Topology Control (TC) aims at tuning the topology of highly dynamic networks to provide better control over network resources and to increase the efficiency of communication. Recently, many TC protocols have been proposed. The protocols are designed for preserving connectivity, minimizing energy consumption, maximizing the overall network coverage or network capacity. Each TC protocol makes different assumptions about the network topology, environment detection resources, and control capacities. This circumstance makes it extremely difficult to comprehend the role and purpose of each protocol. To tackle this situation, a taxonomy for TC protocols is presented throughout this paper. Additionally, some TC protocols are classified based upon this taxonomy.
Deep Dive into A taxonomic Approach to Topology Control in Ad-hoc and Wireless Networks.
Topology Control (TC) aims at tuning the topology of highly dynamic networks to provide better control over network resources and to increase the efficiency of communication. Recently, many TC protocols have been proposed. The protocols are designed for preserving connectivity, minimizing energy consumption, maximizing the overall network coverage or network capacity. Each TC protocol makes different assumptions about the network topology, environment detection resources, and control capacities. This circumstance makes it extremely difficult to comprehend the role and purpose of each protocol. To tackle this situation, a taxonomy for TC protocols is presented throughout this paper. Additionally, some TC protocols are classified based upon this taxonomy.
Multi-hop ad-hoc networks as well as wireless sensor networks are composed of a set of devices that communicate with each other over a wireless medium. Such networks can be formed spontaneously whenever devices are in transmission range. Joining and leaving of nodes occurs dynamically, particularly in the presence of mobility. Two mobile devices out of communication range can use intermediary devices for relaying packets. Such infrastructureless networks are very flexible and easy to deploy in different settings [1]. Potential applications of ad-hoc networks can be found in traffic scenarios, ubiquitous Internet access, collaborative work, and many more.
Those networks, however, suffer from unpredictable factors such as battery lifetime, interference, noise and-in particular-the dynamics in terms of potential mobile nodes or temporary link failures. An inappropriate topology can reduce the impact of network capacity by limiting spatial reuse of the communication channel and decrease network robustness.
Nevertheless, Topology Control picks up controllable factors such as the transmission range to tune the topology in order to get a more efficient communication network while inducing minimal additional overhead [2].
Recently, several TC protocols have been proposed. Designed for preserving connectivity, minimizing energy consumption, maximizing the overall network coverage or network capacity, each TC protocol makes different assumptions about the network topology, environment detection resources, and control capacities using different approaches, different models and having very different optimization objectives. This circumstance makes it extremely difficult to comprehend the role and purpose of each protocol. Due to the high number of characteristics, it has become a challenge to compare the performance of two or more protocols on a reasonable basis.
To overcome this difficult situation, we introduce a taxonomy for TC protocols. A taxonomy is a system for naming and organizing things (objects) into groups that share similar characteristics. A taxonomy helps organize knowledge that can be accessed, navigated, searched, discovered, and compared easier.
Navigating. Traversing the hierarchy of TC protocols; getting an overview of existing approaches in terms of quantity as well as quality.
Searching. Looking up for specific characteristics; applying querying mechanisms on the hierarchy.
Comparing. Elements that fit in one taxonomy are clearly comparable on the basis of the criteria of that taxonomy. Advantages and drawbacks of different approaches become visible by their lateral relationship Hence, a taxonomy is not a theoretical framework, but a method to make knowledge appropriately applicable.
Discovering. By spontaneously browsing through TC protocols something unexpected might be discovered. Deficiencies and shortcomings of TC protocols can be detected. Often new approaches are discovered by combining the categories of a taxonomy.
The contribution of this paper is to introduce a taxonomy for TC protocols focusing on three different purposes. The used categories or taxa are justified in a case study of TC protocols. As a proof-of-concept, a selection of the most important TC protocols is classified based upon this taxonomy.
The remainder of this paper is organized as follows. Section 2 describes taxonomic approaches of TC protocols and shows differences to our approach. Section 3 presents a case study of three TC protocols, namely LINT, LMST, and CBTC. In Sectio 4 the taxonomy is introduced, taxa are justified and described. This taxonomy is applied to the most important TC protocols in Section 5. This paper finishes with the conclusion and future work.
In this section, we describe existing taxonomic concepts and taxonomies of TC. We illustrate the differences between those approaches and ours.
One of the most detailed and applied taxonomy in the literature is introduced by Santi [3]. Santi’s taxonomic approach is incremental. That means it is built on the previous state of the art of the TC techniques classifying the most important TC techniques. Although not explicitly given, the taxa appear to be controlling (action), quality of information (view), and computing character (decentralization). On the first level, there is a distinction between homogeneous and nonhomogeneous TC. In the homogeneous case, the objective is all nodes adjust the transmission power to a minimum value where the network still remains connected. In the heterogeneous case, each node can adjust its transmission power to its own criteria. In contrast, our approach understands this distinction as part of the optimization criteria.
On the second level of this hierarchy, Santi categorizes TC techniques according to the quality of information necessary to run the protocol. With quality of information is meant the understanding that for instance location information or positions require different costs in terms of technology o
…(Full text truncated)…
This content is AI-processed based on ArXiv data.
