Article Preview
TopIntroduction
The exponential growth and accessibility of data in multiple forms is the main driving force for the continuous development of the communication industry. With each passing day, the ever increasing demand for smart devices, mobile multimedia services like e-healthcare, video conferencing, video surveillance, online gaming with High Definition (HD) and Ultra High Definition (UHD) Resolution video, etc. is only rising rapidly. This defines a new phase of development of mobile communications (Kamel et al., 2016). The extraordinary amount of data traffic generated by today’s user requires a fundamental change in all aspects of mobile networks. Many international forecasting agencies project that there shall be around 40 billion wireless connected Internet of Things (IoT) devices by 2025. The 5G cellular networks shall usher in an epoch with over 1Gbps connectivity, around 1mS latency, 50MHz bandwidth, etc. Ultra Dense Network (UDN) is a new paradigm shift in the direction of 5G cellular networks and realization of its true potential. Hence, UDNs are evolving as one of the core challenges and research areas of5G cellular networks that would bring in far reaching modifications in future networks (Yu et al., 2016).
In UDNs, the mobile end user clients would operate on a large number of densely deployed small cells and access nodes in their indoors like buildings, homes or in outdoor hotspot areas like airports, trains, metro/train stations, etc. Small cell networks will synchronize with macro cells, either in the same spectrum or on a dedicated carrier channel.
Figure 1. A generic UDN with mobile users/relay nodes as source/destination pairs and the network is densified with large numbers of small cells of varying sizes co-existing with macro cells.
Figure 1 shows a generic UDN with mobile end users / relay nodes as source/destination pairs with deployment of large numbers of small cells of varying sizes including micro cell, femto cell, pico cell densify the network which co-exist and synchronize with the macro cells are shown.
The primary objective of routing protocols is to select a particular path out of a number of available paths and deliver packets from source to destination. The path for traffic movement can be within the same network or between/across multiple networks. These protocols play an important role to provide seamless connectivity and uninterrupted data communication and transfer between the source and the destination. Selection of an optimum routing protocol is a prerequisite for enhanced performance, reliability and service of the network. The traditional routing processes face several critical challenges in the formation of the routing paths. The process depends upon the type of network in use, the performance metrics and the channel characteristics (Shabbir et al., 2017).
Traditional routing in dynamic wireless networks has a lot of disadvantages and is a challenging issue revolving around many factors starting from network topology that change dynamically, network failures, constraints of resources, designing of routing protocol issues, unfavoured deployment conditions, etc. From the related literature, it has been observed that the as the size of the network grows exponentially, traditional approaches fail to provide the desired results and are more error prone and time-consuming (Kamel et al., 2016; Sharma et al., 2019).
Under such backdrops, lot of research has been done in recent times to search and form a feasible path between source and destination in dynamic environments. This will result in maximizing energy conservation of a network. The key criteria being considered is to design and create routing protocols which takes into account the critical issues of maximizing network lifetime and minimizing energy consumption. Of late, research has focussed on different nature inspired algorithms and met heuristics that imitate the nature for solving various optimization problems opening a new era in computational science. Different optimization techniques have been studied and used for the formation of low cost optimized paths among different available paths. Implementation of Swarm Intelligence (SI) based algorithms has led to the development of various routing protocols for dynamic dense networks. Thus, swarm based intelligent algorithms can be a potential substitute to provide the desired results for routing in dynamic Ultra Dense Networks (UDNs) (Aghbari et al.,2020).