In this article, we consider the problem of distributed offloading in mobile crowdsensing (MCS) by the means of mobile edge computing(MEC). Deploying MEC in MCS can help address many challenges the centralized MCS solutions are facing such as delays in answering real-time requirements due to the centralized nature of the solution, discovering and selecting non-connected devices in the Area of Interest (AoI), and dealing with the complexity of data computation. Specifically, we propose to improve the selection of crowdsourced workers by opting for a distributed mechanism, where the selection is partially offloaded to the Local Edge Nodes (LENs). The proposed framework, OffSEC, relies on a) a Mobile edge computing architecture that defines the Main Edge Node (MEN) and LENs responsible for selecting the local workers available in the AoI and b) a two-layer selection mechanism that helps to offload the selection of crowdsourced workers to the identified LENs. To do this, nodes in the area of interest are first clustered based on their locations, and then for each cluster, one LEN is identified based on the closeness metrics. MEN is then nominated based on a greedy selection. Finally, LENs discover the available nodes in their cluster, including heterogeneous IoT nodes and workers that are not necessarily connected to the Edge server, and select the final list of workers that maximize the quality of service (QoS). The process of selection is dynamic as it is updated according to the requested task. The proposed OffSEC is evaluated using a real dataset and is compared to a centralized approach. The results show that OffSEC outperforms the benchmark by maximizing the QoS of the sensing activities and improving the quality of the collected data readings (QoDR).

中文翻译：

---

用于异构物联网的基于移动边缘的 CrowdSensing 框架

在本文中，我们通过移动边缘计算（MEC）来考虑移动人群感知（MCS）中的分布式卸载问题。在 MCS 中部署 MEC 可以帮助解决集中式 MCS 解决方案面临的许多挑战，例如由于解决方案的集中性质而导致响应实时要求的延迟、发现和选择感兴趣区域 (AoI) 中的未连接设备，以及处理数据计算的复杂性。具体来说，我们建议通过选择分布式机制来改进众包工人的选择，其中选择部分卸载到本地边缘节点 (LEN)。提议的框架 OffSEC，依赖于 a) 定义主边缘节点 (MEN) 和 LEN 的移动边缘计算架构，负责选择 AoI 中可用的本地工作人员，以及 b) 两层选择机制，有助于将众包工作人员的选择卸载到识别 LEN。为此，感兴趣区域中的节点首先根据它们的位置进行聚类，然后对于每个聚类，根据接近度度量确定一个 LEN。然后根据贪婪的选择提名 MEN。最后，LEN 发现其集群中的可用节点，包括异构 IoT 节点和不一定连接到边缘服务器的工作人员，并选择能够最大限度提高服务质量 (QoS) 的工作人员的最终列表。选择过程是动态的，因为它会根据请求的任务进行更新。提议的 OffSEC 使用真实数据集进行评估，并与集中式方法进行比较。结果表明，OffSEC 通过最大化传感活动的 QoS 和提高收集的数据读数 (QoDR) 的质量而优于基准。