Re: about Space time coding

On 30 Mar 2006 22:48:50 -0800, "sairamesh@xxxxxxxxx"
<nammisairamesh@xxxxxxxxx> wrote:

Yes, Alamouti paper is very simple to understand. Which is valid for 2
transmit antenna. Tarokh etc extended that concept to multiple transmit
antennas.


But, one thing I did not convinced is
"How the real channel is Quasi static for the length of whole space
time block code word."
Is this assumption is really valid in all the cases?
Thanks,
-SaiRamesh.

For many S-T codes the block length is very short. e.g., Alamouti's
code is only over two consecutive symbols, so that's not a lot of time
in most systems.

Unless one of the terminals is very highly mobile with respect to the
other (i.e., the Doppler is really extreme) the channel coherence time
is typically quite long compared to a few symbols. If it weren't
then channel estimation schemes would be scrambling to keep up.

Most systems have reasonably long channel coherence times. Consider
that for many systems, like 802.11, the channel is estimated once at
the beginning of the packet, regardless of the length of the packet,
and that channel estimate is assumed to be good for the entire length
of the packet. This is a very reasonable assumption in static or
pedestrian mobile circumstances (and even higher Doppler than that).

So assuming that the channel is static over the length of a typical
S-T codeword is actually a much easier case if the channel estimates
are being updated. If the channel estimates are not being updated
per S-T codeword the situation is still no worse than the scenario I
described for 802.11, i.e., channel estimation once per packet is
almost always good enough.

Eric Jacobsen
Minister of Algorithms, Intel Corp.
My opinions may not be Intel's opinions.
http://www.ericjacobsen.org
.

Relevant Pages

  • LS channel estimation for LTE
    ... I am developing a least square channel estimation in matlab. ... Once the frequency domain estimates are found we will perform FD ... interpolation it should perfrom better. ...
    (comp.dsp)
  • Re: eliminating a doppler shift in a rayleigh fading channel
    ... the amount of doppler shift in the channel is a one which is related to the channel estimation but there is no way to eliminate it since it is already happened. ...
    (comp.dsp)
  • Re: Use of channel estimation in OFDM
    ... Suppose subcarrier 'k' has a channel gain of Hand this subcarrier ... Typically, for OFDM, channel estimation happens in frequency domain. ...
    (comp.dsp)
  • Re: Use of channel estimation in OFDM
    ... Suppose subcarrier 'k' has a channel gain of Hand this subcarrier is ... I read that OFDM needs channel information at the receiver for coherent ... Typically, for OFDM, channel estimation happens in frequency domain. ...
    (comp.dsp)
  • Re: Status of AIO
    ... Specifically the interrupt handler should simply insert and remove packets from the queue and program the hardware registers for DMA access to the packet buffer memory. ... This is how NT and I believe BSD have been doing things for some time now, and the direction the Linux kernel is moving in. ... channel we created in the first step. ...
    (Linux-Kernel)