Link Allocation Scheduling with Service Provision Approach in Wireless Mesh Networks

Master's Thesis of Electrical-Telecommunications Engineering

Wireless mesh networks are one of the most important technologies for creating wireless networks. They are the next generation. Because these networks can provide users with a wide coverage area and high capacity with low power consumption and low cost due to less path loss and also reducing the effect of the shadowing factor, which is due to their multi-step feature. In contrast to these advantages, these networks face the problem of not being easily expandable. Because the traffic that is relayed by multiple interfaces needs more bandwidth, it will suffer more delay and therefore the quality of service will decrease. Increasing the distance of the relays in order to reduce their number will also reduce the speed of the links. An increase in the number of network users also leads to more collisions and as a result, a further decrease in throughput. An increase in the area covered by the network will result in a drop in throughput and an increase in delay due to the need for more relays. Therefore, proper efficiency in a mesh network must be achieved by solving an optimization problem that includes effective factors (such as delay, throughput, etc.). Solving this type of problem in recent years as an NP-Hard problem has attracted a lot of attention in the field of problems related to wireless mesh networks.

In this thesis, a new algorithm is presented in order to improve centralized scheduling and optimal allocation of time windows to network nodes, taking into account the reusability of the frequency space, the end-to-end delay guarantee method for the user is presented. The algorithm proposed in this research for the approximate solution of the scheduling optimization problem is based on the genetic algorithm. The proposed algorithm has the ability to adapt to different parameters (such as efficiency, fairness, etc.) based on the operator's wishes. The result of the implementation confirms the improvement of the results compared to the previous methods. 1-1 Introduction, the perspective of wireless mesh networks The excessive prevalence of the Internet in today's communication world has been such that the high-speed wired access structures are not responsive to the needs of many areas. The number of service centers Today's high-speed Internet services are very low in relation to demand. Cabling high-speed lines for all these service providers is very expensive and time-consuming. Today, new technologies [a1] have been introduced to replace these wired networks. These alternative networks are high-speed wireless networks that provide fast access to the Internet when the wired network structure is unable to meet the needs of users due to the high volume of applicants or old networks, and eliminate the additional costs associated with updating the cabling structure. Traditional wireless systems are often used for commercial purposes in places where high speed and accuracy are needed, and in personal cases or homes, cheap technology should be used. Now, technical advances have made this possible and created many opportunities for Internet service providers. Wireless Mesh Networks [1] (WMN) is one of the key and influential technologies during the next decade that will play a very important role [a2] in the future generations of wireless and mobile networks. With the help of these networks, the dream that has been in the minds of many users of various types of networks around the world for a long time is getting closer to being realized; And this dream is nothing but connecting to the network at any time, at any moment, with the utmost simplicity and the lowest cost.

These networks include mesh routers [a3] and also mesh users, in which mesh routers have the least possible mobility and form the backbone of WMN. They provide network access to both mesh users and regular users.

1.       Wireless Mesh Networks

[a1] Your comment has been applied and the edit has been made

[a2] Your comment has been applied and the word has been corrected.

[a3] Your comment has been applied

The wireless mesh network is completely compatible with the structure of the wired network, and each transmitter provides access to the users connected to it

to the Internet and will act as a part of the network structure. Network traffic will pass through multiple [a1] relays

, allowing different stations to connect even if they are out of network range. Wireless mesh networks are the most flexible and least expensive way to expand high-speed Internet services that can be used mainly for personal use.

Each wireless relay in this network is an element of the network structure and can deliver information from the wireless mesh network to its destination. This type of network eliminates the problems of obstacles in the radio environment and makes the network expandable very cheaply and conveniently, because in this structure each relay only needs to communicate with its adjacent relay. Network traffic can be redirected to another relay in the event of any obstacle, without the need for any change in the location of the central radio to communicate with remote geographical locations.

Since the coverage area of ??each access point can be expanded around obstacles, the number of access points is reduced.

Wireless mesh networks have cheap technology that can be used. They are suitable for expansion and high-speed access in remote geographical areas. RoofNet[a2] is an example of these networks. This network usually includes a number of wireless access points that are installed on the windows and roofs of houses, and the information platforms of home computers are transmitted by wire to the antennas and are transferred from one antenna to another antenna to reach an Internet gateway [1] [a3] .

In wireless mesh networks, the traffic of each SS[2] is sent by network relays to the Internet or another external network to the BS[3] It is directed

Abstract

Wireless mesh network (WMN), is promising technology for next generation networks, because it can provide extended coverage range, power consumption reduction and increased throughput simultaneously. Multi hop concept of these networks leads to less path loss and shadowing effect and so coverage range can be extended. Also these networks need little cabling engineering and this makes their setup cheaper. But one of these networks' problems is scalability issue, because relayed traffic needs more bandwidth and has less QoS (delay and jitter).

Increasing hop distances in order to reduce the number of hops decreases links' rates. Also more users equals more collisions which is equivalent to throughput degradation. Coverage range extension leads to less QoS and throughput, because it needs more hops.

According to these facts, acceptable performance of a WMN is equivalent to solving an optimization problem which takes into account mentioned factors concurrently. This is hot topic research as an NP-hard problem in WMNs nowadays. Scheduling plays an important role in providing QoS support to multimedia communications in WMNs.

In this thesis a novel algorithm based on Genetic algorithm with Spatial Reuse for improving centralized scheduling and optimal time slot allocation in Wireless Mesh Networks is presented. The proposed algorithm considers hard QoS, spatial reuse for efficient use, complete interference model and the impact of node positions in topology.