首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 15 毫秒
1.
2.
We consider a class of stochastic networks with state-dependent arrival and service rates. The state dependency is described via multi-dimensional birth/death processes, where the birth/death rates are dependent upon the current population size in the system. Under the uniform (in state) stability condition, we establish several moment stability properties of the system:
  • (i) 
    the existence of a moment generating function in a neighborhood of zero, with respect to the unique invariant measure of the state process;
  • (ii) 
    the convergence of the expected value of unbounded functionals of the state process to the expectation under the invariant measure, at an exponential rate;
  • (iii) 
    uniform (in time and initial condition) estimates on exponential moments of the process;
  • (iv) 
    growth estimates of polynomial moments of the process as a function of the initial conditions.
Our approach provides elementary proofs of these stability properties without resorting to the convergence of the scaled process to a stable fluid limit model.  相似文献   

3.
4.
5.
6.
When constructing uniform random numbers in [0, 1] from the output of a physical device, usually n independent and unbiased bits B j are extracted and combined into the machine number . In order to reduce the number of data used to build one real number, we observe that for independent and exponentially distributed random variables X n (which arise for example as waiting times between two consecutive impulses of a Geiger counter) the variable U n : = X 2n – 1/(X 2n – 1 + X 2n ) is uniform in [0, 1]. In the practical application X n can only be measured up to a given precision (in terms of the expectation of the X n ); it is shown that the distribution function obtained by calculating U n from these measurements differs from the uniform by less than /2.We compare this deviation with the error resulting from the use of biased bits B j with P {B j = 1{ = (where ] – [) in the construction of Y above. The influence of a bias is given by the estimate that in the p-total variation norm Q TV p = ( |Q()| p )1/p (p 1) we have P Y P 0 Y TV p (c n · )1/p with c n p for n . For the distribution function F Y F 0 Y 2(1 – 2n )|| holds.  相似文献   

7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号