0
$\begingroup$

I need some help with this problem.

A Poisson process is generally used to describe the arrival of calls in a mobile network. Since I am developing a simulator for LTE, i need to implement such a process to realistically randomize the entry requests of users wishing to connect.

After reading about the Poisson process I still can't reach a simple solution just to return the arrival times of each request - such as: "new users will try to connect on seconds: 3, 10, 20, 24" - given the average arrival rate and the simulation duration.

Can someone help me with this?

$\endgroup$
2
  • $\begingroup$ Can you clarify this? What is it you need exactly? Are you wondering what a Poisson process is? Are you wondering how to code one? $\endgroup$ Aug 10, 2015 at 14:37
  • $\begingroup$ Yes, I am wondering how to code it, namely on Matlab. Lets suppose I am running a simulation for 1000 seconds. My problem here is, given a certain arrival rate, getting the arriving times for new calls along those 1000 seconds: as in, on seconds 6, 13, 15, whatever, following a poisson distribution. Am I being clear? $\endgroup$
    – Ricardo
    Aug 10, 2015 at 14:47

1 Answer 1

1
$\begingroup$

One way to go would to be to recognize that a Poisson process has exponentially distributed inter-arrival times with a mean parameter the same as the Poisson average rate. I believe the exprnd function in MATLAB can generate such exponentially-distributed data. Just generate a long list of such data, and calculate the cumulative sums to get arrival times.

$\endgroup$
1
  • $\begingroup$ This solved it, thank you! I didn't realize the correspondence between the poisson process and the exponential distributed inter-arrival times $\endgroup$
    – Ricardo
    Aug 11, 2015 at 8:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.