Advances on Routing Protocols

Abstract

In this modern era of technology, where almost nothing can be achieved without communication. Therefore, it’s important that the information transferred over the internet reaches its destination without any major delays and/or corruption. A simple conversation in human language can be understood by us because we are self-aware and sentient, which isn’t the same case when it comes to computers as they can’t do anything by themselves. We must first define certain set of rules of how the data should be efficiently transferred over in machine language.

Introduction

The Internet is a vast global interconnection of many computers which have many routes between each other which are then again interconnected by routers in between. It is the router’s job to choose the most optimal route to send data towards the destination, that is much shorter in distance and is easier to send. The RPs uses certain software or routing algorithms which defines how the data is transferred over a given network. Initially, when connected, the routers exchange information between neighbours that are in the immediate vicinity. This way they gather data about the general topology of the network. Later when the most optimal path is down temporarily either due to a software/hardware malfunction, the router can still choose the second-best route and so on to send the data without any problems.

Literature Review

Routing Information Protocol (RIP) was the first RP ever created way back in the 1980s and then Interior Gateway Routing Protocol (IGRP), Open Shortest Path First (OSPF), and Intermediate System to Intermediate System (IS-IS) protocols soon followed. These are the examples of Interior Gateway Protocols (IGP) which takes place inside the networks that you administrate. Outside this network, it is handled by the Exterior Gateway Protocol (EGP), example: Border Gateway Protocol (BGP). Other RPs were soon implemented which were designed specifically for certain applications such as: Wireless Sensor Networks (WSN) [1] and Underwater Sensor Networks (UWSN)

It’s clear that based on the applications mentioned above, we can say that the routing protocols can be further implemented and designed for the future applications that demand its need. We will further discuss about them in this report below.

Routing Protocols Applications

Since the recent IEEE 802. 15. 4 standard which led to the rapid growth of WSN. A WSN mostly consists of many sensor nodes of minute power and cost are installed over a region of interest. They carry out simple tasks such as communicating with each other over short distances, wirelessly in accomplishing a specific common task, for example: Environmental monitoring, etc. In such cases battery life is a crucial factor to carry out tasks for a long duration of time, hence, some power-efficient routing protocols were designed. Low throughput affects wireless networks, but the implementation of wireless network coding improves it. From the graph above, we observe that it uses a lot of energy in 3 out of 4 states. We need to optimise it in such a way that the energy is conserved to increase the overall life of a sensor node using the new efficient routing protocols.

Classification of WSN and Routing Protocols used

• Opportunistic Routing reduces redundant transmissions thereby increasing general reliability and efficiency.
• Cross-Layer Routing optimizes the energy consumption between communicating layers to achieve optimal performance.
• Cooperative Routing comes under the above Routing which improves spectral efficiency and general transmitting capacity.
• Biologically Inspired Optimal Routing protocol’s principle were inspired from insects such as Ants and Bees which are generally used in simulations.
• Mobile Sink regulates the power depletion between nodes thus avoiding the formation of energy holes from reckless energy consumption.
• Mobile Sink and Source It manages both Tracking information and routing tables.
• Multiple Mobile Sinks deals with independent or isolated “sensor islands” which cannot move through.

We’ve surveyed the important RPs to understand the power conserved routing for WSN. Where we categorized them into Homogenous and Heterogenous WSN in which they are further divided into Static and Mobile ones. We’ve also discussed about the protocol design and the applications.

UWSN are useful for aquatic applications such as military, disaster prediction, resource tracking, monitoring pollution and marine life, etc. The sensors in this network are installed at various depths to gather data and send it to the destination. It could be a sink or a collective group of sinks. UWSN use acoustic waves where the speed is around 1500 m/s in water which introduces large propagation delays.

When a ship or entity of such size is near a UWSN, it could potentially disrupt the communication between these nodes, which leads to the formation of the void area and it moves along with the obstacle. These void areas occur in shallow depths.

Geographical Dflood (GDflood) protocol utilizes the position of node data to minimize the number of relays that occurs in the process of forwarding. Using this data, the nodes that are closer within destination are involved in the forwarding process. A GDflood, just like a simple network consists of a source address, a destination address, sequence number and hop count. Before actual transmission of a packet, the node first calculates the distance D between itself and the destination [4]. Further ahead the distance is quantized into hop counts.

Network Coding Dflood encodes the incoming packets into either one or more than one output packets instead of the usual store and forward approach. This way, the original data is shared among the encoded packets and the destination has higher chance to receive at least a piece of info instead of the redundant data of the same packet. Void-Handling technique increases the nodes’ density thereby reducing void occurrence to some extent. Flooding involves more nodes to deliver the packets which further increases congestion and wastage of network resources. Hybrid technique involves complexity but improves efficiency by using a variety of void handling techniques.

We’ve discussed that our first idea is to determine the node position in the process of relaying information. By doing so we can stop nodes from involving that are farther away from the destination.

Conclusion and Open Issues

We will be now concluding this report by discussing some of the open issues till date and how they could be improved. We have discussed RPs for homogenous type are more widely discussed over heterogenous and compared to static WSN and mobile WSN, the latter has more advantages for real-time guaranteed delivery along with energy-conservation, large coverage area and efficiency but comes with a cost. Using a reliable routing metric is quite important which should measure capability and overhead routing in WSNs. Heterogeneity requires furthermore study as it’s complex and deployment cost is high as of now. The nodes in WSN are vulnerable for attacks where secure routing require further planning in the future. The current QoS metrics doesn’t have good balancing between guaranteed packet delivery and energy-efficiency, which requires special investigation in this field. Synthetic benchmarks of the newer protocols show promise when compared to the older ones, but they tend to yield additional problems in real-case scenarios that can be foreseen if there are stronger benchmarks studies.

The RPs that use multisink triumph over the RPs that use only a single sink, however doing so introduces security holes which needs attention. In our next application of UWSN: we can further optimize GDflood for a better performance. The less original data content is re-transmitted the second time thereby delivering all the data. This results in reduction of end to end delay and total energy consumption. It means that we could achieve better results with just a single transmission. The other idea is to flood more data packets across nodes using Network Coding. Which in turn increases the chances for the destination to receive the information without any redundant data. Hybrid void handling techniques should be deployed in a new environment as it will become more lucid why a hybrid approach is necessary for UWSN. There should be a dedicated protocol that detects trapped nodes before the transmission to avoid forwarding problems. Void-Handling techniques are applied on shallow regions rather than deeper areas. It would be interesting to know if these techniques can be still relied upon at such temperatures and pressure. We can understand the current hurdles of these void-handling techniques by conducting a real test study using a real testbed. Each void-handling technique is designed for a specific purpose, having its own advantages and disadvantages. In this research report, we’ve discussed the applications where the newly designed RPs are introduced with their classification. In the first application of WSN, we have discussed about the different RPs and their different types along with their short comings and how they can be improved in the future. For our next application, we have discussed about flooding-based protocols or Dflood for UWSN. And finally, their advantages and disadvantages featuring open challenges that are yet to be investigated.