Return to search

Stochastic Resource Control in Heterogeneous Wireless Networks

In the near future, demand for Heterogeneous Wireless Networking (HWN) is expected to increase. HWNs are formed by integration of different communication technologies, for example the integration of wireless LAN and cellular networks, to support mobile users. QoS provisioning in these networks is a challenging issue given the diversity in wireless technologies and the existence of mobile users with different communication requirements. In this thesis, we consider optimal resource planning and dynamic resource management for HWNs.

In the first part of this thesis, we examine the optimal deployment of such networks. We propose a mobility-aware network planning optimization in which the objective is to minimize the rate of upward vertical handovers while maximizing the total number of users accommodated by the network. The optimal placement of Access Points (AP) with respect to these two objectives is formulated as an integer programming problem. Our results show that considering the mobility pattern in the planning phase of network deployment can significantly improve infrastructure performance.

In the second part, we investigate optimal admission control policies employed in maintaining QoS in HWNs. Here we consider two cases: integration of cellular overlay with a single WLAN AP, and integration with a WLAN mesh network. A decision theoretic framework for the problem is derived using a dynamic programming formulation.

In the case of single WLAN AP and cellular overlay, we prove that for this two-tier wireless network architecture, the optimal policy has a two-dimensional threshold structure. Furthermore, this structural result is used to design two computationally efficient algorithms, Structured Value Iteration and Structured Update Value Iteration.
These algorithms can be used to determine the optimal policy in terms of thresholds. Although the first one is closer in its operation to the conventional Value Iteration algorithm, the second one has a significantly lower complexity.

In the second case where the underlay is a complex WLAN mesh network, we develop a Partially Observable Markov-Modulated Poisson Process (PO-MMPP) traffic model to characterize the overflow traffic from the underlaying mesh to the overlay. This model captures the burstiness of the overflow traffic under the imperfect observability of the mesh network states. Then, by modeling the overlay network as a controlled PO-MMPP/M/C/C queueing system and obtaining structured decision theoretic results, it is shown that the optimal control policy for this class of HWNs can be characterized as monotonic \emph{threshold curves}. Moreover, these results are used to design a computationally efficient algorithm to determine the optimal policy in terms of thresholds.
Extensive numerical observations suggest that, in both cases and for all practical parameter sets, the algorithms converge to the overall optimal policy. Additionally, numerical results show that the proposed algorithms are efficient in terms of time-complexity and in achieving optimal performance by significantly reducing the probability of
dropped and blocked calls.

Identiferoai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:OTU.1807/32708
Date21 August 2012
CreatorsFarbod, Amin
ContributorsLiang, Ben
Source SetsLibrary and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada
Languageen_ca
Detected LanguageEnglish
TypeThesis

Page generated in 0.0017 seconds