M-Dimension: Multi-characteristics based routing protocol in human associated delay-tolerant networks with improved performance over one dimensional classic models

Abstract

Human associated delay-tolerant network (HDTN) is a new delay-tolerant network where mobile devices are associated with humans. It can be viewed from both their geographic and social dimensions. The combination of these different dimensions can enable us to more accurately comprehend a delay-tolerant network and consequently use this multi-dimensional information to improve overall network efficiency. Alongside the geographic dimension of the network which is concerned with geographic topology of routing, social dimensions such as social hierarchy can be used to guide the routing message to improve not only the routing efficiency for individual nodes, but also efficiency for the entire network.

We propose a multi-dimensional routing protocol (M-Dimension) for the human associated delay-tolerant network which uses the local information derived from multiple dimensions to identify a mobile node more accurately. Each dimension has a weight factor and is organized by the Distance Function to select an intermediary and applies multi-cast routing. We compare M-Dimension to existing benchmark routing protocols using the MIT Reality Dataset, a well-known benchmark dataset based on a human associated mobile network trace file. The results of our simulations show that M-Dimension has a significant increase in the average success ratio and is very competitive when End-to-End Delay of packet delivery is used in comparison to other multi-cast DTN routing protocols.

Introduction

Although human characteristics and behaviors typical of communities or social networks are constantly increasing their importance in our everyday life, modern routing protocols rarely take these aspects into high consideration, and option instead, for the more well-known physical aspects of the network itself, hence we propose a novel routing protocol, M-Dimension, which takes into extensive consideration both social characters and classic network features.

By successfully quantifying different weights for several metrics from both the physical and social dimensions in order to find the correct routes from source to destination to multi-cast messages, M-Dimension proves to have significantly smaller End-To-End Delay and greater Average Delivery Success Ratio, proving to be a faster and reliable routing protocol for any human associated delay-tolerant network (HDTN).

The human associated delay-tolerant network (HDTN) is a new trend of development in disconnected delay-tolerant network (DTN). Modern mobile devices, such as smart phones, which are now of common use, inarguably reflect social characteristics, typical of the human-to-human communication, such as frequent exchange of messages with similar peers or a mutual friend are in fact these human social behaviors to determine data communications.

The disconnected delay-tolerant network (DTN) (Fall, 2003, Jain et al., 2004, McMahon and Farrell, 2009, Fall and Farrell, 2008), is a new evolution of the mobile ad hoc network (MANET). The traditional MANET routing protocols (Sarkar and Lol, 2010, Kum et al., 2010, Mittal and Kaur, 2009) require a path between source and destination. This assumption may not be satisfactory due to the mobility characteristic of the nodes. However, in the DTNs, nodes can communicate with each other even though the route does not initially exist.

Mobility and limited resources are two main characteristics within the DTNs. The very nature of mobility makes the network unreliable, unpredictable and unstable, causing network disconnection. Additionally, due to limited memory and processing capacity, it is impossible for a single mobile node to store and process the global network information. An efficient routing scheme for the DTNs would need to be adaptable for the mobility of nodes and scalable to the size of the network. Therefore the information of the individual nodes themselves would be treated as a great source to study the whole network character.

To deliver messages in such networks, some existing algorithms, such as Epidemic (Vahdat and Becker, 2000) and PRoPHET (Lindgren et al., 2004), use flooding or partial flooding with probability formulation. There is a high possibility that any flooding may cause network congestion or high interference, and consume too many resources for processing and switching. Context aware routing protocols such as CAR (Musolesi and Mascolo, 2009) and HiBOp (2007) utilize context information such as history, battery status and rate of connectivity change to compute delivery probabilities. Unfortunately, these routing protocols suffer poor performance due to the characteristics of the DTNs or the social behavior behind nodes (Liben-Nowell et al., 2005).

These routing protocols are mainly focused on one characteristic, such as the geographic location of the individual nodes in the network. However, given the dynamic nature of DTNs, taking the other characteristics into consideration, along with the geographic dimension is a viable and beneficial option when looking into ways of increasing efficiency. As a matter of interest, when the mobile devices or nodes are associated with humans, they demonstrate some of the social aspects of the people in their communication. More specifically, the HDTN can also be examined from different points of view. For example, in Fig. 1a, a geographic topology is shown in which a mobile node can only communicate with a node in its neighboring square. The distance between two nodes in the geographic dimension is measured by the minimum number of squares between them. This topology can also be seen from the social dimension with each node's Common Interest, as shown in Fig. 1b, where the social dimension allows for more pathways linking the mobile nodes together. Traditional studies on data networks simply focus on packet transmission and their geographic topologies, but in the last few years, mobile devices have become more affordable, reliable and increasingly pervasive in our social lives. In most developed countries and increasingly in some developing countries, almost every adult carries at least one mobile device, such as a Smartphone or PDA every day. These mobile devices automatically form HDTNs (Hui et al., 2005).

Additionally it is important to note that it is human social behavior which effects the data communications. A mobile node (or person who is carrying the mobile device) in this network has a geographic position in the physical dimension, whilst also occupying a social status in the social dimension. The geographic routing scheme, such as GPSR (Karp and Kung, 2000), has been proven to be an efficient routing scheme in geographic topology in which the geographic positions are stable. However, while the physical topology in the DTNs is unstable, the social characters of mobile nodes are relatively stable for a fixed period of time. For example, a computer science student would be likely to communicate with other computer science students even though their geographic position could change from time to time because they share similar studying interests regardless of distance.

In this paper, we propose the Multi-Dimension Routing Protocol (M-Dimension) for disconnected delay-tolerant networks. M-Dimension identifies mobile nodes using their positions in both physical and social dimensions. That is, taking into account social characteristics behind a node, and adopting the greedy routing scheme by selecting the most suitable neighbors which are closest to the destination.

Our primary contribution is to demonstrate the effectiveness of the M-Dimension Routing Protocol for improving the routing performance in the HDTNs. To sum up, this paper makes the following contributions:

• M-Dimension introduces a new identification system which identifies a mobile node using its positions in several different dimensions. This allows it to more accurately locate a mobile node.

• M-Dimension introduces a new social distance concept, which is used to measure how strongly connected two nodes are in a given network.

• M-Dimension is based on multi-cast, which can avoid the flooding of packets in the network and can simultaneously maintain the same level of efficiency of packet delivery.

The rest of this paper is organized as follows. Section 2 reviews and compares existing routing protocols for DTNs and relevant social concepts. Section 3 describes the M-Dimension structure model and its components in detail. Section 4 presents the evaluation of M-Dimension alongside other benchmark protocols. Finally, Section 5 summarizes this paper and provides suggestions for future work.