论文部分内容阅读
随着WLAN的普及,基于Wi-Fi的室内定位方法逐渐成为研究与应用的热点。虽然,其中基于位置指纹的定位算法研究相对广泛,应用效果较好,然而现有的指纹定位方法或系统仍存在以下3个问题:(1)离线阶段的数据标定和定位模型的训练需要耗费大量人力物力,以及时间消耗,使系统很难得到实际应用;(2)真实环境中WLAN信号波动呈现高动态性,采集的数据存在显著的时效性,无法提供长时间的有效定位保证;(3)实际环境中AP设备变动频繁,导致训练数据与定位数据特征维度不等长,造成模型失效。针对上述问题,本文提出了一种基于众包数据的模型更新方法,通过不断融合增量数据,使定位模型保持实时有效。该方法主要包括半监督极速学习机(SELM)、具有时效机制的增量式定位方法(TMELM)和特征自适应的在线极速学习机(FA-OSELM)3部分。基于上述方法,本文设计并实现了基于众包数据的室内定位平台系统。实际应用表明,本文提出的方法能够显著降低模型训练阶段的数据采集工作量,有效提升模型训练速度,并且长时间保持较高的定位精度。
With the popularization of WLAN, the indoor positioning method based on Wi-Fi has gradually become a hot spot of research and application. However, the existing fingerprinting methods and systems still have the following three problems: (1) The training of data calibration and localization model in offline phase requires a large amount of training (2) WLAN signals fluctuate dynamically in real environment, the collected data has significant timeliness and can not provide long-term effective positioning assurance; (3) AP equipment changes frequently in the real environment, resulting in unequal length of training data and positioning data feature dimensions, resulting in failure of the model. In order to solve the above problems, this paper presents a new model updating method based on crowdsourcing data, which keeps the positioning model in real time by continuously integrating incremental data. The method mainly includes three parts: SELM, TMELM and FA-OSELM. Based on the above method, this paper designs and realizes the indoor positioning platform system based on crowdsourcing data. The practical application shows that the method proposed in this paper can significantly reduce the workload of data collection in the model training stage, effectively improve the model training speed, and maintain a high positioning accuracy for a long time.