In computer networking, Wireless LANs play a pivotal role. As the name suggests the network connections are based on a high frequency of wireless links namely radio waves. Being wireless systems, the interconnections can be extended to certain areas of networks. Thus, these systems are broadly used in the workplace, residence, educational institutions, public areas, etc.
As Wireless LANs are accessed broadly, it’s performance should also be monitored and improved periodically. The primary purpose of this project is to compare different rate adaptation algorithms in the IEEE 802.11ac standard Wireless LANs by changing parameters and evaluating the performance of the system by metrics which is not observed in any works previously.
The network simulator ns-3 is used in this paper which is a distant event network simulator tool where it creates virtual nodes to examine real-world problems.
Rate adaptation algorithms enhance the performance of Wireless LANs. In this paper, two rate adaptation designs are used namely Minstrel, and Ideal. Both these methods are compared with a Constant Rate where no algorithm is used.
Firstly, Minstrel is one of the rate control algorithms which solely selects rates that will provide proper estimations to the best accessible throughput. It is purely based on acknowledgment feedback where the success rate of future network simulations hinges upon the previous efficacies.
An ideal algorithm is completely based on an explicit feedback system which means considering every possibility to improve the performance of the network. The ideal methodology is also known for its instantaneous feedback and rate decision [9]. In this paper, maximum performance was shown by the Ideal algorithm.
In ns-3, mobility models provide the movement of the nodes based on position, velocity, and acceleration [8]. Mobility models used in the proposed system are Random Walk2d, one type of Random Walk that is available and frequently used in transit. It is a mobility pattern lacking memory. The speed of the node currently possessed is not dependent on any characteristics that have been acquired previously.
Secondly, it is one of the mobility models where the current position is fixed once it is set and remains fixed unless explicitly reset to a new value. Thirdly, In the Random Direction2d model, nodes typically choose a path at random and move in that direction until they hit the boundary or epoch [10].
The wireless signal strength at a group of receivers for any packet being delivered by a single transmitter is calculated using the propagation loss model [11]. In the Log Distance path loss model, the path loss across the sender-to-receiver distance is exponential. It is suitable for suburban situations [12]. Quadratic route loss in free space is computed following Harald T. Friis' propagation model [13] referred to as Friis propagation loss.