📄 index.htm
字号:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<!-- saved from url=(0049)http://www.itr.unisa.edu.au/cbt/mobile/index.html -->
<HTML><HEAD><TITLE></TITLE>
<META content="text/html; charset=gb2312" http-equiv=Content-Type>
<META content="MSHTML 5.00.2920.0" name=GENERATOR>
<META content="Inst Telecoms Research" name=AUTHOR></HEAD>
<BODY>
<P><IMG height=21 src="index.files/unisa.gif" width=563></P>
<P><B><FONT color=#800000><FONT size=+2>Mobile Communications
Software</FONT></FONT><FONT color=#0000ff> </FONT></B></P>
<P>Software provided for the Mobile Comms short course in April 1997. Use shift
and left mouse button to save each of the following files. </P>
<UL>
<LI>Flat-Fading Rayleigh Channel Simulations <A
href="http://www.itr.unisa.edu.au/cbt/mobile/ray1.m">ray1.m,
</A><A
href="http://www.itr.unisa.edu.au/cbt/mobile/ray2.m">ray2.m </A>
<LI>Wideband Channel Illustrations <A
href="http://www.itr.unisa.edu.au/cbt/mobile/delay1.m">delay1.m,
</A><A
href="http://www.itr.unisa.edu.au/cbt/mobile/delay2.m">delay2.m,
</A><A
href="http://www.itr.unisa.edu.au/cbt/mobile/delay3.m">delay3.m </A>
<LI>CDMA Demonstration (this version has some graphics) <A
href="http://www.itr.unisa.edu.au/cbt/mobile/cdma_ex.m">cdma_ex.m</A> </LI></UL>
<P>Software on the 'Sat Comms' web page may also be of interest. </P>
<P><B><FONT color=#800000>Comments regarding Assignment 3, Question 1:
(Modulation and Diversity)</FONT></B></P>
<P><FONT color=#000000>Parts a) and b) of this question are very easy, but parts
c) and d) are quite hard to derive from scratch. It will therefore be acceptable
in the assignment to simply give a short outline of the derivations and then
state the results. (Both Yacoub and the Proakis text contain these derivations
and associated results.) </FONT></P>
<P><FONT color=#000000>Although the derivations are tricky, the final results in
these two cases (ie BER for coherent BPSK in Rayleigh channel with a single
path, and with 2 path diversity) are very simple (at least when the 'high SNR'
approximations are employed). In both cases the BER can be written as a simple
function of mean bit energy to noise spectral density. The learning objective
here is to determine these results, understand when they apply and appreciate
the differences compared to unfaded channels. </FONT></P>
<P><FONT color=#000000>You may care to undertake some of the relevant
simulations which are suggested below (parts c) and d)) to get a better
understanding of these results. Inclusion of relevant simulation results in the
assignment will result in bonus marks. </FONT></P>
<P><B><FONT color=#800000>Mobile Communications - extra Matlab simulations in
the channel modelling and modulation topics. </FONT></B></P>
<P><B>a)</B> The "ray1.m" simulation plots fading channel magnitude. From the
channel stored in vector "R", use matlab to plot the channel phase. (Also try
using the "unwrap" function while plotting channel phase.) Is this what you
expect? </P>
<P><B>b)</B> I've added some code ro "ray1.m" to make it easier to change the
fading bandwidth- see the online help and try running the program with (say)
ftype=0.1 Examine the magnitude response and see if it changes as you expect.
</P>
<P><B>c)</B> The third assignment for the Mobile Comms subject requests the
derivation of bit error rate formula for a BPSK signal with coherent detection
in a Rayleigh channel (with negligible fading bandwidth). Use matlab to plot the
ber versus mean Eb/No, and compare the plot to the figure on slide 20 (of slide
set 3). You can also use matlab to plot the ber performance of BPSK in an
unfaded channel, and of the optimum 2 path diversity Rayleigh channel
performance. </P>
<P><B>d)</B> Using the output data from "ray1", simulate a DBPSK mobile
communications system operating in a noisy channel with added noise. </P>
<P><B><FONT color=#800000>Hints</FONT></B>: 1) To make this easy, don't worry
about simulating random bits or implementing differential encoding. If we assume
that the same information bit is transmitted, then the modulator output will
always be unity; thus we can simply add (complex) noise to the Rayleigh channel
vector to form our demod imput signal. </P>
<P>2) The differential detector works out the phase difference between adjacent
symbols in order to make a decision on what symbol was received. The easiest way
to do this in matlab is to multiply each receive sample by the complex conjugate
of the previous sample- and then take the phase of this product. In this
simulation the tx signal is always the same, hence the phase of the product
should be zero. Any phase with magnitude > pi/2 would cause an error. </P>
<P>3) Following the hints above, count the number of errors and so estimate the
ber. Check how the ber changes as the signal to noise ratio and the fading
bandwidth change. [If you want to "calibrate" the simulation in terms of mean
Eb/No then use the relation that (mean Eb)/No = (mean signal power) / (2*
sigma^2), where sigma^2 is the noise power added to each of the real amd imag
parts of the demod input signal. This calibration would allow you to compare
your simulation results to plots given in the notes.] You should use the new
option in ray1.m to reduce the fading bandwidth for these simulations, since
running ray1 with default values produces a Rayleigh channel with very high
fading bandwidth. </P>
<P>WGC, 20/4/97 </P></BODY></HTML>
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -