How to make time shift in a vector using non integer values?

I need to implement this equation

I(t-Ts)= co*n(t) +c1* n(t-Ts)

Where Ts is 1 nanosecond, so I keep getting an error " Array indices must be positive or logical values". How can I implement that shift in time? Thank you.

[Stephen23]

Most likely you are confusing data (e.g. time) with code (e.g. indices). Please show your code.

[Kim Ibrahim]

@Stephen Cobeldick

yes, i'm iterating on t please see the piece of code:

for t =1:10000

nc(t)= sqrt(d0*input(t)+d1)+d2;

I(t)=co*nc(t)+c1*nc(t-Ts)+c2*(nc(t-Ts)^2)+c3*(nc(t-Ts)^3);

end

where Ts is the sampling time choosen to be 1ns, and input is 1x10000 vector

Answer 1

[Adam Danz]

I(t-Ts)= co*n(t) +c1* n(t-Ts)

I'm guessing that t has a value of 1 at some point and ts=1 nanosecond, so (t-ts)=0 at some point. As the error message indicates, the array indices must be positive values (or logical values).

Other possibilities are that (t-ts) results in a positive non-integer value or a negative value, both of which cannot be used as indices.

Note that this index problem happens with I(t-Ts) and with n(t-Ts). Without knowing more about what you're doing, I can't provide a more specific solution.

[Kim Ibrahim]

yes, i'm iterating on t please see the piece of code:

for t =1:10000
    nc(t)= sqrt(d0*input(t)+d1)+d2;
    I(t)=co*nc(t)+c1*nc(t-Ts)+c2*(nc(t-Ts)^2)+c3*(nc(t-Ts)^3);    
end

where Ts is the sampling time choosen to be 1ns, and input is 1x10000 vector

thank you

[Adam Danz]

Looks like my answer was correct.

On the first iteration, t=1 and ts=1 so (t-Ts)=0

In Matlab, indices start with 1 unlike other languages that start with 0.

[Kim Ibrahim]

so is there a work around to fix this and implemet the wanted equation? also Ts is 10^(-9)

[Adam Danz]

10^-9 is a lot different than 1(ns). Providing this level of detail from the start is important and will save lots of time.

Indexing requires positive integers. For example, if 'a' is an array, a(j) is interpreted as the j'th value in a. If j=4, a(4) is the 4th value.

Here's another example, b=[1,2,3,4]; what's the 0.004'th value of b? That question doesn't even make sense. Neither does "nc(t-Ts)".

> is there a work around to fix this and implemet the wanted equation

To answer that, we need to understand what you intended to do with the equation and until mind-reading technology becomes available, we cannot answer that question.

I can make a guess and suggest that you wanted this,

for t =1:10000
    nc(t)= sqrt(d0*input(t)+d1)+d2;
    I(t)=co*nc(t)+c1*nc(t)-Ts+c2*(nc(t)-Ts)^2+c3*(nc(t)-Ts)^3;    
end

but ultimately you have to 1) understand what the line is supposed to be doing and then 2) determine if it's doing that.

It might be helpful to see this color-coded version of how I re-grouped the variables.

[Kim Ibrahim]

Okay, I understand better now, thank you

[Adam Danz]

Glad I could point you in the right direction. If this answered your question, please consider accepting the answer (blue "accept this answer" button). If you have any other questions on this topic, let us know.