Wireless Performance Prediction – Organization Proposal

1
Wireless Performance Prediction Organization
Proposal

• Presented by
• Paul C. Canaan Intel Corporation
• March 11, 2003

2
Proposed Purpose and Scope

• Purpose
• Define the key metrics that impact the
  performance of a wireless ecosystem.
• Scope
• Provide a list of variables that impact the
  performance of 802.11 wireless devices.
• Outline a mechanism in which the performance of
  wireless device can be predicted within a given
  margin of error
• Draft a set of guidelines for persons involved in
  the design of wireless networking components as
  well as the installation of wireless networks to
  use as a means to optimize performance

3
Wireless Ecosystems

• The wireless ecosystem represents the
  environments in which the wireless network
  resides. The ecosystems will be evaluated using a
  three tier approach
• Simple
• Intermediate
• Advanced

Simple
Intermediate
Advanced
4
Performance in the Simple Wireless Ecosystem

• Variables that have a significant impact on
  performance will need to be defined between a
  single access point and client
• These variables have to be measured so latter
  predication methodologies can be devised

Simple
Signal Strength
Antenna
Protocol
Packet Error Ratio
Multi-path
5
Issues with Predicting Wireless Performance

• The performance experience of wireless systems is
  impacted by additional variables such as the test
  environment, access point position, antenna
  placement and other devices
• Each test environment is unique and results
  obtained in one environment may not correlate to
  another.
• Environmental factors localized to a specific
  location make it difficult to fully assess the
  test results
• No known standard industry test methodology or
  best known methods exist for testing of
  wireless systems. Thus making it difficult to
  replicate results

6
Questions to Ask When Developing Guidelines For
Predicting Wireless Experience

• Correlation
• What can be measured
• What variables count
• What equipment is needed
• Ensure results are repeatable
• Document Guidelines
• Wireless User Experience
• Determine which experiences to focus on
  (business, consumer)
• Determine what kind of usage models are
  associated with those experiences
• Determine what can be measured
• Determine how to repeat results using correlation
  guidelines
• Document Guidelines for a particular wireless
  user experience

These are questions that will have to be answered
before we can predict wireless performance