Changeset 9736

Show
Ignore:
Timestamp:
11/12/09 15:39:02 (4 years ago)
Author:
hvolos
Message:

latest changes

Files:
1 modified

Legend:

Unmodified
Added
Removed

  • ossiedev/branches/hvolos/docs/MIMOlab/MIMOLab.tex

    r9697 r9736  
    8585\end{enumerate} 
    8686 
     87\subsection{Extra Notes} 
     88\begin{enumerate} 
     89\item Make sure to stop the waveform's execution, if using WAVEDASH to change any property other than the ``noise power'' 
     90\item If you have extra time, you may use ALF to plot the waveform's signals 
     91\end{enumerate}  
     92 
    8793 
    8894\subsection{Getting Started} 
    … …  
    99105The waveform's structure is the same for all methods: MRC, STBC, and VBLAST. \osref{waveformblockdiagram} provides the block diagram of the waveform. The following components are used: 
    100106\subsubsection{TxDemo \& RxDemo} 
    101 TxDemo generates known QPSK modulated symbols which are used to the test the communication methods. The RxDemo demodulates the received symbols and calculates the number of bit errors. 
     107TxDemo generates known QPSK modulated symbols which are used to the test the communication methods. RxDemo demodulates the received symbols and calculates the number of bit errors. 
    102108 
    103109\subsubsection{MIMOEnc} 
    … …  
    116122\item File $\rightarrow$ New $\rightarrow$ OSSIE Waveform 
    117123\item Project name: ``MIMOTest'' 
    118 \item Add the following components to the waveform panel:TxDemo,MIMOEnc,MIMOChannel,MIMODec,RxDemo 
     124\item Add the following components to the waveform panel: TxDemo, MIMOEnc, MIMOChannel, MIMODec, RxDemo 
    119125\item Add the default\_GPP\_node to the platform panel 
    120126\item Expand the the node, and add all the components, one by one, to the GPP device. 
    … …  
    135141\subsection{Component Properties} 
    136142For each component set the listed properties. 
    137 \subsubsection{MIMOEnc \&MIMODec} 
     143\subsubsection{MIMOEnc \& MIMODec} 
    138144\begin{verbatim} 
    139145Rx Antennas = 2 
    … …  
    144150\subsubsection{ChannelMIMO} 
    145151\begin{verbatim} 
    146 Noise Power = 1225000  
     152Noise Power = 50000000 
    147153No of Rx Antennas = 2 
    148154No of Tx Antennas = 1 
    149155\end{verbatim}  
    150156 
    151  
    152  
    153157\subsection{Waveform Testing} 
    154158\begin{enumerate} 
    155  \item Built and run the waveform using the default\_GPP\_node. The each component outputs the number of antennas and MIMO method used. The RxDemo component outputs the number of erroneous bits. 
    156 \item The AWGN noise power of 1225000 results to an SNR of approximately 16 dB (ignoring the fading aspect). At this level there should be severe error counts. 
    157 \item Change the number (in all components) of receive antennas to 2, there should be an improvement in the number of errors, however, they are still severe. 
    158 \item Now change the number of antennas to 8. There should be bursts of 0 errors and bursts of high number of errors. 
    159 \end{enumerate} 
    160  
    161   
     159 \item Built and run the waveform using the default\_GPP\_node. Each component outputs the number of antennas and MIMO method used. The RxDemo component outputs the number of erroneous bits. 
     160\item The AWGN noise power of 50000000 results to an SNR of approximately 3 dB (ignoring the fading aspect). At this level there should be several bursts of high error counts. 
     161\item Change the number (in all components) of receive antennas to 4, the frequency of high error count packets should be reduced. 
     162\item Now change the number of receive antennas to 8. Most packets should now have zero or near zero error counts. 
     163\item (Optional) change the number of receive antennas to 16, there should be no errors. 
     164\item For the last number of receive antennas, increase the noise power so that the error occurrence approximately matches what was observed in the first step. Try to estimate the SNR gain in dB when using the multiple antennas. Assume a signal power of $1\times10^8$. 
     165\end{enumerate} 
    162166 
    163167 
    … …  
    167171\subsection{Component Properties} 
    168172For each component set the listed properties. 
    169 \subsubsection{MIMOEnc \&MIMODec} 
     173\subsubsection{MIMOEnc \& MIMODec} 
    170174\begin{verbatim} 
    171175Rx Antennas = 1 
    … …  
    177181\subsubsection{ChannelMIMO} 
    178182\begin{verbatim} 
    179 Noise Power = 1225000  
     183Noise Power = 25000000 
    180184No of Rx Antennas = 1 
    181185No of Tx Antennas = 2 
    … …  
    186190\subsection{Waveform Testing} 
    187191\begin{enumerate} 
    188  \item Built and run the waveform using the default\_GPP\_node. The each component outputs the number of antennas and MIMO method used. The RxDemo component outputs the number of erroneous bits. 
    189 \item In this setup, the AWGN noise power of 1225000 results to an SNR of approximately 13 dB (ignoring the fading aspect). At this level there should burst of errors. 
    190 \item Change the number (in all components) of receive antennas to 2, there frequency of the error bursts should be reduced. 
    191 \item Now change the number of antennas to 8. The bursts of errors should be even sparser. 
     192 \item Built and run the waveform using the default\_GPP\_node. Each component outputs the number of antennas and MIMO method used. The RxDemo component outputs the number of erroneous bits. 
     193\item In this setup, the AWGN noise power of 25000000 results to an SNR of approximately 3 dB (ignoring the fading aspect). At this level there should several bursts high error counts. 
     194\item Change the number (in all components) of receive antennas to 2, the frequency of the error high error bursts should be reduced and most error counts should be around 20 
     195\item (Optional) change the number of receive antennas to 16, there should be mostly no errors 
     196\item For the last number of receive antennas, increase the noise power so that the error occurrence approximately matches what was observed in the first step. Try to estimate the SNR gain in dB when using the multiple antennas. Assume a signal power of $5\times10^7$. 
     197 
    192198\end{enumerate} 
    193199\section{Spatial Multiplexing} 
    194200\subsection{Component Properties} 
    195201For each component set the listed properties. 
    196 \subsubsection{MIMOEnc \&MIMODec} 
     202\subsubsection{MIMOEnc \& MIMODec} 
    197203\begin{verbatim} 
    198204Rx Antennas = 2 
    … …  
    204210\subsubsection{ChannelMIMO} 
    205211\begin{verbatim} 
    206 Noise Power = 24500  
     212Noise Power = 25000 
    207213No of Rx Antennas = 2 
    208214No of Tx Antennas = 2 
    … …  
    212218\subsection{Waveform Testing} 
    213219\begin{enumerate} 
    214  \item Built and run the waveform using the default\_GPP\_node. The each component outputs the number of antennas and MIMO method used. The RxDemo component outputs the number of erroneous bits. 
    215 \item In this setup, the AWGN noise power of 24500 results to an SNR of approximately 30 dB (ignoring the fading aspect). At this level there should be a lot of burst of errors. 
    216 \item Change the noise power to 2450 (40dB), the error occurrence should be now reduced. 
     220 \item Built and run the waveform using the default\_GPP\_node. Each component outputs the number of antennas and MIMO method used. The RxDemo component outputs the number of erroneous bits. 
     221\item In this setup, the AWGN noise power of 25000 results to an SNR of approximately 33 dB (ignoring the fading aspect). At this level there should be a lot of high error bursts 
     222\item Change the noise power to 5000 (40dB), the error bursts occurrence should be now reduced 
    217223\end{enumerate} 
    218224\end{document}