integer programming technique, an approximation approach is also proposed to find approximate solutions for large-scale EAP-CR problems. We evaluate EAP- <inline-formula><tex-math notation="LaTeX">$OPT$</tex-math></inline-formula> and EAP- <inline-formula><tex-math notation="LaTeX">$APX$</tex-math></inline-formula> against three relevant approaches through experiments conducted on a widely-used real-world data set and a synthetic data set. The results demonstrate that our proposed approaches can solve the EAP-CR problem effectively and efficiently.

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.

Mobile applications can leverage high-quality deep learning models such as convolutional neural networks and deep neural networks to provide high-performance cognitive services. Prior work on deep learning models-based mobile applications in a cloud-edge computing environment focuses on performing lightweight data pre-processing tasks on edge servers for cloud-hosted cognitive servers. These approaches have two major limitations. First, it is uneasy for the mobile applications to assure satisfactory user experience in terms of network communication delay, because the intermediary edge servers are used only to pre-process data (e.g., images and videos) and the cloud servers are used to complete the tasks. Second, these approaches assume the pre-trained deep learning models deployed on cloud servers are static, and will not attempt to automatically upgrade in a context-aware manner. In this article, we propose a cloud-edge collaboration framework that facilitates delivering cognitive services with long-lasting, fast response, and high accuracy properties. We fist deploy a shallow model (i.e., EdgeCNN) on the edge server and a deep model (i.e., CloudCNN) on the cloud server. EdgeCNN can provide durable and rapid response cognitive services, because edge servers not only provide computing resources for mobile applications, but also close to users. Then, we enable CloudCNN to assist in training EdgeCNN to improve the performance of the latter. Thus, EdgeCNN also provides high-accuracy cognitive services. Furthermore, because users may continue to upload data to edge servers in real-world scenarios, we propose to use the ongoing assistance of CloudCNN to further improve the accuracy of the shallow model. Experimental results show that EdgeCNN can reduce the average response time of cognitive services by up to 55.08 percent and improve accuracy by up to 26.70 percent.

This paper shows security and privacy of user data stored in cloud services is of utmost concern especially for commercial public cloud platforms and service providers. The user data stored on cloud is encrypted, and it is desired that data mining operations on encrypted data must be executed efficiently. The proposed system implements a data mining operation namely clustering on encrypted data on cloud platforms using TCP / IP protocol. The clustering operation is performed using standard K-Means, Spectral, Hierarchical, Hybrid DBSCAN and simple DBSCAN clustering technique optimized for encrypted data.

A Novel Lyapunov based Dynamic Resource Allocation for UAVs-assisted Edge Computing

Over the Virtual Top. Digital Service Value Chain Disintermediation Implications for Hybrid Heterogeneous Network Regulation

One of the challenges in autonomous driving is limited sensing from a single vehicle that causes spurious warnings and dead-lock situations. We posit that cloud-based vehicle control system[1] is promising when a number of vehicles must be controlled, since we can collect information from sensors across multiple vehicles for coordination. However, since cloud based control has inherent challenge in long-haul communication susceptible to prolonged latency and packet loss caused by congestion, mobile edge computing (MEC)[2] recently attracts attention in ITS in the next generation mobile network such as 5G. Although edge servers can perform data processing from the vehicles in ultra low latency in MEC, computational resources at edge servers are limited compared to cloud. Therefore, dynamic resource allocation and coordination between edge and cloud servers are necessary. In this paper, we propose infrastructure-based vehicle control system that shares internal states between edge and cloud servers, dynamically allocates computational resources and switches necessary computation on collected sensors according to network conditions in order to achieve safe driving. We implement a prototype system using micro-cars and evaluate the stability of infrastructure-based vehicle control. We show that proposed system mitigates instability of cloud control caused by latency fluctuation. As a result, when controlled from the cloud with 150ms latency, micro-cars deviate by over 0.095m from the course for the 40% of the entire trajectory possibly causing car accidents. On the other hand, MEC-based control stabilizes the driving trajectory. Also, our proposed system automatically switches control from cloud and from edge server according to the network condition without degrading the stability in driving trajectory. Even when the ratio of time of control by edge server to that by cloud is suppressed to 54%, we can achieve almost the same stability as in full control by edge controller.

With the rapid development of the Internet of Things, a large number of smart devices are being connected to the Internet while the data generated by these devices have put unprecedented pressure on existing network bandwidth and service operations. Edge computing, as a new paradigm, places servers at the edge of the network, effectively relieving bandwidth pressure and reducing delay caused by long-distance transmission. However, considering the high cost of deploying edge servers, as well as the waste of resources caused by the placement of idle servers or the degradation of service quality caused by resource conflicts, the placement strategy of edge servers has become a research hot spot. To solve this problem, an edge server placement method orienting service offloading in IoT called EPMOSO is proposed. In this method, Genetic Algorithm and Particle Swarm Optimization are combined to obtain a set of edge server placements strategies, and Simple Additive Weighting Method is utilized to determine the most balanced edge server placement, which is measured by minimum delay and energy consumption while achieving the load balance of edge servers. Multiple experiments are carried out, and results show that EPMOSO fulfills the multiobjective optimization with an acceptable convergence speed.

Cloud computing is a prevailing approach for running various data-analysis algorithms to obtain insight from massive raw data. However, the current cloud computing architecture suffers from large computation and communication overhead when processing data in big size. In this work, we develop a systematic approach to efficiently process large-scale image processing tasks under an edge-assisted cloud computing architecture. We use a two-stage processing mechanism, which is to pre-process raw data on edge servers and feed only the most relevant information to the cloud, for the purpose of reducing system overhead. We model and formulate a problem that jointly optimizes the responding strategy and task placement strategy. We then propose a solution with a queuing mechanism to capture the impact of background workload on heterogeneous edge devices. We implement our system and use extensive experiments to validate its efficacy and efficiency in practice.

Today, with the presence of 5G communication systems, including Internet of Things (IoT) technology, there is a high demand for mobile devices (especially smartphones, tablets, wearable technology, and so on). Regarding this proliferation and high demand, the massive adoption of mobile devices (MDs) has led to an exponential increase in network latency; the heavy demand for cloud servers causes the degradation of data traffic, which considerably impacts the real-time communication and computing aspects of mobile devices. As a result, mobile edge computing (MEC), an efficient framework capable of enhancing processing, optimizing energy usage, and offloading computation tasks, is considered a promising solution. In current research, numerous models have been implemented to achieve resource allocation and task offloading. However, these techniques are ineffective due to privacy issues and a lack of sufficient resources. Hence, this study proposes secure task offloading and resource allocation strategies in mobile devices using the Probit Mish–Gated Recurrent Unit (PM-GRU) and Entropic Linear Interpolation-Serval Optimization Algorithm (ELI-SOA). Primarily, the tasks to be offloaded and their attributes are gathered from mobile users and passed to a local computing model to identify the edge server. Here, the task attributes and the server attributes are compared with a cache table using the Sorensen–Dice coefficient. If the attributes match, then details about the appropriate edge server are produced. If the attributes do not match, then they are inputted into a global scheme that analyzes the attributes and predicts the edge server based on the Probit Mish-Gated Recurrent Unit (PM-GRU). Then, the server information is preserved and updated in the cache table in the local scheme. Further, the attributes, along with the predicted edge server, are inputted into a system for privacy-preserving smart contract creation by using Exponential Earth Mover’s Distance Matrix-Based K-Anonymity (EEMDM-KA) to develop a secure smart contract. Subsequently, the traffic attributes in the smart contract are extracted, and the request load is balanced by using HCD-KM. Load-balanced requests are assigned to the edge server, and the optimal resources are allocated in the cloud server by using the Entropic Linear Interpolation-Serval Optimization Algorithm (ELI-SOA). Finally, the created smart contract is hashed based on KECCAK-512 and stored in the blockchain. With a high accuracy of 99.84%, the evaluation results showed that the proposed approach framework performed better than those used in previous efforts.

A tremendous amount of visual data are bing collected by the Internet of Video Things (IoVT) systems in which ubiquitous cameras deployed in cities enable new applications in the domains of smart transportation and public security. However, the limited resources in terms of communication, computing, and caching (3C) in the conventional cellular network make it challenging to adopt centralized artificial intelligence (AI) to conduct real-time video-based data analytics. In this work, based on the 5G network architecture with edge servers, a three-phase resource-effective solution is proposed to perform surveillance operations in a large-scale wireless IoVT. The proposed strategy integrates front-end cameras with simple on-chip neural networks performing real-time object-of-interest segmentation, edge servers, and cloud servers with AI functionality carrying out image-based target recognition and video-based target analytics tasks. More importantly, we design the optimal 3C strategy to achieve the best video analytics performance constrained by computing offload ratio, network resource allocation and video-related parameters. Extensive simulations with deep neural networks implemented both at the front-end cameras and in the cloud server have validated the effectiveness of the proposed solution.

Federated Learning (FL) enables cooperative model training across dispersed edge devices while protecting data privacy and providing localized insights without the need for centralized data aggregation. In the Internet of Things (IoT), federated learning enables cooperative model training among dispersed edge devices while protecting data privacy and providing localized insights without the need for centralized data aggregation. Nevertheless, Federated Learning&rsquo;s local model sharing approach makes it susceptible to Man-in-the-Middle (MITM) attacks. In FL, attackers have the ability to manipulate local models. As a result, a global model produced from the altered local models may be inaccurate. In this paper, we suggest a blockchain-based FL architecture to prevent intermediaries from readily altering the FL parameters throughout the transmission process. All of the clients&#039; parameters are combined by a cloud server, which is acting as the federated parameter server. By integrating blockchain technology into the overall architecture, we link all cloud and edge servers. This paper uses a PoC algorithm based on the SHA-256 hashing function in order to validate the data. The results and comparison analysis show that the suggested framework has a low false-positive rate and a high accuracy in detecting MITM attacks in their early stages, with a detection rate ranging from 98 to 100%.