Unveiling the Missed 4.5G Performance In the Wild
Unveiling the Missed 4.5G Performance In the Wild
复制标题
揭示野外错过的 4.5G 性能
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Chunyi Peng
中科院分区：
文献类型：
--
作者：
Haotian Deng;Kai Ling;Junpeng Guo;Chunyi Peng
在本文中，我们研究了操作蜂窝网络中一个重要但未探索的问题：每个移动设备实际上与设备充其量的实际性能有多近？从根本上讲，这为设备的可跨度性能差距增添了数量。由于设备无法实现可行的，更高的性能，因此被认为是错过的性能。在这项工作中，我们报告了我们的初步结果，这些结果揭示了，量化和理解这种错过的野外表现。我们已经在西拉斐特的小城市进行了广泛的测量，以量化和分析四个顶级美国航空公司的性能差距，所有这些绩效差距都运行4.5克LTE高级。我们的研究表明，重要的表现确实发生了，并且经常发生。我们已经看到，下行链路吞吐量仅为750 kbps，而一个位置可用且可行的47Mbps（60倍）；在超过20％的位置，四个载体的丢失数据吞吐量（比率）不少于30 Mbps（7.4倍，AT＆T），22.6 Mbps（6.8倍，Verizon），12.1 Mbps（6.9倍，T-Mobile）和25 Mbps和25 Mbps（1.7倍，Sprint）。更糟糕的是，与当前基础架构兼容的某些操作（如我们的研究所证实），可以很容易地发挥这种遗漏的性能潜力。我们进一步分析了其根本原因，并得出结论，当前选择疗法细胞的细胞实践应承担责任。
In this paper, we study an important, yet unexplored problem in operational cellular networks: How close is the actual performance each mobile device gets in reality from what the device could have got at best? Fundamentally, this gauges the bridgeable performance gap for the device. It is deemed as the missed performance because the device fails to achieve the feasible, higher performance. In this work, we report our preliminary results that uncover, quantify, and understand such missed performance in the wild. We have conducted extensive measurements in the small city of West Lafayette, IN to quantify and analyze such performance gaps over four top-tier US carriers, all running 4.5G LTE-Advanced. Our study shows that, significant performance miss indeed happens and happens quite frequently. We have seen that, downlink throughput is merely 750 Kbps, while 47Mbps is available and feasible (60x miss) at one location; At more than 20% places, the missed data throughput (ratio) for four carriers is no less than 30 Mbps (7.4x, AT&T), 22.6 Mbps (6.8x, Verizon), 12.1 Mbps (6.9x, T-Mobile) and 25 Mbps (1.7x, Sprint), respectively. Even worse, such missed performance potentials can be readily reached with certain operations compatible with the current infrastructure (as confirmed by our study). We further analyze its root cause, and conclude that the current cellular practices on selecting serving cells should be held accountable.