Now, we're in a position to formulate our method for calculating the mathematical expectation of ENt from the distribution of size. Let me just recall, that due to the theorem which I showed some time before, a mathematical expectation of ENt is equal to the function U of t, which is defined as the sum of the following series, so it's n from one to infinity, F n star of t. Recall that F n star of t is n forth convolution of the function F. It is F convoluted as F and so on n times. From here, it immediately follows that this sum is equal to F of t plus the same sum, 1 to infinity F n star, convoluted towards the function F. Here, we apply the property of [inaudible]. Okay, what we have here is actually exactly the function U. And therefore, from here, we conclude that the function U of t is equal to F plus U convoluted with F. And here, the convolution is understood in the sense of the distribution functions. Now, I would like to apply the Laplace transform to both parts of this equality. And you know that the Laplace transform from the convolution is equal to the product of Laplace transforms of corresponding functions. Nevertheless, in that property, the convolution is understood in the sense of densities and therefore we should somehow change this operation in order to apply the Laplace transform to this object. And I claim here that actually this object is equal to F plus a convolution between U and density of p if the density exists. So here, we assume additionally that the function capital F has a density. And if it's so, then this equality is fulfilled and the convolution in the right hand side of this equality is actually a convolution in the sense of densities. Why is this equality holds? This is nothing more than the application of the corresponding definitions. In fact, what is written in the left-hand side is the integral over R, U, x minus y, dF in the point Y. And what is written in the left-hand side is the integral over R, U, x minus y, p of y dy. Definitely, this is to all [inaudible] as it say. So, this equality can look a little bit strange because what you have here is the same operation. But basically, it is understood in different senses. So, if we know that the function U is equal to F plus U convoluted with p, then we can take Laplace transforms from both parts of this equality and finally get the following, Laplace transform as a function U at the point s is equal to Laplace transforms of the function F plus Laplace transform of the function U multiplied with Laplace transform of the function p. Here we should apply one more property of the Laplace transform. Actually, I have shown that Laplace transform of F is equal to Laplace transform of p divided by s. And from this formula we actually get the following, we conclude that the Laplace transform of the function U is equal to Laplace transform of the function p divided by s multiplied by 1 minus Laplace transform of the function p. And this formula gives rise for the following method. So, in order to solve our problem to calculate mathematical expectation of ENt from F, we should make the following three steps. Step number one, from F, we can calculate Laplace transform of p. Basically, we are awful we have not f by p, if you don't have p you can take the derivative from f. So, this isn't a quite simple operation. Second step, from Laplace transform of p, we should calculate Laplace transform of U based on this formula. And third step, from the Laplace transform of U we should guess which U was used for this Laplace transform. In other words, which U corresponds to this Laplace transform. This is the most difficult operation, because theoretically, we know that only one U corresponds to the Laplace transform. There is one by one correspondence between functions and their Laplace transforms. But on the other side, the exact form of the inverse operation is rather difficult since the so-called Bromwich integral and in most cases it can be calculated but with a lot of difficulties. And therefore, normally you shall simply guess which U corresponds to this L of U. I guess it will be a nice idea to provide an example how I can use this scheme in practice. And this is exactly the topic of the next subsection.