ON THE REFLECTED FRACTIONAL BROWNIAN MOTION PROCESS ON THE POSITIVE ORTHANT: ASYMPTOTICS FOR A MAXIMUM WITH APPLICATION TO QUEUEING NETWORKS

被引:5
|
作者
Delgado, Rosario [1 ]
机构
[1] Univ Autonoma Barcelona, Dept Matemat, E-08193 Barcelona, Spain
关键词
Fluid queue; Heavy traffic; Maximum; Reflected fractional Brownian motion; State space collapse; Workload; INPUT; LIMIT;
D O I
10.1080/15326341003756437
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Let W be a J-dim. reflected fractional Brownian motion process (rfBm) on the positive orthant IR(+)(J), with drift theta is an element of IR(J) and Hurst parameter H is an element of (0, 1), and let a is an element of IR(+)(J), a not equal 0, be a vector of weights. We define M(t) = max(0 <= s <= t) a(T) W (s) and prove that M(t) grows like t if mu = a(T) theta > 0, in the sense that its increase is smaller than that of any function growing faster than t, and if a restriction on the weights holds, it is also bigger than that of any function growing slower than t. We obtain similar results with t(H) instead of t in the driftless case (theta = 0). If mu < 0 we prove that the increase of M(t) is smaller than that of any function growing faster than t and also that (log t)(1/2(1-H)) is a lower bound for M(t). Motivation for this study is that rfBm appears as the workload limit associated to a fluid queueing network fed by a big number of heavy-tailed On/Off sources under heavy traffic and state space collapse; in this scenario, M(t) can be interpreted as the maximum amount of fluid in a queue at the network over the interval [0, t], which turns out to be an interesting performance process to describe the congestion of the queueing system.
引用
收藏
页码:272 / 294
页数:23
相关论文
共 50 条
  • [21] A SUFFICIENT CONDITION FOR THE POSITIVE RECURRENCE OF A SEMIMARTINGALE REFLECTING BROWNIAN MOTION IN AN ORTHANT
    Chen, Hong
    ANNALS OF APPLIED PROBABILITY, 1996, 6 (03): : 758 - 765
  • [22] Asymptotics of minimum distance estimator of the parameter of stochastic process driven by a fractional Brownian motion
    Kouame, Louis
    N'Zi, Modeste
    Yode, Armel Fabrice
    RANDOM OPERATORS AND STOCHASTIC EQUATIONS, 2008, 16 (04) : 399 - 407
  • [23] Reflected fractional Brownian motion in one and higher dimensions
    Vojta, Thomas
    Halladay, Samuel
    Skinner, Sarah
    Janusonis, Skirmantas
    Guggenberger, Tobias
    Metzler, Ralf
    PHYSICAL REVIEW E, 2020, 102 (03)
  • [24] On the supremum of γ-reflected processes with fractional Brownian motion as input
    Hashorva, Enkelejd
    Ji, Lanpeng
    Piterbarg, Vladimir I.
    STOCHASTIC PROCESSES AND THEIR APPLICATIONS, 2013, 123 (11) : 4111 - 4127
  • [25] Perturbative expansion for the maximum of fractional Brownian motion
    Delorme, Mathieu
    Wiese, Kay Joerg
    PHYSICAL REVIEW E, 2016, 94 (01)
  • [26] Asymptotics for the Green's functions of a transient reflected Brownian motion in a wedge
    Franceschi, Sandro
    Kourkova, Irina
    Petit, Maxence
    QUEUEING SYSTEMS, 2024, 108 (3-4) : 321 - 382
  • [27] Singular control of (reflected) Brownian motion: a computational method suitable for queueing applications
    Ata, Baris
    Harrison, J. Michael
    Si, Nian
    QUEUEING SYSTEMS, 2024, 108 (3-4) : 215 - 251
  • [28] Asymptotics of Karhunen-Loeve Eigenvalues for Sub-Fractional Brownian Motion and Its Application
    Cai, Chun-Hao
    Hu, Jun-Qi
    Wang, Ying-Li
    FRACTAL AND FRACTIONAL, 2021, 5 (04)
  • [29] Non-Gaussian behavior of reflected fractional Brownian motion
    Wada, Alexander H. O.
    Warhover, Alex
    Vojta, Thomas
    JOURNAL OF STATISTICAL MECHANICS-THEORY AND EXPERIMENT, 2019,
  • [30] Distribution of maximum loss of fractional Brownian motion with drift
    Caglar, Mine
    Vardar-Acar, Ceren
    STATISTICS & PROBABILITY LETTERS, 2013, 83 (12) : 2729 - 2734