基于贪婪算法的参与式感知激励分配机制

Incentiveallocation mechanism in participatory sensing based on greedy algorithm

针对参与式感知激励机制,提出一种基于贪婪算法的以任务发布者的有限预算为约束,以每块子区域的激励分配为优化问题的激励机制。考虑在数据收集过程中样本的数量和分布的情况,引入加权熵作为定量指标来评估样本的分布,发现数据样本的分布是感知结果准确性的重要因素。仿真结果表明,与逆向拍卖机制和进化算法激励模型(incentive-based evolutionary algorithm,IEA)相比,所提激励机制能得到更加精确的感知结果。

According to incentive mechanisms in participatory sensing,an incentive mechanism based on greedy algorithm was proposed.The constraint was the limited budget of the task publisher.The optimization goal was incentive allocation in each subregion.The method considered both the amount and distribution of samples in data collection,and introduced weighted entropy as a quantitative metric to evaluate the distribution of samples and found that the distribution of data samples was another important factor to the accuracy of sensing result.Experimental results show that the proposed incentive mechanism can obtain more accurate sensing results compared with the reverse auction mechanism and IEA.