Stochastic Forecasts Achieve High Throughput and Low Delay over Cellular Networks

By Keith Winstein, Anirudh Sivaraman, Hari Balakrishnan. 2013

418 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

ROBOTURK: A Crowdsourcing Platform for Robotic Skill Learning through Imitation (2018, methodology): “We used the Cellsim network simulation tool [42] to regulate the network capacity and delay between the platform and the users.”
Glimpse: Continuous, Real-Time Object Recognition on Mobile Devices (2015, methodology): “In our evaluation, we use wireless network traces described in two recent papers [59, 16], which included Wi-Fi, Verizon Wireless’s LTE, and AT&T’s LTE network.”
Copa: Practical Delay-Based Congestion Control for the Internet (2018, methodology): “Implementations: We compare the performance of our user-space implementation of Copa with Linux kernel implementations of TCP Cubic, Vegas, Reno, and BBR [8], and user-space implementations of Remy, PCC [9], PCC-Vivace [10], Sprout [44], and Verus [46].”
PCC Vivace: Online-Learning Congestion Control (2018, methodology): “An even more challenging network scenario, as suggested by [32, 36], is the LTE environment where very deep queues are accompanied by drastically changing available bandwidth in a matter of milliseconds.”
Resource Management with Deep Reinforcement Learning (2016, background): “Examples include job scheduling in compute clusters [17], bitrate adaptation in video streaming [23, 39], relay selection in Internet telephony [40], virtual machine placement in cloud computing [20, 6], congestion control [38, 37, 13], and so on.”
A Deep Reinforcement Learning Perspective on Internet Congestion Control (2019, background): “Network conditions, such as available link capacities, packet loss rates, and end-to-end latency, might be highly dynamic and change considerably over time (e.g., in mobile/cellular networks, Winstein et al. 2013).”
Neural Adaptive Video Streaming with Pensieve (2017, background): “However, selecting the right bitrate can be very challenging due to (1) the variability of network through-put [18, 42, 49, 52, 53]; (2) the conﬂicting video QoE requirements (high bitrate, minimal rebuffering, smoothness, etc.); (3) the cascading effects of bitrate decisions (e.g., selecting a…”
Mahimahi: Accurate Record-and-Replay for HTTP (2015, methodology): “We also perform evaluations over cellular networks using modified versions of the Verizon and AT&T traces collected in [40].”