Cooperative MIMO in Wireless Networks: Where are we? - PowerPoint PPT Presentation

1 / 16
About This Presentation
Title:

Cooperative MIMO in Wireless Networks: Where are we?

Description:

Emulate an antenna array -- use of the antenna elements of different ... Jumped on the bandwagon! Assumed that PHY. Layer benefits are. directly applicable at ... – PowerPoint PPT presentation

Number of Views:632
Avg rating:3.0/5.0
Slides: 17
Provided by: srikanthkr
Learn more at: http://anrg.usc.edu
Category:

less

Transcript and Presenter's Notes

Title: Cooperative MIMO in Wireless Networks: Where are we?


1
Cooperative MIMO in Wireless Networks Where are
we?
Srikanth Krishnamurthy
2
What is cooperation?
  • Nodes cooperate to jointly transmit information.
  • Instead of contention.
  • Emulate an antenna array -- use of the antenna
    elements of different nodes jointly.
  • Possible use of Space Time Block Codes
  • PHY layer studies suggest that there is a
    diversity gain
  • Diversity gain --gt reduced SNR requirement for a
    specific BER requirement.

3
What did we do ?
  • Jumped on the bandwagon!
  • Assumed that PHY
  • Layer benefits are
  • directly applicable at
  • higher layers

4
Assumptions
  • The lowered SNR requirement can help in either
  • Increased range
  • Increased rate
  • In our work, we seek to exploit the increase in
    range.

5
Our work in a nutshell
  • Design of MAC layer extensions to the 802.11 DCF
    to enable cooperation.
  • Exchange of channel state information.
  • Intelligent selection of neighbors.
  • Design of an tightly coupled routing scheme to
    exploit increased range.
  • Compute SISO route.
  • Bridge this route with longer cooperative MISO
    hops.

6
The story in pictures
  • Paper in INFOCOM 2006

7
A closer look
  • Do we really get these gains ?
  • Or is this an artifact of the models ?

8
Experiments with 802.11n
  • Use of STBC

MIMO STBC
SISO
  • Not much to be gained at lower rates !!!

9
To summarize
  • At the very least, in commercial off the shelf
    hardware, the gains from MIMO (and possibly with
    cooperative MIMO) are not as one might expect!

10
Where did we go wrong ?
  • Long story -- but in short -- PHY layer issues
    need to be well understood.

11
A list of issues
  • Packet error rate does not linearly change with
    BER!
  • Bursty errors can make PHY layer models less
    accurate.
  • Interactions with FEC codes and OFDMA
  • We cannot consider one PHY layer technology in
    isolation if we want to determine what happens in
    practice.

12
But there is hope
  • MIMO seems to help in coping with far away
    interferers.
  • Increased connectivity true to some extent
  • higher rates.
  • Can help in power savings.

13
Need to rethink design
  • Simultaneous transmissions due to capture are
    possible
  • However these are precluded by 802.11 DCF.
  • Can we somehow enable this ?
  • Facilitate cooperation
  • Tune CCA ? -- there are challenges.
  • Need to carefully understand when benefits are
    possible.

14
The Road Ahead
  • Slippery but driving can be fun!
  • Need to understand PHY implications on higher
    layers.

15
  • Thank you

16
Model
  • Assume that nodes use cooperation to jointly
    transmit data.
  • Use of STBC codes.
  • We calculate the additional distance that a node
    can communicate given a fixed value of the
    diversity gain due to cooperative MIMO.
  • Diversity gain was obtained from SNR vs BER
    graphs from PHY layer studies on MIMO
  • Turns out that this gain is significant.
Write a Comment
User Comments (0)
About PowerShow.com