Comparative Analysis of Green Algorithm within Active Queue Management for Mobile Ad Hoc Network

The Queue Management Techniques play an important role in improving the Quality of Service in Mobile Ad-hoc Networks. The need of congestion control and Queue management is inevitable in Mobile Ad-hoc Network. Understanding and analyzing the latest AQM Techniques give an insight to improving throughput, packet loss rate, average end to end delay and other parameters that severely affect the performance. The performance of mobile ad-hoc networks is evaluated by varying the Channel Error Rate, Bandwidth and Delay at different fragment sizes of different traffic flows. The proposed method performs the network analysis of new Routing protocols. The work proposes to evaluate various Active Queue Management Techniques in a multi traffic environment i.e., sources generating both TCP and UDP traffic classes. The UDP is propagated via CBR and TCP is propagated via FTP. The implementation of the proposed queue management technique has been done in NS-2. In the proposed work we have make NS-2 compatible with GREEN by integrating source code into the ns-2 installation files and using various AQMs in comparison with GREEN to calculate the throughput at various Bandwidths across network. Using various AQMs in comparison with GREEN we have calculated the throughput at various Delays across the network and we have also calculated the throughput at various channels Error Rate across the network. The proposed method has been compared with other primitive AQM techniques so that a repository of best available results can be obtained for design and research purposes. The proposed method has outperformed the existing queue management techniques in number of scenarios.


INTRODUCTION
Mobile Ad hoc Networks (MANETs) are self organizing networks formed by a collection of autonomous mobile wireless nodes without the support of any fixed infrastructure or central administration. The remote nodes in MANETs communicate directly with the help of the intermediate nodes which operate as routers. As MANETs eliminate the cost and time of infrastructure deployment, setup, and administration, they have numerous applications, especially in the military and emergency services. Although MANETs have unique applications, realizing them is difficult because of several intrinsic constraints. Unlike their wired counterparts, mobile devices usually have limited resources (battery power, memory, CPU) which limit their transmission range. In addition to this, wireless networks have limited bandwidth and are more prone to errors than wired networks which further impose limits on the amount of data that can be sent. Hence, in order to conserve the limited resource, it is highly desirable that transmission should be as efficient as possible (minimal losses and retransmissions).

Fig 1: Mobile Ad hoc Network
The mobile network nodes today use Drop tail queue management where packets are discarded when the queue is full. When multiple packets are dropped due to queue overflow, multiple TCP flows (to which these packets belong to) back off. However, as there is a considerable amount of

IJTSRD23950
International Journal of Trend in Scientific Research and Development (IJTSRD) @ www.ijtsrd.com eISSN: 2456-6470 @ IJTSRD | Unique Paper ID -IJTSRD23950 | Volume -3 | Issue -4 | May-Jun 2019 Page: 936 delay between the packet drop at the queue and the notification at the sender, large number of packets may be dropped as the sender continues sending at its current rate. Multiple flows backing off results in under utilization of the queue (no congestion). When these flows sense this (no congestion), they start increasing their sending rates as a result of which after some time the queue overflows again. This phenomenon of alternating periods of under utilization and over utilization is caused by the Drop Tail queue management policy and is called 'Global Synchronisation.
Active Queue Management (AQM) 1) Random Early Detection (RED) was proposed as a solution to the 'Global Synchronization' problem and this opened up a new area of research called Active Queue Management (AQM). The key aims of AQM are to prevent global synchronization, reduce queuing delays and improve resource utilization. AQM has been extensively studied in the context of wired networks (to quote a few). A few approaches have also studied the implications of employing AQM in the fixed wireless side (WLANs in particular) [3] with encouraging results. However, to the best of our knowledge no approach has ever studied the implications of AQM in mobile environments. If benefits in terms of improved performance (reduced packet loss) are to be reaped by deploying AQM in MANETs, it is imperative that any such scheme should be Lightweight, Requires less memory, Processing power and energy. There are few AQM Techniques used in MANETs which are REM (Random Exponential Marking) SFQ (Stochastic Fair Queuing). Random exponential marking (REM) is an attractive adaptive queue management algorithm. It uses the quantity known as 'price' to measure the congestion in a network. REM can achieve high utilization, small queue length, and low buffer overflow probability.
Fair Queuing is a queuing mechanism that is used to allow multiple packets flow to comparatively share the link capacity. Routers have multiple queues for each output line for every user. When a line as available as idle routers scans the queues through round robin and takes first packet to next queue. FQ also ensure about the maximum throughput of the network.

Literature Review Chung and Claypool et al. (2017) has discussed that in
Internet routers, active queue management (AQM) is a technique that consists in dropping or explicit congestion notification (ECN) marking packets before a router's queue is full. An Internet router typically maintains a set of queues, one per interface, that hold packets scheduled to go out on that interface. Historically, such queues use a drop-tail discipline: a packet is put onto the queue if the queue is shorter than its maximum size (measured in packets or in bytes), and dropped otherwise.

Methodology
In order to perform the analysis of AQM techniques for the performance evaluation requires a network to be designed that contains a router to router link on which the parameters can be changed. The multiclass traffic network generating both TCP and UDP traffic, topology of the network created is as follows: Fig2: Network Scenario with both TCP and UDP sources.
The network has 2 source nodes (S1 and S2), one generating TCP traffic and the other generating UDP traffic to be transferred at the destination node (d). The network parameters are changed on the link (r1-r2) as shown in the  The AQM Techniques are analyzed and the following parameters are changed in the network. In first case, the bandwidth of the main router to router link (r1-r2) is changed keeping the other network parameters unchanged. In the second scenario the delay of the main router to router link (r1-r2) is changed keeping the other network parameters unchanged. In the third scenario the error rate of the main link is changed and in the fourth scenario the packet size is changed.

Experimental Results
The simulations are performed on ubuntu 14.14. NS2 is installed with NSG and trace graph to complete the full working model to obtain the results. NSG is used to write TCL scripts from graphical user interface and trace craft is used to analyze the output trace files.