GPS感測器在移動設備上的普及化，使得新一代基於位置的服務越來越流行。這些應用程式讓用戶可以記錄自己的旅行路線，並上傳上自己的GPS軌跡點去做一些即時社交分享，慢跑日記，記錄生活，醫療保健。然而，這些GPS路徑點(GTPs-GPS track points)常夾帶了多餘或不精確的資料點，這些冗餘的GTPs不僅僅占用存儲空間，直接上傳所有GTPs 更額外的耗費電池能量，占用頻寬。為了解決這個問題，我們提出了此論文，稱之為 CET: 有效率的GPS軌跡特徵點擷取技術。目的是從一個GPS trace中過濾出有特徵點的GTPs，用來表徵和重建該用戶行經的路線。透過僅儲存壓縮後的有特徵點的GTPs，使移動設備需要更少的存儲空間，並降低發送它們所需要的頻寬和能量消耗，從而導致優化的GPS功能的移動應用。這樣特性更是符合新興的可穿戴設備設計原則。最後我們的實驗進行了兩個trace driven模擬和真實的使用者經驗去比較CET和其他演算法的優劣。實驗結果顯示：（1）CET可壓縮GPS traces 的比例高達33倍，（2）經由CET壓縮後的GTPs能夠精確地代表道路路段，和（3）CET最多可節省能源消耗達72％。

The popularity of GPS sensors in mobile devices has enabled a new generation of location-based services, in which users record their traveling routes and upload the GPS track points (GTPs) on-the- fly for real-time social sharing, jog journaling, life logging, health care, and map generation. However, directly uploading all GTPs wastes battery energy, storage space, and network bandwidth on the mobile device, because GTPs are highly redundant. To reduce the redundancies, we propose in this thesis the Corner Extraction Technique (CET) to extract the corner points from the GTPs of individual users that can be used to characterize and reconstruct the routes that the user has traveled. Storing corner GTPs requires much less storage space, and transmitting them allows saving in both network bandwidth and energy consumption, leading to optimized GPS-enabled mobile applications. Such features are especially attracted for the emerging wearable devices. We have conducted both trace- driven simulations and real experiments to demonstrate the merits of the proposed CET approach. The experimental results indicate that: (1) CET results in up to 33 times of compression ratio, (2) CET closely follows the original road segments, and (3) CET saves energy consumption by up to 72%.

1 Introduction 1
2 Related Works 4
3 Corner Extraction Technique 8
3.1 Problem Statement 8
3.2 Basic Idea 9
3.3 Rectangle Method 10
3.4 Solution for Clustering of GTPs 13
3.5 Solution for Back-and-Forth Patterns 14
3.6 Solution for CET under Lower Sampling Rate 16
4 Experiments 18
4.1 Experimental Settings 18
4.1.1 Dataset 18
4.1.2 Error Metric 19
4.1.3 Parameter Selection 19
4.2 Experimental Results 20
4.2.1 Data Compression Ratio 20
4.2.2 Distance of Road-Matching 21
4.2.3 CET under Lower Sampling Rate 21
4.2.4 How does speed affect CET 23
4.2.5 Power Saving Ratio 23
4.2.6 CET in Field Test 24
5 Conclusion 27

[1] BALASUBRAMANIAN, N., BALASUBRAMANIAN, A., AND VENKATARAMANI, A. Energy consumption in mobile phones: a measurement study and implications for network applications. In Proc. of the ACM SIGCOMM Conference on Internet Measurement Conference (IMC) (2009).
[2] BANFIELD, J. D., AND RAFTERY, A. E. Model-based Gaussian and Non-Gaussian clustering. Biometrics (1993).
[3] BIAGIONI, J., AND ERIKSSON, J. Map inference in the face of noise and disparity. In Proc. of the International Conference on Advances in Geographic Information Systems (2012).
[4] CAO, L., AND KRUMM, J. From GPS traces to a routable road map. In Proc. of the ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (2009).
[5] DAVICS, J., BERESFORD, A., AND HOPPER, A. Scalable, distributed, real-time map genera- tion. IEEE Pervasive Computing (2006).
[6] DOUGLAS, D., AND PEUCKER., T. Algorithms for the reduction of the number of points re- quired to represent a digitised line or its caricature. The Canadian Cartographer. 1973.
[7] DummyNet. http://info.iet.unipi.it/~luigi/dummynet/.
[8] ET AL, E. A density-based algorithm for discovering clusters in large spatial databases with noise. In Proc. of the Second International Conference on Knowledge Discovery and Data Mining (KDD-96) (1996).
[9] JUAN, S., CHUNG, Y., KING, C., AND HSU, C. CEGF: Corner extraction by GPS filtering for power-efficient location uploading. In Proc. of the International Conference on Mobile Systems, Applications, and Services (MobiSys) (2013).
[10] LOU, Y., ZHANG, C., ZHENG, Y., XIE, X., WANG, W., AND HUANG, Y. Map-matching for low-sampling-rate GPS trajectories. In Proc. of the ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (GIS) (2009).
[11] MACQUEEN, J. Some methods for classification and analysis of multivariate observations. In Proc. of Berkeley Symposium on Mathematical Statistics and Probability (1967).
[12]MAZHELIS.,O.Using recursive Bayesian estimation for matching GPS measurements to imperfect road network data. In Proc. of International IEEE Conference on Intelligent Transportation Systems (ITSC) (2010).
[13] MERATNIA, N., AND BY, R. Spatiotemporal compression techniques for moving point objects. In Proc. of International Conference on Extending Database Technology (EDBT). 2004.
[14] MUCKELL, J., HWANG, J., PATIL, V., LAWSON, C., PING, F., AND RAVI, S. SQUISH: An online approach for GPS trajectory compression. In Proc. of the International Conference on Computing for Geospatial Research (2011).
[15] OpenStreetMap. http://www.openstreetmap.org/.
[16] POTAMIAS, M., PATROUMPAS, K., AND SELLIS, T. Sampling trajectory streams with spa-
tiotemporal criteria. In Proc. of the Scientific and Statistical Database Management (2006).