Designing High Performance Enterprise Wi-Fi Networks - PowerPoint PPT Presentation

Harvard University

JitendraPadhye, Ranveer Chandra, Alec Wolman, and Brian Zill

Microsoft Research

Designing High Performance Enterprise Wi-Fi Networks

• Increased adoption and usage [Forrester]
• Culture of mobility: Users tend to use Wi-Fi even when wired connections are available [Gartner, Forrester, Economist]
• Move towards an all wireless office

Users want wire-like performance

from wireless networks

• Corporate WLAN Study:
• 12 users
• < 1 Mbps each

• Focus on coverage
• Fewer APs than clients
• Clients talk to APs far away; worsens rate anomaly
• Clients pick APs to associate with
• Use RSSI of beacon packets
• Agnostic to channel load at APs
• Lack adaptive behavior
• No load balancing; fixed channel assignments
• Congestion and hotspots worsen

• Focus on capacity
• Lots of APs; densely deployed
• Clients can talk to APs near by; mitigates rate anomaly
• Infrastructure picks client-AP associations
• Global view of network conditions (channel load, interference, etc.)
• Adaptability
• Load balance associations; Dynamic channel assignment
• Redistributes load away from local hotspots

• No client modifications
• Works with legacy clients
• Changes limited to the infrastructure
• Easy to deploy
• Self-managing

Interface with clients

Send summaries to DC

DenseAP Nodes (DAPs)

Summarized

Data

Commands

Wired Network

Commands to DenseAP nodes

Summarized Data

from DenseAP nodes

Associations

Channel Assignments

Load Balancing

DenseAP Central

Controller (DC)

• Controlling Associations
• Mechanisms
• Policy
• Dynamic Channel Assignment
• Mechanism
• Policy
• Load Balancing
• Mechanism
• Policy

Probe Request

Probe Request

00:09:5B:5A:1F:4F

Probe Request

Probe Request

MAC = 00:09:5B:5A:1F:4F

RSSI = 30

Probe Request

MAC = 00:09:5B:5A:1F:4F

RSSI = 42

00:09:5B:5A:1F:4F

Probe Request

MAC = 00:09:5B:5A:1F:4F

RSSI = 40

Probe Response

Accept Client

00:09:5B:5A:1F:4F

Client only sees one DAP at any given time

• What is the quality of a connection between a client and a DAP? (rate)
• How busyis the medium around each DAP?

Overall goal: Associate client with a DAP

that will yield good throughput

Expected

Transmission-Rate

(Mbps)

Available

Capacity (AC)

(Mbps)

Free Air Time

(%)

=

X

RSSI = 10

Free air

time = 0.22

DAP2

Probe Request

Probe Request

Probe Response

Accept Client

RSSI = 30

DAP3

Probe Request

Free air

time = 0.45

DAP1

RSSI = 20

Free air time = 0.35

• Correlation between
• RSSI of Probe Request packets
• Avg. throughput between a DAP-client pair
• Rough approximation - ordering of DAPs
• Online profiling method that builds RSSI to data-rate estimates

Upload and RSSI correlation = 0.71

Download and RSSI correlation = 0.61

• Estimate how busy is the medium around at a DAP
• Technique similar to ProbeGap*
• Measure time taken to finish a packet transmission
• Estimates match up closely with offered traffic load

*Lakshminarayan et al., 2004

*Vasudevan et al., 2005

• Integrated into the association process
• DAPs not discovered by clients don’t need channels
• A DAP is assigned a channel only when it goes from being passive (no clients)to active (services at least one client)
• Central controller assigns channel with least load

• So far, associations when a new client joins the network
• No association is perfect
• Client traffic demands change
• Local hotspots created

• Central controller monitors load on every DAP
• When channel load on a DAP crosses a certain threshold
• Client causing most load is determined
• Moved to less loaded DAP nearby
• Ensure client continues to get at least as much available capacity at the new DAP
• Load balancing achieved via handoffs
• Use association control; manipulate ACLs on DAPs

1 Corp AP

24 DAPs

24 Clients

802.11 a/bg

• Performance
• Density
• Channels
• Intelligent Association
• Load Balancing

• Gains due to
• More channels
• DAP density
• Intelligent associations

1250% gain

Why?

• Put all DAPs on the same channel
• Factors out
• Channels
• Intelligent Associations: same load on all DAPs
• Single out impact of
• Density

Higher density provides better performance

Is intelligent association control necessary?

• Client-Driven
• Disable intelligent association control
• Let clients pick DAP to associate with (conventional WLANs)
• Compare with DenseAP
• Factors out
• Channels
• Density
• Single out impact of
• Intelligent association

160% gain

Intelligent association policy is necessary

Client 1 improves

Clients 2 & 3 improve

Client 1 moved

Client 2 moved

• Load balancing algorithm and mechanism
• Mobility
• Performance
• Fewer DAPs
• Fewer channels
• 802.11g
• …..
• Scalability

• Plenty of prior work on static channel assignment, power control and associations
• Each studied each aspect in isolation
• Require client modifications [Ramani and Savage, Infocom 2005]
• SMARTA [Ahmed et al., CoNext 2006]
• Examines channel and power control
• Increase overall network capacity
• Does not consider associations, load balancing
• MDG [Broustis et al., MOBICOM 2007]
• Identified tuning channel, power and associations
• Studies the order in which these knobs must be tuned
• Requires client modifications

• Practical system
• How do density, intelligent association, and more channels affect capacity?
• Adaptive system
• Future directions
• Impact of hidden terminals
• Heterogeneous mix of client traffic patterns
• Other backhauls: e.g. Wireless, powerline