A Multiobjective Optimization Model for Dynamic Reconfiguration of Ring Topologies with Stochastic Load

Abstract

Multiobjective optimization, Optical wireless network, Topology control, Dynamic reconfiguration,

Interval Uncertainty-Based Robust Optimization for Convex and Non-Convex Quadratic Programs with Applications in Network Infrastructure Planning

“…A typical example of MILPs in networking planning can be a centralized topology control problem in free space optical (FSO) networking (Shim et al 2008). FSO communications are used to form a wireless backbone network in which a set of mobile base stations are connected through high data rate directional links (see http://www.cnis.umd.edu for an example of a backbone-based wireless network).…”

“…Subsequently, the PAT techniques create a new topology in accordance with the solution by steering FSO transceivers. A central node is designated to this topology control process; it acts as a network planner based on the periodically updated link and traffic states of all nodes in the network (Shim et al 2008). Gabriel et al (2008) proposed an optimization model for the dynamic reconfiguration of a bi-connected ring network topology, jointly minimizing the cost and congestion given a cost vector and traffic demands.…”

“…(3)) with a linear cost and congestion term in the objective function and with binary decision variables for link and directional flow. They realize the uncertainty in the traffic demands (r w¼ðs;dÞ ) between source (s) and destination (d) nodes in the ring network with a discrete scenario; refer to Gabriel et al (2008) for more detailed mathematical formulation of this case study. Our robust optimization method can contribute to this optimization problem in FSO networking with different realizations of the uncertainty in the traffic demands, assuming the demands exhibit variation within bounds (i.e., r w AE Δr w , where Δr w is a constant for source-destination pair of nodes ω=(s,d)).…”

Molecular inspired models for prediction and control of directional FSO/RF wireless networks

“…Reconfigure topology: if there is a better topology or an essential link cannot be retained, the network topology should go into a reconfiguration phase to try to re-gain/improve connectivity. Low complexity algorithms for dynamic topology optimization have been developed [14][15][16][17].…”

Autonomous reconfiguration and control in directional mobile ad hoc networks

“…communications cost in terms of BER or transmitted power) and the network layer cost (e.g., congestion) jointly considered in the optimization problem. Specifically in [18] we presented a multi-objective optimization model to find a ring topology which jointly minimizes total communications cost (in terms of transmitted power) and network congestion. Such an optimization problem balances these two competing objectives.…”