UWB Propagation Phenomena

1
Project IEEE P802.15 Working Group for Wireless
Personal Area Networks (WPANs) Submission
Title Status report of the 802.15.4a channel
modeling subgroup Date Submitted 8 January,
2004 Source Andreas F. Molisch Company
Mitsubishi Electric Address 201 Broadway,
Cambridge, MA Voice 617 621 7558 E-Mail
Andreas.Molisch_at_ieee.org Re Contributions on
UWB Channel Models Abstract This contribution
describes the status of the IEEE 802.15.4a
channel modeling subgroup, the currently
envisioned generic model structure, and an
outlook for future work Purpose Notice This
document has been prepared to assist the IEEE
P802.15. It is offered as a basis for discussion
and is not binding on the contributing
individual(s) or organization(s). The material in
this document is subject to change in form and
content after further study. The contributor(s)
reserve(s) the right to add, amend or withdraw
material contained herein. Release The
contributor acknowledges and accepts that this
contribution becomes the property of IEEE and may
be made publicly available by P802.15.
2
Contents

• Background
• Radio Environments
• Generic Model Structure
• Future workplan
• Interaction with SG4a plenary

3
Background

• Subgroup established in Singapore (Sept. 2003)
• Mandate establish a channel model for system
  evaluation of a 802.15.4a standard
• Timeline finish by May 2004
• Considered bandwidth 2-10 GHz
• Model has to work for UWB as well as non-UWB
  proposals that fulfill the CFA requirements
• UWB will be among the proposals
• Any UWB model can be downfiltered to a narrowband
  system
• 2.45 GHz ISM band should be included in the
  considered range
• Working mode biweekly teleconferences
• Very informal approach (within the rules of
  IEEE) participants always welcome

4
Radio Environments

• Follow from CFA (Nov. 2003, Albuquerque)
• List of 10 radio environments established by
  subgroup
• 1. Indoor office
• 2. Indoor residential
• 3. Indoor industrial
• 4. Indoor open spaces
• 5. Warehouses
• 6. Body-worn devices
• 7. Out door hand held peer to peer device
• 8. Hand held communicating to fixed location
  devices
• 9. Agricultural areas/farms
• 10. Sport stadiums
• 11. Disaster areas (houses filled with rubble,
  avalanches, .)

5
Generic model structure

• Why not reuse 802.15.3a? (i) operating
  environments are different, (ii) longer range,
  (iii) more measurements are now available in the
  literature
• First step time-continuous model that is valid
  within the considered bandwidth
• Explicit discretization rules will follow later,
  together with a possible MATLAB implementation
• Basic structure (tentative, status Jan. 8th)
• Saleh-Valenzuela model multiple clusters, with
  discrete multipaths in each of them, plus a
  diffuse component
• Number of clusters is assumed fixed (no random
  variable)
• Movement not modeled explicitly, but rather via
  angular spectrum. That allows separation of
  effects due to movement of MS and movement of
  scatterers

6
Generic model structure (contd.)

• Parameterization of
• Pathloss and shadowing (including frequency
  dependence)
• Arrival rates of multipath components (delay
  dependent?)
• Average power delay profiles (per-cluster delay
  spreads)
• Average angular spectra (including dependence on
  delay)
• Small-scale fading statistics (Rayleigh,
  lognormal, or Nakagami)

7
Future workplan

• Finalization of generic model structure end of
  Feb. 2004
• Measurement campaigns to establish parameters
  Dec. 2003-April 2004
• Parameter extraction March-May 2004
• Presentation of model (2004)
• Writing of MATLAB program for general use
  (optional) May-June 2004

8
Interaction with SG4a plenary

• Minutes of teleconferences are sent out to SG4a
  reflector
• Status reports at every 4a meeting
• We need more input from SG4a!!!
• Without the model, no proper selection of
  candidate PHY layers can be made
• Any radio environment for which we do not have
  sufficient data will be eliminated from the list.
  In that case, the suitability of the PHY layer
  for a specific application cannot be tested
• At the moment, most work is being done by just a
  few participants. We need a fair distribution of
  the workload.
• We have a unique chance to get the channel model,
  and thus the basis for 4a, to a good start. Lets
  not waste that chance!