Evaluation Methodology for 802.20 MBWA

1
(No Transcript)
2
802.20 Evaluation Criteria and Traffic Models
Status UpdateUpdated per Conference call Dec. 7,
2004By Anna Tee

• Farooq Khan
• IEEE 802.20 Plenary Meeting
• San Antonio, Texas, USA 
• November 15-19, 2004

All changes show in Green.
3
Evaluation Criteria Status

• Four conference calls since September Interim
• September 28, 2004 Phased Approach
• October 12, 2004 Evaluation Criteria Document
  Review
• October 26, 2004 Document Review Channel
  Models
• November 9, 2004 Document Review Simulation
  Calibration
• Major open issues
• VoIP (and wireless multi-party Gaming 04/86)
  Traffic model
• System simulation calibration 04/83r1
• Channel 04/82r1 and Traffic mix
• Details of Phase 2 simulations 04/85r1
• Updated 802.20 Evaluation Criteria Version 12 now
  available
• Included Phased Approach Table
• Updates based on the document review over the
  conference calls
• Number of occurrences of the word TBD in the
  document 22 (18 of them located in Phased (2)
  approach Table)
• Reference to related IEEE 802.20 contributions
  included in .

4
Traffic Models

• Specification of traffic mix
• Phase 1 use full buffers model
• Traffic mix scenarios need to be defined for
  Phase 2 of the simulations
• VoIP Traffic Model
• Need to finalize on VoIP source traffic model
• Contributions invited on Wireless multi-party
  Gaming traffic models 04/86
• Video Streaming Model
• Need to determine if video streaming data rate
  need to be different than 32Kb/s currently
  assumed. 04/88

5
Phase 2 Simulations Details 04/85r1

• The details of phase 1 are currently being
  discussed in the evaluation criteria
• Agreed to use 19-cells 3-sector wrap-around
  configuration, Full buffers (hungry) traffic,
  simulation calibration, link-system interface
  etc.
• Current Recommendation is to use suburban macro,
  3 Km/h pedestrian B and 120Km/h Vehicular B
  channel models.
• The issues that need further consideration
• Full-duplex simulation, traffic mix, channel mix,
  control signaling and handoff modeling etc.

6
Link Budget Criteria

• Consensus on most of the link budget parameters
• Open issue Should maximum range (link budget) or
  equivalently maximum pathloss be used as a
  performance metric for proposal comparison or
  not? 04/64r4
• Clarification text on the section for link budget
  required so that technology proponents will know
  how to provide the data requested (How the data
  will be used will be covered in the Technology
  Selection Process document.)
• Performance metrics specified in section 13 can
  refer to the link budget template

7
Application specific criteria
In the evaluation of spectral efficiency and in
order to make a fair comparison of different
proposals, it is important that all mobile users
be provided with a minimal level of throughput.
The fairness for best effort traffic (HTTP, FTP
and full buffers) is evaluated by determining the
normalized cumulative distribution function (CDF)
of the user throughput, which meets a
predetermined function. For applications other
than best effort, application specific outage
criteria are defined. The proposals will also
provide additional fairness metrics. The details
of the additional fairness metrics are TBD (see
for example IEEE C802.20-04/05).

• A fairness criteria is defined for the best
  effort data traffic
• application specific outage and QoS (FER, delay
  etc.) criteria need to be defined for other
  applications!
• Contributions are also invited on additional
  fairness metrics
• For other applications such as VoIP, gaming or
  video streaming, what would be the criteria in
  each case to ensure the spectral efficiency is
  computed based on system resources being shared
  fairly amongst the simulated users in the same
  sector?
• Contributions are required if additional metrics
  are to be adopted

8
System simulation calibration 04/83r1

• The evaluation criteria would specify a system
  simulation calibration process.
• Calibration would be done as part of phase 1 of
  simulations
• However, it is not clear, at this stage, to what
  level of detail simulations need to be
  calibrated.
• The group discussed a contribution on this issue
  over the November 9, 2004 conference call
• Further discussions planned during the Plenary
  meeting

9
Channel Models Mix 04/82r1

• Decided to address the Channel models mix issue
  in evaluation criteria.
• Need to decide if Channel Models mix is necessary
  for evaluation
• If it is necessary, what would be the appropriate
  channel models mix?
• Further discussions planned based on an open
  contribution discussed over the November 8, 2004
  channel models CG call.