Rebera: Real-time Bandwidth Estimation and Rate Adaptation...

Rebera: Real-time Bandwidth Estimation andRate Adaptation for Video Calls

over Cellular Networks

E. Kurdoglu, Y. Liu, Y. Wang 1

Y. Shi, C. Gu, J. Lyu 2

1NYU Tandon School of Engineering

2Tencent Co. Ltd.

May 11, 2016

Kurdoglu, Liu, Wang, Shi, Gu, Lyu NYU Tandon School of Engineering, Tencent Co. Ltd.

Rebera: Real-time Bandwidth Estimation and Rate Adaptation for Video Calls over Cellular Networks

Outline

1 Introduction and Problem Setting

2 Rebera ComponentsMeasuring the Current BandwidthPredicting the Future BandwidthDetermining the Sending Rate (Bit-budget)Encoder & Video Stream StructureDynamic Frame Selection

3 Experiment Results

Video calls: very low-delay, interactive applications

1 Need to adapt transmission rate wrt network speed:Tx rate > Bandwidth ⇒ intolerable delaysTx rate < Bandwidth ⇒ under-utilized link

Accurate measurement and forecast of bandwidth is hard!

2 Need to adapt video encoding rate wrt to network speed

Accurate low-delay video rate control is hard!

3 Need resiliency against late or lost packets

Cellular links can be highly volatile: link speeds, delays...

A Video Call Scenario

ISP1 Internet ISP2

Challenge: measuring and predicting the e2e bandwidth

Assume: cellular links are the only bottlenecks

Model: cellular links as isolated queues (no cross-traffic)

uplink buffer in the device, downlink buffer in the BS

Outline

Rebera Overview

PacketTransmission

H.264/AVCHierP

CapacityPrediction

PacketReception

CapacityMeasurement

Decode &Display

sender

receiver

We periodically insert I-frames

Intra-period: time between two successive I-frames (≈1 sec)

Rate control performed at the beginning of each intra-period

Measuring the Current Bandwidth

Observation: We can measure bandwidth by sending packet trainsand measuring packet inter-arrival times at the receiver.

a2 a3 ap

time⇒ mn ,

z2 + · · ·+ zpa2 + · · ·+ ap

Key idea: Use the packets belonging to a frame as a packet train.

1 Estimate avg. ABW in thelast intra-period

2 Feed it back to sender

3 Slide the window

sliding window

Predicting the Future Bandwidth

Observation: Bandwidth measurement values are correlated.

Key idea: Forecast future bandwidth from past values with onlinelinear adaptive filtering (e.g. RLS)

time(sec)200 250 300 350 400 450 500

1500MeasurementPrediction (RLS)

ck+1 =M−1∑i=0

wi (k)ck−i

M number of filter tapsw(k) filter tap vector of length Mck measured capacity

Determining the Sending Rate (≡ Bit-budget)

Observation: Forecast errors are inevitable.

Key idea: Estimate the (almost) worst-case scenario to find the(almost) safest max sending rate.

rk sending rateεk exact-to-predicted ratio (= ck

qk backlogged bits

Pr(rk+1 > ck+1) ≤ δ ⇒ Pr

(εk+1 <

)≤ δ

r safek+1 = ck+1 × δ-quantile of ε

Bit-budget for intra-period k + 1

bk+1 = r safek+1 × T − qk

Determining the Sending Rate (≡ Bit-budget)

Observation: Forecast errors are inevitable.

Key idea: Estimate the (almost) worst-case scenario to find the(almost) safest max sending rate.

rk sending rateεk exact-to-predicted ratio (= ck

qk backlogged bits

Pr(rk+1 > ck+1) ≤ δ ⇒ Pr

(εk+1 <

)≤ δ

r safek+1 = ck+1 × δ-quantile of ε

Bit-budget for intra-period k + 1

bk+1 = r safek+1 × T − qk

Encoder & Video Stream Structure

Observation: Resiliency against packet losses is essential.

Key idea: Use hierarchical-P temporal layering.

GoP GoP

Intra-period1 Enables UEP X

2 Complexity & overheadkept minimum X

3 DFS does not causejitter. X

Dynamic Frame Selection for Hierarchical-P Video

Observation: When video rate > target bitrate, we can discardhigher layer frames and maintain a stable frame rate.

Requirement: Must make the “send/discard” decision on the fly.

Decision for frame P1?

Cannot wait for {P2,P4,P6}

GoP GoP

Intra-period

Sequential decision makingalgorithm considers:

1 Qualities of layers

2 Decoding dependencies

3 Smoothness of frameintervals

Outline

Our Testbed

NYU Network

eth0eth1

Implemented in C++ (platform: Linux)Encoder: x264, modified to generate hierarchical-P w/ 3 TLsDecoder: FFmpegCellular links emulated by CellSim

FaceTime used as a benchmark

Performance metrics: ABW utilization, 95-percentile packet &frame delays

Evaluation under Cellular Bandwidth Traces

0 200 400 600 800 1000

0.5 Tr1

0 200 400 600 800 1000

0.5 Tr2

0 200 400 600 800 1000

012 Tr4

0 200 400 600 800 1000

time (sec)0 200 400 600 800 1000

ABW utilization percentage10 20 30 40 50 60 70

lays (

A-BA-C B-A

B-CC-A

att-tmobile

att-verizon3g

att-verizon4g

tmobile-att

tmobile-verizon3g

tmobile-verizon4g

verizon3g-att

verizon3g-tmobile

verizon4g-att

verizon4g-tmobile

A-BA-CB-A

att-tmobile

att-verizon3g att-verizon4g

tmobile-att

tmobile-verizon3g

tmobile-verizon4g

verizon3g-att

verizon3g-tmobile

verizon4g-att

verizon4g-tmobile

Double-link scenarios

FaceTimeRebera

Rebera avg. util.: 55%, avg. 95-percentile frame delay: 865 msecFaceTime avg. util.: 32%, avg. 95-percentile frame delay: 2.8 sec

ABW utilization percentage10 20 30 40 50 60 70

lays (

B-C C-AC-B

att-tmobile

att-verizon3g

att-verizon4g

tmobile-att

tmobile-verizon3g

tmobile-verizon4g

verizon3g-att

verizon3g-tmobile

verizon4g-att

verizon4g-tmobile

A-BA-C

B-A B-C

att-tmobileatt-verizon3g

att-verizon4gtmobile-att

tmobile-verizon3g

tmobile-verizon4g

verizon3g-att

verizon3g-tmobile

verizon4g-attverizon4g-tmobile

Double-link scenarios

FaceTimeRebera

Rebera avg. util.: 55%, avg. mean frame delay: 295 msecFaceTime avg. util.: 32%, avg. mean frame delay: 620 msec

sec0 100 200 300 400 500 600 700 800

2000verizon3g-att

CapacityFaceTimeRebera

sec0 100 200 300 400 500 600 700 800 900 1000

2000verizon3g-tmobile

sec0 100 200 300 400 500 600 700 800

2000att-verizon4g

sec0 100 200 300 400 500 600 700 800

2000tmobile-att

Conclusions

Proposed a real-time bandwidth estimation for video calls→ can be integrated with existing CC mechanisms

Sequential linear prediction allows forecasting→ can be further improved: ML, etc...

Hierarchical-P structure enables easy rate control and someresiliency

Proposed method integrated into last version of WeChat

Thanks!

Find the code and reproduce our work at

http://vision.poly.edu/

Rebera: Real-time Bandwidth Estimation and Rate Adaptation...

Documents

Transcript of Rebera: Real-time Bandwidth Estimation and Rate Adaptation...

I/O Emulated UART Baud Rate Calibration Application Note · I/O Emulated UART Baud Rate Calibration Application Note . AN0475E 2 / 13 January 11, 2018 . START Bit 0 Bit 1 Bit 2 Bit

ECU with emulated partial networking functionality · iCC 2013 CAN in Automation 05-6 ECU with emulated partial networking functionality An alternative approach to ISO 11898-6 CAN

Implementation Validation of Cellsim Haifa07...Implementation and validation of a Cell simulator (Cellsim) using UNISIM Alejandro Rico1, Felipe Cabarcas1,3, David Ródenas1,2 Xavier

Error Detection and Correction in SRAM Emulated TCAMs

APPLICATION LAYER SYSTEM DESIGN FOR REAL-TIME VIDEO ...vision.poly.edu/index.html/uploads/eymen_thesis_v6.pdf · REAL-TIME VIDEO COMMUNICATIONS DISSERTATION Submitted in Partial Ful

PLoS - Why It is a Model to be Emulated

CellSIM OS Overview 1.0

Implementing an emulated UART on STM32F4 microcontrollers · Implementing an emulated UART on STM32F4 microcontrollers Introduction This application note describes how to implement

Wavefront shaping through emulated curved space in ...slab.nju.edu.cn/upload/uploadify/20160512/Wavefrontshapingthrough... · Wavefront shaping through emulated curved space in waveguide

An Automated System for Emulated Network Experimentationconferences.sigcomm.org/co-next/2013/program/p235.pdf · An Automated System for Emulated Network Experimentation Simon Knight

An Automated System for Emulated Network Experimentation

Robust Vehicle Tracking for Urban Trafﬁc Videos at ...vision.poly.edu/index.html/uploads/robust_tracking_final.pdfRobust Vehicle Tracking for Urban Trafﬁc Videos at Intersections

IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL …vision.poly.edu/papers/before_2010/HoiShutoy2007STSP.pdf · UEP through channel coding, without cooperation), coopera-tive transmission

Implementing an emulated UART on STM32F4 microcontrollers · August 2016 DocID026046 Rev 2 1/24 1 AN4457 Application note Implementing an emulated UART on STM32F4 microcontrollers

Emulated EEPROM Function for Data storage (ref. 908LJ12)

Virtual Peripheral Interfaces in Emulated Embedded ...

Emulated Services over PSN – IPmux, Gmux

Improved clustering of spike patterns through video …vision.poly.edu/papers/2013/BugraAkyildiz2013SPMB.pdf · 2015-07-15 · Improved clustering of spike patterns through video

Basics of Video Courtesy of Professor Yao Wang Polytechnic University, Brooklyn, NY11201 yao@vision.poly.edu.

Storage/Emulated/0/DCIM/HP Connected Photo/Edited Photos/IMG_201502185048.Jpg