Task offloading in Edge and Cloud Computing: A survey on mathematical, artificial intelligence and control theory solutions

Cloud computing (CC) is a centralized computing paradigm that accumulates resources centrally and provides these resources to users through Internet. Although CC holds a large number of resources, it may not be acceptable by real-time mobile applications, as it is usually far away from users geographically. On the other hand, edge computing (EC), which distributes resources to the network edge, enjoys increasing popularity in the applications with low-latency and high-reliability requirements. EC provides resources in a decentralized manner, which can respond to users' requirements faster than the normal CC, but with limited computing capacities. As both CC and EC are resource-sensitive, several big issues arise, such as how to conduct job scheduling, resource allocation, and task offloading, which significantly influence the performance of the whole system. To tackle these issues, many optimization problems have been formulated. These optimization problems usually have complex properties, such as non-convexity and NP-hardness, which may not be addressed by the traditional convex optimization-based solutions. Computational intelligence (CI), consisting of a set of nature-inspired computational approaches, recently exhibits great potential in addressing these optimization problems in CC and EC. This paper provides an overview of research problems in CC and EC and recent progresses in addressing them with the help of CI techniques. Informative discussions and future research trends are also presented, with the aim of offering insights to the readers and motivating new research directions.

/pdf/a-review-on-computational-intelligence-techniques-in-cloud-goandaedec.pdf

A Review on Computational Intelligence Techniques in Cloud and Edge Computing

Review of channel assignment approaches in multi-radio multi-channel wireless mesh network

Cluster‐based mobility management algorithms for wireless mesh networks

The mobile computing world is migrating from 4G to 5G and one of the major offering of 5G is the seamless computing power and it is the major set back in the current scenario. The major difficulties that need to be addressed are computing, quality of services. Speed, power and security. This research paper aims in addressing the issue of task management in the mobile systems that is directly related to quality. The article proposes a deep learning-based algorithm that performs dynamic task offloading in the mobile cloudlet since cloudlet aids in the reduction of the delay that occur in the WLAN. The delay in performing tasks is one of the major drawback of cloudlet that it is deprived of resources when compared to cloud server due to which the tasks that are to be performed are divided and is designated to mobile devices, different cloud servers and cloudlet itself. Therefore, to determine the combination of devices required to perform different tasks, deep learning algorithms are considered. The algorithm is responsible to identify the subtasks, the subtasks that has to be computed/executed in which device or cloudlet or cloud server. The proposed algorithm is named Deep Learning based Dynamic Task Offloading in Mobile Cloudlet (DLDTO). The algorithm is implemented and compared with Cloudlet based Dynamic Task Offloading (CDTO). The overall analysis and comparison with the existing CDTO for job allocation proved that the performance of the proposed DLDTO algorithm is better in terms of energy consumption and completion time.

Deep learning based dynamic task offloading in mobile cloudlet environments

An efficient algorithm for dynamic task offloading using cloudlets in mobile cloud computing

The prediction of abnormality in the heart functionality at an early stage increases the chances of saving the life of people. Thus, this paper proposes a technique which predicts the abnormality in the functionality of the heart using heart rate in the form of beats per minute using wearable devices and deep learning model. The devices used are wrist strap and devices that can be fixed near the chest of the person or back of the person where heart beat can be detected. The proposed system is divided into 3 modules: (1) data collection and processing module, (2) prediction module and (3) communication module. First module is used to collect data and process, while prediction module predicts the abnormal functionality of the heart using deep learning model. One of the advantage of the proposed work in this paper is communication module as the communication is given to the doctor who can perform analysis before the patient reaches the hospital. A message is sent to the ambulance so that it reaches the destination on time. The message related to first aid is sent to two dear ones and the patient such that appropriate measures can be taken. The proposed technique is evaluated in terms of sensitivity, specificity, F1–Score, time and ROC curve metrics. It is also compared with the Two Stage Neural Network and TSNN and proved to be performing better.

Prediction of heart abnormalities using deep learning model and wearabledevices in smart health homes

Problem statement: In the previous researches, steps were taken to resolve the problems of multicasting, after several discussions. Actually these issues were raised, while multicasting packets from Internet Protocol (IP) to Mobile Nodes (MN). On the other hand, there is very little concern about the problem of packet loss reduction. Sometimes the occurrence of multicast service chaos is ignored during handovers. Therefore in this study it has been tried to explain the optimal multicast technique for Mobile IPv6 (MIPv6) to diminish the required amendment to the existing fast handover. Approach: An alternate method is suggested in Fast handover for MIPv6 (FMIPv6) to lessen the packet losses during handovers, before tunneling. Based on the qualities of the multicast subscription techniques for the Mobile IPv6 (MIPv6), multicast upholds method for FMIPv6. MIPv6 in general comprise two types of multicasting techniques, which are related to Home Agent (HA) and Foreign Agent (FA). These techniques help the MN to obtain the packets in roaming location referred as Foreign Network (FN), which is being forwarded by the Core Network (CN). But due to the handover latency problem MN have to experience packet loss while switching between any two Access Routers (AR). A protocol has been designed to conquer the issue of packet loss in MIPv6. Results and Conclusion: This protocol can be used during joining process in MIPv6 before tunneling to eliminate the IP connectivity time. This implementation allows a mobile node to be connected more quickly at a fresh point of connection, when that mobile node moves with less packet losses.

/pdf/a-fast-handover-scheme-for-multicasting-in-ipv6-based-mobile-1i9jh7hbio.pdf

A Fast Handover Scheme for Multicasting in IPv6 based Mobile Ad hoc Networks

In wireless mesh networks, the availability of spectrum should be utilized efficiently because the exchanging of traffic rate of multimedia applications is increasing day to day. The traffic of multimedia content may degrade the network performance in terms of high packet loss and throughput degradation because of asynchronous behavior between the sender and the receiver. The proposed Efficient Channel Reservation and Assignment EchRA results in successful originating and handing off of multimedia call transmissions using the four dimensional Markov chain process. The EChRA algorithm, proposed in this paper, achieves the maximum throughput performance considering the parameters of blocking probability, dropping probability, and synchronized dwell time between the source and destination. The quality of service is guaranteed by sharing the available slots and by using efficient queuing model to service the incoming and handoff calls, and henceforth, the successful completion of handoff calls rate is increased. Copyright © 2015 John Wiley & Sons, Ltd.

M. Pounambal

Papers

Deep learning based dynamic task offloading in mobile cloudlet environments

An efficient algorithm for dynamic task offloading using cloudlets in mobile cloud computing

Prediction of heart abnormalities using deep learning model and wearabledevices in smart health homes

A Fast Handover Scheme for Multicasting in IPv6 based Mobile Ad hoc Networks

Efficient channel assignment method for multimedia traffic in wireless mesh networks