<< Chapter < Page Chapter >> Page >

Shown across the top of [link] is a stylized version of that seen in the bottom portion of [link] . The uniform weight shown on the top right leads to the bandpass filter shapes shown on the left. Note that the filters are separated in frequency by f s N Hz.

Figure nine is a four-part figure. Part a contains a graph with vertical axis labeled transfer function. It is a series of uniform waves that begin and end at the horizontal axis. They cross each other and continue for half the figure, then continue with some dots in a horizontal line, then finish with more waves. Below the graph is a distance measured between the beginning of a wave and its apex, and the width is labeled Δf = f_s/N. To the right of part a is a horizontal axis containing evenly-spaced line segments, titled Uniform weighting. Part b contains a graph with waves that have wide, flat peaks. The width of half of the wave is measured as Δf = f_s/N. the vertical axis of the graph is labeled Transfer Function. to the right of the graph is a second graph with one major positive wave and two minor negative waves with their area drawn down or up to the horizontal axis shaded black, and labeled Non-uniform weighting. Part c contains another graph labeled Transfer function. Its waves are similar to the waves in a, but are aligned closer together. The distance measured on the bottom is labeled δf = Δf/Q = f_s/QN. The graph to the right is similar to the graph in part b, except that it is wider, and labeled Non-uniform weighting; N' = QN. Part d is wider than parts a and c, but narrower in peaks than part b, and its width is measured as Δf = f_s/N, with the identical graph to the right of waves as part c, labeled Non-uniform Weighting; N' = QN; Thinning Output. Below all four figures is a label describing the horizontal axes as Frequency.
Effects of Changing Data Weighting and DFT Size

Now suppose we employ non-uniform weighting to improve the shape of the bandpass filters. As discussed in Appendix A , such non-uniform weighting can be used to attain the desired transfer function shape, but virtually always at the expense of the bandpass filter's bandwidth. In fact, to obtain the desired characteristic shown in [link] , with its flat passband, sharp skirts, and high-attenuation stopband, the minimum passband bandwidth is more than a factor of ten larger than the unweighted response. Thus the use of a non-uniform weighting, as shown on the right of [link] (b), results in the situation shown on the left side. There are still N bandpass filters, and their center frequencies are still separated by integer multiples f s N Hz, but each filter has been widened considerably, leading to a high degree of overlap.

The first problem to deal with is not the overlap, but rather the fact that the individual bandpass filters are far wider than the original goal of about f s N Hz. This is dealt with by returning to [link] and simply letting the delay line length, the number of weighting coefficients, and the DFT order grow until the filters are sufficiently narrowband to meet our objectives. Again using the example of the desired frequency response seen in [link] , the dimensions must grow by more than a factor of ten.

While the resulting dimensions can take on rather arbitrary values (above some minimum value) we'll assume here that the new size N' is an integer multiple of N . In particular, we assume that the delay line, and the weighting and DFT with it, are extended to the length N' where:

N ' = Q · N ,

where Q is a positive integer. We further assume that Q is chosen to be large enough that a weighting function of length N ' can be designed to produce not only the desired shape but also a bandwidth of about f s N Hz. The resulting situation is shown in [link] (c). The weighting function is now longer than before (by a factor of Q ). On the left we see that there are now N ' filters in the filter bank. Each one of them now has the desired nominal bandwidth of f s N Hz, but their center frequencies are now separated by δ f = f s N ' = f s N Q Hz instead of f s N Hz. The overlap seen just above still exists but now there is a factor of Q more filters, a factor of Q narrower, and a factor of Q more closely spaced. Thus the positive effect of expanding the delay line dimension to Q N is that the resulting filter bank includes the desired bandpass filters, both in bandpass characteristics and center frequencies. The negative aspects include the fact that the amount of weighting and DFT computation have gone up by a factor of Q and that there are now ( Q - 1 ) · N superfluous bandpass filters.

Suppose now that we choose to compute only every Q-th point of the DFT. The delay line is still Q N samples long, there are still Q N coefficients in the weighting function, and the DFT still has order Q N , but we'll choose to only compute those output bins X m ( k ) where m is an integer multiple of Q . This results in the situation shown in [link] (d). The same QN-point weighting function is used as immediately above. This case, with N filters of nominal bandwidth f s N Hz and spaced f s N Hz apart, was our objective. To achieve it, however, required expanding the dimensions of the preceding operations quite considerably.

Questions & Answers

how do I set up the problem?
Harshika Reply
what is a solution set?
Harshika
hello, I am happy to help!
Shirley Reply
please can go further on polynomials quadratic
Abdullahi
I need quadratic equation link to Alpa Beta
Abdullahi Reply
find the value of 2x=32
Felix Reply
divide by 2 on each side of the equal sign to solve for x
corri
X=16
Michael
Want to review on complex number 1.What are complex number 2.How to solve complex number problems.
Beyan
use the y -intercept and slope to sketch the graph of the equation y=6x
Only Reply
how do we prove the quadratic formular
Seidu Reply
hello, if you have a question about Algebra 2. I may be able to help. I am an Algebra 2 Teacher
Shirley Reply
thank you help me with how to prove the quadratic equation
Seidu
may God blessed u for that. Please I want u to help me in sets.
Opoku
what is math number
Tric Reply
4
Trista
x-2y+3z=-3 2x-y+z=7 -x+3y-z=6
Sidiki Reply
Need help solving this problem (2/7)^-2
Simone Reply
x+2y-z=7
Sidiki
what is the coefficient of -4×
Mehri Reply
-1
Shedrak
the operation * is x * y =x + y/ 1+(x × y) show if the operation is commutative if x × y is not equal to -1
Alfred Reply
An investment account was opened with an initial deposit of $9,600 and earns 7.4% interest, compounded continuously. How much will the account be worth after 15 years?
Kala Reply
lim x to infinity e^1-e^-1/log(1+x)
given eccentricity and a point find the equiation
Moses Reply
A soccer field is a rectangle 130 meters wide and 110 meters long. The coach asks players to run from one corner to the other corner diagonally across. What is that distance, to the nearest tenths place.
Kimberly Reply
Jeannette has $5 and $10 bills in her wallet. The number of fives is three more than six times the number of tens. Let t represent the number of tens. Write an expression for the number of fives.
August Reply
What is the expressiin for seven less than four times the number of nickels
Leonardo Reply
How do i figure this problem out.
how do you translate this in Algebraic Expressions
linda Reply
why surface tension is zero at critical temperature
Shanjida
I think if critical temperature denote high temperature then a liquid stats boils that time the water stats to evaporate so some moles of h2o to up and due to high temp the bonding break they have low density so it can be a reason
s.
Need to simplify the expresin. 3/7 (x+y)-1/7 (x-1)=
Crystal Reply
. After 3 months on a diet, Lisa had lost 12% of her original weight. She lost 21 pounds. What was Lisa's original weight?
Chris Reply
Got questions? Join the online conversation and get instant answers!
Jobilize.com Reply

Get the best Algebra and trigonometry course in your pocket!





Source:  OpenStax, An introduction to the fdm-tdm digital transmultiplexer. OpenStax CNX. Nov 16, 2010 Download for free at http://cnx.org/content/col11165/1.2
Google Play and the Google Play logo are trademarks of Google Inc.

Notification Switch

Would you like to follow the 'An introduction to the fdm-tdm digital transmultiplexer' conversation and receive update notifications?

Ask