Improving performance in delay/disruption tolerant networks through - PDF document

Wireless Netw (2012) 18:9–31 DOI 10.1007/s11276-011-0384-1 Improving performance in delay/disruption tolerant networks through passive relay points Saeed Shahbazi • Shanika Karunasekera • Aaron Harwood Published online: 30 September 2011 � Springer Science+Business Media, LLC 2011 Abstract In this paper, we study the case of a limited indicate that our approach can achieve competitive mes- number of mobile nodes trying to communicate in a large sage delay and delivery rates. We also demonstrate several geographic area, forming a delay/disruption tolerant net- techniques for optimizing the stationary relay node place- work (DTN). In such networks the mobile nodes are ment, namely relay pruning, probability based relay dis- disconnected for significantly long periods of time. Tradi- tribution and a genetic algorithm; the genetic algorithm is tional routing protocols proposed for mobile ad hoc net- shown to provide the best solutions to this problem. works or mesh networks, which assume at least one path between each source and destination, are ineffective in Keywords Ubiquitous network connectivity � Delay/disruption tolerant networks � DTNs. One approach to improve communication is through Performance evaluation � RFID tags gossip based protocols because these protocols do not rely on a fixed path. Another approach is to control the move- ment of the mobile nodes and/or use special mobile nodes called ferry nodes. Others try to employ a fixed infra- 1 Introduction structure including stationary relay points. One scheme in stationary relay point approach is to use base stations as The number of connections present in a mobile network at relay points which need their own power supply. In this any one time is an important topological property because paper, we study a passive approach where mobile nodes such connections allow communication to take place. We deposit/retrieve messages to/ from known stationary loca- categorize mobile networks as delay/disruption tolerant tions in the geographic region. Messages are delivered network (DTN) depending on the degree to which con- from a source by being deposited at one or more locations nections are available. In DTNs, the number of nodes per that are later visited by the destination. A proposed unit area, or the node density, is small and the nodes do not implementation of our approach using read/writable pas- frequently connect. The network may remain partitioned sive Radio Frequency Identification (RFID) tags, one per into individual nodes for relatively long periods of time. point location, is considered in this work. Passive RFID DTNs arise naturally from applications such as wildlife technology is desirable because it operates wirelessly and tracking [25], vehicle-based disruption-tolerant networks without the need for attached power. Our simulation results (VDTN) [12, 31, 38], rural kiosks in developing countries [47], delay-tolerant mobile sensor network [57], and environmental monitoring including metropolitan areas S. Shahbazi ( & ) � S. Karunasekera � A. Harwood [27] and underwater [1, 40], or from fragility and failures The University of Melbourne/NICTA, 3.08, 111 Barry St., in the network itself due to disasters, jamming and noise, ICT Building, Carlton, VIC 3053, Australia and power failure. DTNs are also referred to as sparse e-mail: shahbazi@csse.unimelb.edu.au mobile networks, extreme wireless networks, or intermit- S. Karunasekera tently connected networks in the literature. In DTNs, if two e-mail: shanika@csse.unimelb.edu.au nodes are within the broadcast range of each other and the A. Harwood link between them is up then we say they are connected. e-mail: aaron@csse.unimelb.edu.au 123

10 Wireless Netw (2012) 18:9–31 In the literature, the connected link is referred to as a Stationary relay point approaches can be categorized as contact [15]. active and passive based on the type of relay nodes. If the Perur et al. [44] define sparseness of a network by relay node can initiate the communication we refer to the measuring connectivity of the network, i.e., the probability approach as an active relay point; otherwise, we refer it as a that the network graph forms a single connected compo- passive one. Further, in active approaches [6, 22, 66], nent, if it is less than 0.95, then it is referred to be as a stationary relay nodes have their own supply of power sparse network. From another perspective, we consider a while in passive approaches [49], they are powered by mobile network as a DTN for a given time interval, if the readers, i.e., mobile nodes. Also, in active approaches, the average number of contacts is less than 5% of all potential number of relays is less than the passive approaches while contacts, i.e., all pairs of nodes, over the interval. The their broadcasting range is usually bigger. details of calculating this threshold is referred to - 1. In this paper, we propose a passive stationary relay point As there is no fixed infrastructure in DTNs to deliver the based protocol to improve the delivery performance of the messages from source nodes to destination nodes, mobile DTNs. In contrast to the active approaches proposed in the nodes have to participate in routing and they have to act as literature, in our protocol, the stationary points do not a router similar to mobile ad hoc networks (MANETs). require their own power supply. Specifically, we propose However, traditional routing protocols proposed for an alternative approach where mobile devices deposit/ MANETs [24, 41–43, 48] are ineffective in DTNs as they retrieve messages to/from known point locations in the typically make an assumption that the underlying network geographic region. The point locations act like ‘‘mail- is connected. A connected network in this context means boxes‘‘. Mobile nodes are assumed to know the position of that there exists at least one (possibly multi-hop) path the mailboxes. As a mobile node moves around the region, between each pair of nodes and that exists for a long- it checks the mailboxes that it meets. Messages can be enough period of time to allow a packet to traverse it. retrieved from a mailbox and copied into another, subse- Furthermore, these protocols assume that if the path is quently met mailbox. This mechanism allows all nodes in disrupted it can be repaired or replaced in a relatively short the network to help push messages over the geographic period of time. The assumption of connectivity is clearly region and thereby expedites message delivery. In fact, in ineffective in DTNs because of the lack of instantaneous our protocol nodes never communicate messages directly end-to-end paths in such networks which prevents estab- to each other, rather they communicate messages only via lishing any routes to forward the data packets. these mailboxes which are in known places. This helps In order to overcome the lack of instantaneous end- saving energy because the transmission can be off when to-end paths in DTNs, a routing protocol can use a mobile nodes are not in the communication range of the store-and-forward paradigm. Therefore, a new class of relay nodes, which is a challenging issue in DTNs as the routing protocols, referred to as store-carry-and-forward mobile nodes are battery powered. Banerjee et al. [5] show [10, 23, 25] has emerged. This class of routing protocols that using 802.11 radio to search for contacts in a DTN exploit the mobility of the nodes in the network to forward devotes 99.5% of the total energy of mobile nodes just to the data packets by relaying packets to intermediate nodes. find other nodes which is leading to have a short network The intermediate nodes then keep the data and deliver it to lifetime. Additionally, the possible communication between the final destination or to another intermediate node. mobile nodes is, by definition of a DTN, infrequent and Therefore, the data is incrementally distributed throughout therefore its effects are negligible. Although we augment the the network, i.e., in the intermediate nodes, leading to network with a passive unconnected infrastructure, mobile facilitate the data delivery process. nodes still can make a network on the fly without a priori Recently there has been focus on augmenting DTNs connected infrastructure. with some low cost, easily deployable fixed relay nodes. A proposed implementation of our approach using pas- We refer to this emerging class of protocols as stationary sive read/writable Radio Frequency Identification (RFID) relay point approaches. In these approaches some station- tags, one per point location, is considered in this work to ary relay nodes are added to the network in order to evaluate the network performance; hence we refer to our improve connectivity [6, 22, 49, 66]. This class of proto- protocol as a Tag-based Routing (TBR) Protocol. RFID cols can increase contact opportunities among mobile technology is desirable because it operates wirelessly and nodes, consequently improving DTNs performance. For without the need for attached power. This makes its example, Banerjee et al. [5] show experimental results deployment relatively easy and sustainable. It would be collected from the UMassDieselNet DTN [10] that adding equally valid to consider the results of our work on a net- a fixed relay node, called throwbox, to the network work where the point locations are wireless base stations improves the packet delivery by 37% and reduces the (with no connections between the base stations), if such an message delivery latency by at least 10%. infrastructure was feasible for the application. The increased 123