Convolution between two different distribution

2015 10 6
1 个回答
更新时间：2015 10 6
4 次查看（30 天）

0 个投票

Hello
I want to generate C(u) in the following:
C(u)=a(-u)*b(u) {a(-u) convolution b(u)}
which a is impulsive (for example [0 0 0 0 0 1 0 0 0 0]) and b is a probability distribution (for example a Normal distribution)
Thank you

6 个评论
显示 4更早的评论 隐藏 4更早的评论

Walter Roberson
Walter Roberson 2015-10-6
What do you intend a(-u) and b(u) to mean? Is a(-u) subscripting with a "u" that happens to be negative? If so then if b(u) is to indicate subscripting, then it would be with a negative value, and subscripting a probability distribution at a position would not appear to make sense.
Image Analyst
Image Analyst 2015-10-6
To know what a(-u) is, you need to know which element is the "origin" so we can flip the vector about that origin. Anyway, the answer will of course just be b again, with some padding of zeros on the outer edges.
Adam
Adam 2015-10-6
Can you not just use the builtin conv function?
jafar
jafar 2015-10-6
To Walter Roberson, Image Analyst and Adam
a(u) and b(u) are two probability density functions ( any pdf could be used e.g two gaussian distributions with different mean, or any other distributions)
Explanation of whole question is attached
thank you

请先登录，再进行评论。

请先登录，再回答此问题。

回答（1 个）

Image Analyst
Image Analyst 2015-10-6

0 个投票

If the origin is halfway between elements 5 and 6 of "a", then just do
C = conv(fliplr(a), b, 'full');

1 个评论
显示 -1更早的评论 隐藏 -1更早的评论

jafar
jafar 2015-10-6
Thank You
No!
The origin is not halfway between elements 5 and 6 of "a", it is on zero,

请先登录，再进行评论。

类别

在 帮助中心 和 File Exchange 中查找有关 Uniform Distribution (Continuous) 的更多信息

标签

提问:

jafar
jafar
2015-10-6

评论：

jafar
jafar
2015-10-6

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by