WITHDRAWN: A Survey on Local Transport Mode Detection on The Edge

Abstract

We present a survey of solutions for smartphone-based transport mode detection. These are divided into local and remote approaches being the first ones addressed in this article. A local approach performs the following steps in the smartphone (and not in some far away cloud server): 1) data collection or sensing, 2) preprocessing, 3) feature extraction, and 4) classification (with a previous machine learning based training phase). Solutions are presented taking into account the above mentioned four steps, and analyzed according to the most relevant requirements (accuracy, delay taken to detect a transport mode, resources consumption, and generalization).

Intelligent Scene Recognition Based on Deep Learning

“…It's the most widely used to leverage built-in sensors in almost every smartphone. A hybrid approach using multiple sensors outperformed ones that only used one single sensor [20]. Remote-based scene recognition pays little attention to power control and computational complexity [2].…”

“…As far as delay is concerned, [20] claimed all the local TMD approaches have implicitly met this requirement by performing classification locally (in comparison with remote-based methods). However, their definition of delay mostly meant the classification delay except for [25].…”

“…With the increasing capability of mobile computing, smartphones are gradually competent to deal with self-generated data rather than sending data to a remote cloud server. Local TMD solutions surveyed in [20] outperform the remote ones greatly in many ways, especially detection delay [21]. Apart from travel modes, other complicate scenes are also worthy of consideration, such as the elevator and…”