Eric, i've found out that what u mentioned previously was something
about IQ mod/demod right? If i implement that then the complex
conjugation for the synchronization at the receiver would be
straightforward.
But what if i decide only to transmit the real signal without IQ
modulation? Meaning after DAC i modulate it with my desired carrier
frequency and transmit into the power line channel. Then such
conjugation in synchronization cant be performed right?
Then how should i attempt this?
Reply by Ant_Magma●December 19, 20052005-12-19
Ah, got it.
Thx Phil =)
Reply by Phil●December 19, 20052005-12-19
Ant,
What I had in mind when I said continuous was a signal that is
transmitted in a stream. There is no real beginning or end when it
comes to receiving it. Burst transmission is when the signal is broken
up into segments (bursts). These usually have a preamble or some other
similar structure to help detect, acquire, power adjust, ... so that
the data can be demodulated successfully.
So as you said, it is governed by how the transmission is designed.
Reply by Ant_Magma●December 19, 20052005-12-19
Phil wrote:
> Eric,
>
> I think part of the confusion in this conversation stems from your
> familiarity of the wireless OFDM standards. You've been assuming that
> the baseband signal is complex. Until I began working for a wireline
> company this year, I would have made the same assumption.
>
> Since Ant mentioned that he is looking at power line, I'm assuming that
> he is looking basband signals that go over the line, i.e., -ve
> frequency bins are complex conjugate of +ve frequency bin.
>
> Ant, you did not mention whether you are looking at OFDM continuous
> signals or if you are considering bursty signals that contain
> preambles. This, along with burst durations and the dynamics of the
> channel will impact the detection and synchronization algorithms that
> you should choose.
>
> I hope that was an appropriate interruption to the string.
>
> Phil
Phil, could you explain abit how do i differenciate continuous and
burst transmission in Simulink? or is it just governed by the way my
transmission packet is designed?
Reply by Ant_Magma●December 15, 20052005-12-15
>Since Ant mentioned that he is looking at power line, I'm assuming that
>he is looking basband signals that go over the line, i.e., -ve
>frequency bins are complex conjugate of +ve frequency bin.
I'm already doing that in my Simulink model. I'm using a 64 IFFT with
24 data subcarriers, with another 24 as conjugate carriers. Thus, the
output of my transmitter in the model is a real signal.
>Ant, you did not mention whether you are looking at OFDM continuous
>signals or if you are considering bursty signals that contain
>preambles. This, along with burst durations and the dynamics of the
>channel will impact the detection and synchronization algorithms that
>you should choose.
Well, i'm not very sure how would i determine a continuous or burst
transmission in my model. With HomePlug, the spec for power line
communication. Data is transmitted as burst signals, with preambles
SYNCM and SYNCP. In Simulink, how do i model such that my transmission
would be continuous or bursty? since now everything is just output to a
column vector.
That is what i'm trying to do. However, since the complete specs is
only available to members, i do not know what is SYNCM and SYNCP.
>I kinda suspected that. The burden is really on the OP to explain
>their problem in enough detail that they'll get useful answers. In
>any case, I think Ant needs to fall back to some basics a bit. Either
>a beginning comm or good DSP text (like Rick's) should provide enough
>info regarding DFT symmetry properties, real to complex conversion,
>etc., to get him rolling.
I did in fact read a couple of comm books such as Proakis' Digital
Communication and Van Nee's Wireless OFDM. However, i'm having a hard
time understanding the math and how exactly does all those things work
in the real-world.
I truly apologise if i sound like i didn't make an effort to do my
research before asking, but the truth is i did (well, i guess i din do
enough). The fact is, i'm an Electronics major and i havent took any
communication related subjects/courses.
Reply by Ant_Magma●December 15, 20052005-12-15
>Since Ant mentioned that he is looking at power line, I'm assuming that
>he is looking basband signals that go over the line, i.e., -ve
>frequency bins are complex conjugate of +ve frequency bin.
I'm already doing that in my Simulink model. I'm using a 64 IFFT with
24 data subcarriers, with another 24 as conjugate carriers. Thus, the
output of my transmitter in the model is a real signal.
>Ant, you did not mention whether you are looking at OFDM continuous
>signals or if you are considering bursty signals that contain
>preambles. This, along with burst durations and the dynamics of the
>channel will impact the detection and synchronization algorithms that
>you should choose.
Well, i'm not very sure how would i determine a continuous or burst
transmission in my model. With HomePlug, the spec for power line
communication. Data is transmitted as burst signals, with preambles
SYNCM and SYNCP.
That is what i'm trying to do. However, since the complete specs is
only available to members, i do not know what is SYNCM and SYNCP.
>I kinda suspected that. The burden is really on the OP to explain
>their problem in enough detail that they'll get useful answers. In
>any case, I think Ant needs to fall back to some basics a bit. Either
>a beginning comm or good DSP text (like Rick's) should provide enough
>info regarding DFT symmetry properties, real to complex conversion,
>etc., to get him rolling.
I did in fact read a couple of comm books such as Proakis' Digital
Communication and Van Nee's Wireless OFDM. However, i'm having a hard
time understanding the math and how exactly does all those things work
in the real-world.
I truly apologise if i sound like i didn't make an effort to do my
research before asking, but the truth is i did (well, i guess i din do
enough). The fact is, i'm an Electronics major and i havent took any
communication related subjects/courses.
Reply by Jerry Avins●December 15, 20052005-12-15
Eric Jacobsen wrote:
> On 14 Dec 2005 10:35:59 -0800, "Phil" <phil_no_spam@sympatico.ca>
> wrote:
>
>
>>Eric,
>>
>>I think part of the confusion in this conversation stems from your
>>familiarity of the wireless OFDM standards. You've been assuming that
>>the baseband signal is complex.
...
> I kinda suspected that. The burden is really on the OP to explain
> their problem in enough detail that they'll get useful answers.
...
I keep saying that it takes two circuits to get a "complex" signal at
baseband. Why won't anyone believe me? :-(
Jerry
--
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������
Reply by Eric Jacobsen●December 15, 20052005-12-15
On 14 Dec 2005 10:35:59 -0800, "Phil" <phil_no_spam@sympatico.ca>
wrote:
>Eric,
>
>I think part of the confusion in this conversation stems from your
>familiarity of the wireless OFDM standards. You've been assuming that
>the baseband signal is complex. Until I began working for a wireline
>company this year, I would have made the same assumption.
>
>Since Ant mentioned that he is looking at power line, I'm assuming that
>he is looking basband signals that go over the line, i.e., -ve
>frequency bins are complex conjugate of +ve frequency bin.
>
>Ant, you did not mention whether you are looking at OFDM continuous
>signals or if you are considering bursty signals that contain
>preambles. This, along with burst durations and the dynamics of the
>channel will impact the detection and synchronization algorithms that
>you should choose.
>
>I hope that was an appropriate interruption to the string.
>
>Phil
Phil,
I kinda suspected that. The burden is really on the OP to explain
their problem in enough detail that they'll get useful answers. In
any case, I think Ant needs to fall back to some basics a bit. Either
a beginning comm or good DSP text (like Rick's) should provide enough
info regarding DFT symmetry properties, real to complex conversion,
etc., to get him rolling.
Eric Jacobsen
Minister of Algorithms, Intel Corp.
My opinions may not be Intel's opinions.
http://www.ericjacobsen.org
Reply by Phil●December 14, 20052005-12-14
Eric,
I think part of the confusion in this conversation stems from your
familiarity of the wireless OFDM standards. You've been assuming that
the baseband signal is complex. Until I began working for a wireline
company this year, I would have made the same assumption.
Since Ant mentioned that he is looking at power line, I'm assuming that
he is looking basband signals that go over the line, i.e., -ve
frequency bins are complex conjugate of +ve frequency bin.
Ant, you did not mention whether you are looking at OFDM continuous
signals or if you are considering bursty signals that contain
preambles. This, along with burst durations and the dynamics of the
channel will impact the detection and synchronization algorithms that
you should choose.
I hope that was an appropriate interruption to the string.
Phil