This paper presents a dynamic bandwidth manage- ment scheme for real-time traffic on multiple MPLS paths. We first propose a type of utility function to describe the relationship between end user's satisfaction degree and resources allocated to user's application. Then a heuristic is introduced to perform admission control, path selection and load balancing based on the utility function to maximize the sum of the satisfaction degree of all users. Performance of our heuristic is compared with existing multiple paths routing algorithms. Simulation results show that our scheme can achieve higher total utility and lower blocking probability than conventional algorithms.

How the QoS performance of a communication network is affected by the disturbances, such as channel impairments link failures, node mobility, etc.? In other words, how robust is the QoS performance? Before answering these questions, we need a clear definition of QoS performance robustness and a method to quantify it. This paper contributes in the two aspects. It first proposes a new approach to define performance robustness before outlining a method to quantify and calculate robustness as a metric. Later, a case study is presented where the proposed robustness metric is used to compare two existing routing algorithms for a wireless multi-hop network

Frequent changes in network topologies caused by mobility in mobile ad hoc networks (MANETs) impose great challenges to designing routing schemes for such networks. Various routing schemes each aiming at particular type of MANET (e.g., flat or clustered MANETs) with different mobility degrees (e.g., low, medium, and high mobility) have been proposed in the literature. However, since a mobile node should not be limited to operate in a particular MANET assumed by a routing scheme, an important issue is how to enable a mobile node to achieve routing performance as high as possible when it roams across different types of MANETs. To handle this issue, a quantity that can predict the link status for a time period in the future with the consideration of mobility is required. In this paper, we discuss such a quantity and investigate how well this quantity can be used by the link caching scheme in the dynamic source routing protocol to provide the adaptability to variable topologies caused by mobility through computer simulation in NS-2.

In scenarios where wireless ad hoc networks are deployed, sometimes it would be desirable that ad hoc nodes can communicate with servers in wired networks to upload or download data. In these cases TCP connections will span both wireless ad hoc and wired domains. However, TCP often faces severe unfairness in this type of connection scenario, which forces some TCP flows to completely stop transferring any data despite all links being in good states. In this paper, we propose a simple scheduling scheme, which helps competing TCP connections to achieve fairness without much throughput loss. Simulation results show that our scheme successfully eliminates the extreme unfairness existing in above-mentioned scenarios.

REM (random exponential marking) is an active queue management algorithm that is designed to achieve a desirable equilibrium, without explicit consideration to its dynamic property especially in the presence of feedback delay. We provide sufficient conditions for REM to be locally stable around an equilibrium, in a multi-link multi-source setting, when sources have identical delay of one or two-step discrete time.

We introduce the concept of generalized stochastically bounded burstiness (gSBB) for Internet traffic, the tail distribution of whose burstiness can be bounded by a decreasing function in a function class with few restrictions. This new concept extends the concept of stochastically bounded burstiness (SBB) introduced by previous researchers to a much larger extent - while the SBB model can apply to Gaussian self-similar input processes, such as fractional Brownian motion, gSBB traffic contains non-Gaussian self-similar input processes, such as /spl alpha/-stable self-similar processes, which are not SBB in general. We develop a network calculus for gSBB traffic. We characterize gSBB traffic by the distribution of its queue size. We explore the property of sums of gSBB traffic and the relation of input and output processes. We apply this calculus to a work-conserving system shared by a number of gSBB sources, to analyze the behavior of output traffic for each source and to estimate the probabilistic bounds for delays. We expect this new calculus to be of particular interest in the implementation of services with statistical qualitative guarantees.