Research On Resource Optimization Allocation In Wireless MIMO Mesh Networks

Wireless Mesh Networks is widely applied in the next generation of wireless network system. It’s a key enabling technology, which can provide high capacity wireless link within a wide scope of geographical area, and it has a low cost of deployment, high scalability, flexibility,and fault tolerance at the same time. In recent years, since the WMNs have the advantage of high bandwidth, easy networking, low cost, wide coverage, stable performance, etc., it has attracted a wide research and investment from academia and industry. Multiple Input Multiple Output system also has gained a lot of attention and research, because this method can increase channel capacity, thus improve transmission rate of the wireless link effectively. Thus, if the MIMO technology used in the wireless Mesh network, network performance can be improved efficiently.At the same time, the Cognitive Radio(CR) can sense the spectrum is idle or not through testing and adapting the surrounding environment, therefore significantly improve the efficiency of spectrum utilization. The CR technology is also a hot spot of research in the case of spectrum resource scarce.Under the precondition that there is a good understanding about the study status of resources optimization allocation of wireless MIMO Mesh network at home and abroad now,aiming at the resources optimization allocation problem of wireless MIMO Mesh network in different network model, this paper designs the distributed algorithm and the implementation scheme respectively. The main work of this paper includes:(1) Propose the model of resources optimization allocation problem of wireless MIMO Mesh network with fixed bandwidth. According to the flow rate conservation constraint applied in the research of wireless network, the shortcoming of wireless MIMO Mesh network that the existing of network loop is pointed out. The problem of network loop is solved by adding power weighted coefficient. After introducing the power weighted coefficient, the problem of cross layer resources optimization allocation based on combining the network utility and node energy constraint is constructed. It can be solved using the subgradient algorithm based on dual decomposition. The simulation verifies the existing of network loop. In the case of affecting the network utility slightly, the problem of network loop is solved and the power consumption of network node is reduced. They illustrate the importance and efficiency of power weighted coefficient.(2) Propose the model of resources optimization allocation problem of wireless MIMOMesh network with dynamic bandwidth. Based on(1), and taken the vary of network bandwidth into account, that is to say, viewing the bandwidth as available allocation resources, then the problem of cross layer resources optimization allocation based on combining the network utility and node energy constraint is constructed. In the event of allocating the bandwidth dynamically,the bandwidth resources can be used more efficient and the network throughput and utility are improved. But the original optimization problem in this scene is changed from the physical link layer subproblem after dual decomposition to multivariable coupling and complex convex problem. The subproblem can be solved by alternate iteration updating algorithm. Then original problem of the resources optimization allocation is solved by subgradient algorithm based on dual decomposition. Thus distributed solution is realized. The simulation verifies the convergence and efficiency of the aforementioned algorithm, and the effect of variable power weighted coefficient to network utility.(3) Propose the model of resources optimization allocation problem of MIMO Mesh network for cognitive radio. Based on(2), considering the cross layer resources optimization allocation problem of MIMO Mesh network model under the background of cognitive radio technology. The empty spaces of primary users are utilized efficiently and the efficiency of spectrum resource is improved after introducing the cognitive radio technology. But the existence of mutual interference between primary users and second users and that the second users in the case of utilizing the spectrum resource should guarantee the need of the minimum flue rate of primary users, make the optimization problem to non-convex problem with high complexity. Then the physical link layer subproblem after dual decomposition turns into non-convex problem which can get approximate feasible solution using alternate iteration updating algorithm relied on linearization. Then cross layer dual problem is solved by subgradient algorithm based on dual decomposition. At last, the feasible approximate optimal solution can be recovered using the recovery algorithm.