Taguchi Approach for Recognizing the Most Influential Factor in Improving IEEE 802.15.4 Performance

WPAN using IEEE 802.15.4 protcol operated in non beaconing mode an attempt has made to find out the most significant factor of defacto parameters of IEEE 802.15.4 to enhance the performance by minimizing energy consumption and to enhance the network lifetime of the Wireless Personal Area Networks(WPAN). The factors include Buffer size, Beacon Interval, Back-off-transmission has an ideal impact on QoS metrics in IEEE 802.15.4 protocol. A Design of experiments have been simulated to an optimum level using the taguchi approach. The experimental results from the taguchi approach reveales that Back-of-transmission as the most significant factor for IEEE 802.15.4 in minimizing the power consumption in the WPAN.

NetXplain: Real-time explainability of graph neural networks applied to networking

Recent advancements in Deep Learning (DL) have revolutionized the way we can efficiently tackle complex optimization problems. However, existing DL-based solutions are often considered as black boxes with high inner complexity. As a result, there is still certain skepticism among the networking industry about their practical viability to operate data networks. In this context, explainability techniques have recently emerged to unveil why DL models make each decision. This paper focuses on the explainability of Graph Neural Networks (GNNs) applied to networking. GNNs are a novel DL family with unique properties to generalize over graphs. As a result, they have shown unprecedented performance to solve complex network optimization problems. This paper presents NetXplain, a novel real-time explainability solution that uses a GNN to interpret the output produced by another GNN. In the evaluation, we apply the proposed explainability method to RouteNet, a GNN model that predicts end-to-end QoS metrics in networks. We show that NetXplain operates more than 3 orders of magnitude faster than state-of-the-art explainability solutions when applied to networks up to 24 nodes, which makes it compatible with real-time applications; while demonstrating strong capabilities to generalize to network scenarios not seen during training.

QoS Metrics Analysis of Routing Protocols in MANET using FIS and ANOVA Test with Varying Speed of Nodes

In this paper, we analyzed AODV,DSR and DSDV routing protocol using different parameter of QoS metrics such as packet delivery ratio(PDR), Normalize Routing overhead, and Energy. The goal of this work is to determine if there is a difference between routing protocol performance when operating in a large-area MANET with high-speed mobile nodes. After the simulations, we will use Fuzzy Infurrence System to plot the performance metric. After that we use one-way ANOVA tools for that the result is correct or not. We use Matlab for simulation work. The comparison analysis will be carrying out about these protocols and in the last the conclusion will be presented, that which routing protocol is the best one for mobile ad hoc networks

Proportional interactive buffer management and scheduling

The proportional differentiated services model has been gaining attention in recent years. Many effective algorithms have been proposed to provide delay and packet loss differentiated services to different traffic classes according to their pre-defined constraints. In this article, we propose three novel QoS mechanisms to enhance the control of class-based QoS. First, we introduce a Scheduling to Dropping feedback mechanism, which is called SDF model. The main usage of the SDF is to provide dynamic trade-off between different basic QoS metrics. SDF provide differentiated services based on both proportionality constraints and absolute constraints. We also introduces an additional feature called Adaptive Safety margin (ASM). Usually, The Earlier Due Date (EDD) [7] employs a fixed safety margin to pre-assign more bandwith for time-critical traffics. ASM can adaptively changes this margin according to the queuing length or packet head-drop rate of the time-critical traffics when the congestion happened. By employing ASM with SDF model, it is found that the total packet drop can greatly be reduced. Finally, we introduce two sorting methods, called one-sort and complete-sort scheme, for the queuing systems in the end-to-end QoS environment. By sorting the order of the packets in the queue based on their deadlines, we can reduce the probability of end-to-end deadline violation.