Examination of reasons why someone might vote MB-OFDM

1
Project IEEE P802.15 Working Group for Wireless
Personal Area Networks (WPANs) Submission
Title Reasons to vote for the MB-OFDM
proposal Date Submitted July, 2004 Source
Michael Mc Laughlin Company decaWave
Ltd Address 8133 Leesburg Pike Vienna, VA
USA Voice703-269-3000, E-Mailmichael_at_decawav
e.com Re Abstract Examines the five most
popular reasons for voting for the MB-OFDM PHY
proposal and finds that they are actually reasons
to vote for the DS-UWB proposal. Purpose Provid
e technical information to the TG3a voters
regarding PHY proposals. 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
Summary of reasons why someone might vote for
MB-OFDM instead of DS-UWB

• 5. Complexity MB-OFDM lt 4 x DS-UWB.
• 4. Power MB-OFDM lt 4 x DS-UWB.
• 3. Range MB-OFDM almost as good as DS-UWB on
  many channels.
• 2. OFDM previously chosen for other, different
  modulation schemes.
• 1. MB-OFDM is backed by TI / Intel /Sony
  /Philips and others.

3
Reason 5 Complexity

• MB-OFDM Digital Complexity less than 4 times
  DS-UWB Digital Complexity
• Digital complexity of MB-OFDM is between 2 and 4
  times that of DS-UWB depending on bit rate
• True, but then why not choose lower complexity
  proposal

4
Reason 5 Complexity
Component MB-OFDM (Doc 03/268r3) DS-UWB 32-Finger Rake Architecture
Matched filter rake or FFT 100K 45K
Viterbi decoder 108K 54K
Synchronization 247K 30K
Channel estimation 247K 24K
Other Miscellaneous including RAM 247K 30K
Equalizer Freq Domain 20K
Total gates _at_ 85.5 MHz 455K 203K
5
Reason 4 Power Consumption

• MB-OFDM Power less than 4 times DS-UWB Power
• Digital power consumption at a given speed is
  proportional to the number of gates
• MB-OFDM proposal is 2 to 4 times digital
  complexity complexity for same speed gt digital
  power consumption of MB-OFDM is between 2 and 4
  times that of DS-UWB.
• True, so why not vote for lower power proposal

6
Reason 4 Power Consumption
Channel Model DS-UWB 220Mbps MB-OFDM 200Mbps DS-UWB 500Mbps MB-OFDM 480Mbps
Rx Digital Power 130nm 51mW 106mW 57mW 202mW
7
Reason 3 Range

• MB-OFDM range is almost as good as DS-UWB on many
  channels
• Almost is not good enough.
• It is true that MB-OFDM ranges come fairly close
  to DS-UWB for the 110Mbps mode, but as conditions
  get worse and as bit rates rise, the DS-UWB
  advantage increases.
• e.g. DS-UWB outstrips MB-OFDM by more than 60
  for the 220/200Mbps over CM4.

8
Reason 3 Range
Channel Model DS-UWB 110Mbps 10 Outage Range MB-OFDM 110Mbps 10 Outage Range DS-UWB 220Mbps 10 Outage Range MB-OFDM 200Mbps 10 Outage Range
CM1 13.5 11.4 8.4 6.9
CM2 11.7 10.7 7.2 6.3
CM3 11.4 11.5 7.0 6.8
CM4 10.8 10.9 7.1 4.7
9
Reason 2 OFDM used in other standards

• OFDM modulation has previously been chosen for
  other different modulation schemes.
• True, e.g. ADSL and 802.11 but the key word here
  is different.
• In ADSL the fixed channel allows the number of
  bits per tone to be varied according to SNR for
  that part of the channel. This is not currently
  possible for wireless systems.
• In 802.11a/g the SNR assumed is very high
  compared to UWB and the bandwidth is much lower.
• For both of these reasons, the rayleigh fading
  effects are far less damaging.

And its a big but
10
Reason 1 Many large companies support MB-OFDM

• Giants like Intel/Philips etc. wouldnt pick an
  inferior scheme, right?
• Wrong! They already admitted this when they
  abandoned their original Multiband proposal in
  favour of TIs MB-OFDM proposal

11
Summary

• The five top reasons for voting for MB-OFDM have
  examined and found to actually be reasons to vote
  for DS-UWB
• This is without even considering the obvious
  reasons not to vote for MB-OFDM (Interference,
  SOP performance, Time to Market, Scalability)
• There are no good technical reasons to vote for
  the MB-OFDM proposal.