Incentive Mechanism and Content Provider Selection for Device-to-Device-Based Content Sharing

Incentive Mechanism and Content Provider Selection for Device-to-Device-Based Content Sharing
复制标题

DOI:
10.1109/tvt.2019.2895326
复制
发表时间:
2019-01
影响因子:
6.8
通讯作者:
Jianzhang He;Haibo Wang;Xiaoli Chu;Zhang Tao
Jianzhang He;Haibo Wang;Xiaoli Chu;Zhang Tao
中科院分区:
计算机科学2区
文献类型:
--
作者:
Jianzhang He;Haibo Wang;Xiaoli Chu;Zhang Tao

文献摘要

被引文献

相似文献

基于设备到设备(D2D)通信的内容共享被认为是从过载的蜂窝网络中分流流量的一种很有前途的技术。高效的D2D内容共享需要一种激励机制来鼓励移动设备参与,如果存在多个候选提供商,则还需要最佳的内容提供商选择方案。在本文中,我们提出了一种综合评分机制(CSM),它根据每个候选内容提供商的历史内容提供记录、当前传输速率和预期回报来计算评分。CSM建立历史内容供应记录和预期报酬之间的关系,并且使得能够选择具有适合于所请求的内容的可实现传输速率的内容提供商。基于CSM和匈牙利算法,我们提出了一种内容共享激励和提供商选择算法(CIPS)来优化多个并发内容请求者的内容提供商选择。通过大量的仿真实验表明,该算法能够有效地激励移动设备参与内容共享,并能从多个候选者中选择最合适的内容提供商(S)。
Content sharing based on device-to-device (D2D) communications has been regarded as a promising technology to offload traffic from the overburdened cellular networks. Efficient D2D content sharing requires an incentive mechanism to encourage mobile devices to participate, and the optimal content-provider selection scheme is also necessary if multiple candidate providers exist. In this paper, we propose a comprehensive scoring mechanism (CSM), which calculates a score for each candidate content provider based on their historical content supply record, current transmission rate, and expected reward. The CSM establishes the relationship between the historical content supply record and the expected reward, and makes it possible to select the content provider with an achievable transmission rate appropriate for the requested content. Based on the CSM and the Hungarian algorithm, we propose a Content-sharing Incentive and Provider Selection (CIPS) algorithm to optimize the selection of content providers for multiple concurrent content requesters. Through extensive simulations, we show that the proposed CIPS algorithm can effectively motivate mobile devices to participate in content sharing and can select the most appropriate content provider(s) from multiple candidates.