improving wireless network performance using sensor...

Improving Wireless Network Performance Using Sensor Hints

Lenin Ravindranath,Calvin Newport, Hari Balakrishnan, Sam Madden MIT Computer Science and Artificial Intelligence Laboratory

NSDI'11 Proceedings of the 8th USENIX conference on Networked systems design and implementation 2011

Outline •Introduction

•Architecture

•Hint-aware bit rate adaptation

•Access point (AP) association

•Mobile topology maintenance

•Path selection in vehicular mesh networks

•Conclusion

Introduction(1/2) •Big Changes in Access Devices

297M smartphones sold worldwide in 2010

31% of US phone market;

50% by this year Smartphones and tablets exceeding PC sales

Introduction(2/2) •The Problem • Protocols need to adapt to different settings

-Mobility mode impacts wireless performance

• Most protocols optimized for static settings

-They perform poorly during mobility

Accelerometer

Proximity Sensor Camera

Ambient Light Sensor Microphone

GPS

Compass

Gyro

Architecture(1/11)

Application

Transport

Network

MAC

PHY

Wireless Radio

Wireless Protocol Stack

Accelerometer

Proximity Sensor Camera

Ambient Light Sensor Microphone

GPS

Compass

Gyro

Ignored by Protocols!

Architecture(2/11)

Application

Transport

Network

MAC

PHY

Wireless Radio

Wireless Protocol Stack

Accelerometer

GPS

Compass

Gyro

Architecture(3/11)

Application

Transport

Network

MAC

PHY

Wireless Radio

Wireless Protocol Stack

GPS

Compass

Accl

Gyro

Hints

• Movement

• Direction

• Speed

Use hints to adapt to different mobility

modes

Architecture(4/11)

Application

Transport

Network

MAC

PHY

Wireless Radio

Wireless Protocol Stack

GPS

Compass

Accl

Gyro

Hints

• Movement

• Direction

• Speed

Use hints to adapt to different mobility

modes differently

Architecture(5/11)

Application

Transport

Network

MAC

PHY

Wireless Radio

Wireless Protocol Stack

GPS

Compass

Accl

Gyro Rate Adaptation Movement

Heading AP Association

Speed

Vehicular Routing

Walking

Architecture(6/11)

Application

Transport

Network

MAC

PHY

Wireless Radio

Wireless Protocol Stack

GPS

Compass

Accl

Gyro Rate Adaptation Movement

Heading AP Association

Speed

Vehicular Routing

Walking

Topo Maintenance

Architecture(7/11)

Application

Transport

Network

MAC

PHY

Wireless Radio

Wireless Protocol Stack

GPS

Compass

Accl

Gyro Rate Adaptation Movement

Heading AP Association

Speed

Topo Maintenance

Packet Scheduling

Power Saving

Adapt Cyclic Prefix

Network Monitoring

Speed

Walking

Location Vehicular Routing

Architecture(8/11)

Application

Transport

Network

MAC

PHY

Wireless Radio

Wireless Protocol Stack

GPS

Compass

Accl

Gyro Rate Adaptation Movement

Heading AP Association

Speed

Vehicular Routing

Walking

Architecture(9/11)

Accl

Movement

Is the device static or moving?

Architecture(10/11)

GPS

Compass

Accl

Gyro Movement

Heading

Speed

Walking

Walking Hint

Accelerometer

Heading

Outdoor - GPS

Indoor – Compass + Gyro + Accelerometer

Speed

Outdoor - GPS

Indoor – Accelerometer

Architecture(11/11)

6 Mbps 9 Mbps 12 Mbps 18 Mbps 24 Mbps 36 Mbps 48 Mbps 54 Mbps

802.11a bit rates

Packet encoded at a particular bit rate

Rate Adaptation: Finding the best bit rate to transmit a packet

Hint-aware bit rate adaptation (1/8)

RapidSample

1. When a packet fails, probability that the next few packets at the same bit rate will fail is high

Immediately reduce bit rate on packet loss

2. Coherence time of the channel is a few ms

Never retry any higher rate for this period of

time

Hint-aware bit rate adaptation (2/8)

3. If the channel is not degrading

After a few successes at the current bit rate, sample higher rates that have not recently failed (in the last few milliseconds)

If we are wrong about the channel improving and the sampled higher rate fails, revert to the original rate

Hint-aware bit rate adaptation (3/8)

6 Mbps 9 Mbps 12 Mbps 18 Mbps 24 Mbps 36 Mbps 48 Mbps 54 Mbps

6 Mbps 9 Mbps 12 Mbps 18 Mbps 24 Mbps 36 Mbps 48 Mbps 54 Mbps

6 Mbps 9 Mbps 12 Mbps 18 Mbps 24 Mbps 36 Mbps 48 Mbps 54 Mbps

1. After a single loss Reduce rate

2. Short history Don’t retry a any higher rate

3. Sender attempts to sample a higher bit rate After few successes, sample

higher rate not failed

If incorrect, come back to the original rate

[failed – within last 10ms]

Hint-aware bit rate adaptation (4/8) RapidSample

Implementation and Evaluation

Linux (Click) Android

Movement hint

SampleRate RRAA RapidSample Hint-Aware

1000 byte packets (at a bit rate)

ACK

Hint-aware bit rate adaptation (5/8)

when device is moving

Hint-aware bit rate adaptation (6/8)

when static…

Hint-aware bit rate adaptation (7/8)

Hint-aware bit rate adaptation (8/8)

RapidSample when moving SampleRate when static

1. Maximize throughput

File download

2. Minimize handoffs (scans)

VOIP – minimize disruptions

AP association(1/3)

1. Static – Stop Scanning 2. Walking – Scan Periodically

3. Walking to Static – Scan once

Maximize throughput

AP association(2/3)

Heading

Minimize Handoff

Training-based approach

Background Android application

Training: WiFi scans + Heading hint

Maps:<heading, AP>

AP association(3/3)

How frequently should nodes probe?

Probes

Mobile topology maintenance(1/3)

Static vs. Mobile

Mobile case requires more probes to maintain acceptable estimation error

Mobile topology maintenance(2/3)

Adaptive Probing Protocol • Adapt probing based on movement hints

• When a node is static – Probe infrequently (1 probe every 2 seconds)

• When a node is mobile – Probe frequently (10 probes every 1 seconds)

Mobile topology maintenance(3/3)

“V2V”

Path selection in vehicular mesh networks(1/2)

• Longevity of links useful – avoids expensive repairs

• Connection Time Estimate (CTE) – Use heading and speed to predict connection duration

– Link between nodes heading in the similar direction tend to last longer

– Speed is inversely correlated to connection duration

Path selection in vehicular mesh networks(2/2)

Large difference in headings predicts short-lived link

Small difference in headings predicts long-lived link

Conclusion This paper introduced a network architecture that uses

sensor hints to augment and improve wireless protocols.

The key idea is to use these hints to infer the context in

which communication is occurring, and to use that con-

text to adapt the behavior of protocols.