多样性和规模问题：使用单个蜂窝塔信息的普适交通模式检测
The Diversity and Scale Matter: Ubiquitous Transportation Mode Detection using Single Cell Tower Information
摘要：
检测用户的运输模式对于广泛的应用是重要的。虽然许多最近的系统使用无处不在的移动电话来解决交通模式检测问题，但这些研究利用GPS、惯性传感器和/或多个蜂窝塔信息。然而，这些不同的电话传感器具有高能量消耗，限于电话的小子集（例如，高端电话或支持相邻蜂窝塔信息的电话），不能在某些区域中工作（例如，在用于GPS的隧道内），和/或仅从用户侧工作。在本文中，我们提出了一个交通模式检测系统，MonoSense，利用手机服务小区的信息。其基本思想是，手机速度可以与从服务蜂窝塔ID和接收信号强度中提取的特征相关联。为了利用这些有限的信息实现高检测精度，MonoSense利用沿多个轴的多样性来提取新特征。具体地，MonoSense从不同滑动窗口大小上的服务小区塔可用的时域和频域信息两者中提取特征。更重要的是，我们还表明，对数和线性RSS尺度可以提供不同的信息，手机的运动，进一步丰富的特征空间，并导致更高的准确性。MonoSense使用覆盖485公里的135小时蜂窝轨迹进行评估，并由四名用户使用不同的Android手机收集，结果表明，它可以在仅使用服务蜂窝塔信息的情况下区分静止，步行和驾驶模式时分别实现89.26%和89.84%的平均准确率和召回率，突出了MonoSense能够实现广泛的智能交通应用的能力。
英文摘要：
Detecting the transportation mode of a user is important for a wide range of applications. While a number of recent systems addressed the transportation mode detection problem using the ubiquitous mobile phones, these studies either leverage GPS, the inertial sensors, and/or multiple cell towers information. However, these different phone sensors have high energy consumption, limited to a small subset of phones (e.g. high-end phones or phones that support neighbouring cell tower information), cannot work in certain areas (e.g. inside tunnels for GPS), and/or work only from the user side. In this paper, we present a transportation mode detection system, MonoSense, that leverages the phone serving cell information only. The basic idea is that the phone speed can be correlated with features extracted from both the serving cell tower ID and the received signal strength from it. To achieve high detection accuracy with this limited information, MonoSense leverages diversity along multiple axes to extract novel features. Specifically, MonoSense extracts features from both the time and frequency domain information available from the serving cell tower over different sliding widow sizes. More importantly, we show also that both the logarithmic and linear RSS scales can provide different information about the movement of a phone, further enriching the feature space and leading to higher accuracy. Evaluation of MonoSense using 135 hours of cellular traces covering 485 km and collected by four users using different Android phones shows that it can achieve an average precision and recall of 89.26% and 89.84% respectively in differentiating between the stationary, walking, and driving modes using only the serving cell tower information, highlighting MonoSense ability to enable a wide set of intelligent transportation applications.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
