Amazon Technologies, Inc.

Satheesh Ramalingam

Rahul Urgaonkar

Mohan Padmanabhan

Yongjun Wu

Amarsingh Buckthasingh Winston

Manish Rao

Moping Dou

Hargun Kaur

Adhvik Arun Shetty

Yang Yang

Alon Dulce

Zahaib Akhtar

Kevin Matthew McCaw

Viet Anh To

Tianyu Chen

Adaptive bitrate techniques are used to select bitrates of segments to download when streaming video content. Different bitrate paths may be evaluated to optimize an objection function. Filtering bitrate paths that are unlikely to be optimal increases the computational speed, facilitating deployment.

Adaptive video streaming algorithm for optimized model predictive control

Systems, methods, and computer-readable media are disclosed for optimizing network performance on a computer device to improve quality of experience by determining which network settings on the computing device to adjust. A clustering algorithm may identify various classes of networks and a classification algorithm may determine a network class specific to a network on a computing device. The effects of certain network settings for that networks class may be determined and the network setting and/or settings that optimizes the network performance may be promoted. The system may periodically analyze network data to recalculate the appropriate networks class and may determine different network settings based on the recalculation, facilitating mid-session improvements to the quality of experience.

Machine learning-based playback optimization using network-wide heuristics

Systems and methods for optimizing network settings

Systems, methods, and computer storage media are described herein for optimizing adaptive bitrate streaming for content (e.g., streams of live events). Bitrate selections may be made based at least in part on the current state of the playback. When the playback is in a transient state (e.g., when the user has recently played or navigated within the content), bitrate selections may be determined utilizing a virtual buffer that simulates a playback buffer. The virtual buffer may be initialized with a predetermined fullness value. As content fragments are received and queued on the playback buffer, the fullness values for the virtual buffer and the playback buffer may be adjusted differently. When the fullness of the playback buffer exceeds the virtual buffer, subsequent bitrate selections may be determined utilizing the playback buffer. The virtual buffer may be utilized to simulate a steady state playback buffer while the playback buffer populates.

Techniques for short buffer adaptive bitrate streaming

Systems, methods, and computer-readable media are disclosed for determining and applying a timeout for streaming video. Processing circuitry of a device may request video fragments to be downloaded and played, and if a requested fragment is not downloaded entirely before expiration of a timeout, the device may adjust the timeout to allow the download to complete. A timeout may be set based on available bandwidth and a size of a fragment requested for download, and the timeout may be extended based on network and/or device conditions.

Timeout optimization for streaming video

Growing community of inventors

Sammamish, WA, United States of America

Satheesh Ramalingam