Deep learning is a promising approach for extracting accurate information from raw sensor data from IoT devices deployed in complex environments. Because of its multilayer structure, deep learning is also appropriate for the edge computing environment. Therefore, in this article, we first introduce deep learning for IoTs into the edge computing environment. Since existing edge nodes have limited processing capability, we also design a novel offloading strategy to optimize the performance of IoT deep learning applications with edge computing. In the performance evaluation, we test the performance of executing multiple deep learning tasks in an edge computing environment with our strategy. The evaluation results show that our method outperforms other optimization solutions on deep learning for IoT.
4 May 2016 in IEEE Transactions on Vehicular Technology
The vehicle ad hoc network (VANET) is an emerging network technology that is expected to be cost-effective and adaptable, making it ideal to provide network connection service to drivers and passengers on today's roads. In the next generation of VANETs with fifth-generation (5G) networks, software-defined networking (SDN) technology will play a very important role in network management. However, for infotainment applications, high latency in VANET communication imposes a great challenge for network management, whereas direct communication through the cellular networks brings high cost. In this paper, we present an optimizing strategy to balance the latency requirement and the cost on cellular networks, in which we encourage vehicles to send the SDN control requests through the cellular networks by rebating network bandwidth. Furthermore, we model the interaction of the controller and vehicles as a two-stage Stackelberg game and analyze the game equilibrium. From the experimental results, the optimal rebating strategy provides smaller latency than other control plane structures.
16 April 2019 in ACM Transactions on Internet Technology
Mobile crowdsensing becomes a promising technology for the emerging Internet of Things (IoT) applications in smart environments. Fog computing is enabling a new breed of IoT services, which is also a new opportunity for mobile crowdsensing. Thus, in this article, we introduce a framework enabling mobile crowdsensing in fog environments with a hierarchical scheduling strategy. We first introduce the crowdsensing framework that has a hierarchical structure to organize different resources. Since different positions and performance of fog nodes influence the quality of service (QoS) of IoT applications, we formulate a scheduling problem in the hierarchical fog structure and solve it by using a deep reinforcement learning–based strategy. From extensive simulation results, our solution outperforms other scheduling solutions for mobile crowdsensing in the given fog computing environment.
5 September 2014 in IEEE Transactions on Cloud Computing
Software-defined network (SDN) is the next generation of networking architecture that is dynamic, manageable, cost-effective, and adaptable, making it ideal for the high-bandwidth, dynamic nature of today's applications. In SDN, network management is facilitated through software rather than low-level device configurations. However, the centralized control plane introduced by SDN imposes a great challenge for the network security. In this paper, we present a secure SDN structure, in which each device is managed by multiple controllers, not just a single as in a traditional manner, with the dynamic and isolated instance provided by the cloud. It can resist Byzantine attacks on controllers and the communication links between controllers and SDN switches. Furthermore, we study a controller minimization problem with security requirement and propose a cost-efficient controller assignment algorithm with a constant approximation ratio. From the experiment result, the secure SDN structure has little impact on the network latency, provide better security than general distributed controller, and the proposed algorithm performs higher efficiency than random assignment.
Edge-centric computing (ECC) and content- centric networking (CCN) will be the most important technologies in future 5G networks. However, due to different architectures and protocols, it is still a challenge to fuse ECC and CCN together and provide manageable and flexible services. In this article, we present ECCN, an orchestrating scheme that integrates ECC and CCN into a hierarchical structure with software defined networking (SDN). We introduce the SDN technology into the hierarchical structure to decouple data and control planes of ECC and CCN, and then design an SDN protocol to control the data forwarding. We also implement two demonstration applications in our testbed to evaluate the ECCN scheme. The experimental results from the testbed applications, and extensive simulations show ECCN outperforms original structures.
A smart evacuation needs a scalable and flexible system to provide service in both emergency and normal situations. A single cloud service is usually limited to support scaling up requirements in an emergency, especially one with a large geographic scope. In this article, the authors propose MCES, a multicloud architecture that deploys smart evacuation services in multiple cloud providers and that can tolerant more pressure than single cloud-based services. This system maintains basic service to support monitoring, but during an emergency, visits to the service will scale up enormously, which means MDSE must support a rapid scaling up of service capacity in a short time. The authors use a three-layer cloud instance management to support rapid capacity scaling in MCES. By conducting extensive simulations, the authors demonstrate that their proposed MCES significantly outperforms single cloud solutions under various emergency settings.