Content Sharing over Smartphone-Based Delay-Tolerant Networks

L. Ramya Rekha et al Int. Journal of Engineering Research and Applications www.ijera.com
ISSN : 2248-9622, Vol. 4, Issue 10( Part - 2), October 2014, pp.39-45
www.ijera.com 39 | P a g e
Content Sharing over Smartphone-Based Delay-Tolerant
Networks
1
L. Ramya Rekha 2
G. Surekha ³ Ravi Mathey
1 M,Tech Student- Vidya Jyothi Institute of Technology. Aziz nagar Hyderabad,
2 Assistant Professor of CSE Department, Vidya Jyothi Institute of Technology, Hyderabad,
3 Associate Professor and HOD of CSE Department, Vidya Jyothi Institute of Technology, Hyderabad,
Abstract—
With the growing number of smartphone end users, peer-to-peer ad hoc content giving is likely to occur often.
Thus, new articles sharing mechanisms must be developed since traditional information delivery schemes will
not be efficient with regard to content sharing due to the sporadic connectivity between smartphones on the
market. To obtain data delivery such challenging environments, researchers include proposed the employment
of store-carry-forward methodologies, in which a node stores a communication and holds it until a forwarding
prospect arises through an encounter together with other nodes. Most past works in this field have dedicated to
the conjecture of whether two nodes could encounter the other, without thinking about the place and also time
from the encounter. In this particular paper, we propose to her discover-predict-deliver as a possible efficient
articles sharing scheme for delay-tolerant touch screen phone networks. In this proposed scheme, contents are
usually shared while using the mobility information of people. Specifically, our strategy employs the mobility
understanding algorithm to spot places inside your own home and outdoor.
Keywords— Wireless Networks, Location Dependency, DTN
I. Introduction
In the last few several years, the amount of
smartphone customers has speedily increased. As
smart phone interfaces are convenient and user
friendly, users can create various types of content. On
the other hand, content sharing remains worrisome. It
requires several person actions, such as registration,
uploading to key servers, as well as searching as well
as downloading subject matter. One approach to
reduce a user’s burden would be to rely while on an
ad hoc approach to peer-to-peer articles sharing. With
this method, subject matter are in an instant
discovered as well as shared. The effectiveness of
this sharing method is determined by the productivity
of sharing and also the significance on the shared
subject matter. In this kind of paper, we mainly focus
on the productivity of articles sharing, and we
provide suggestions upon creating substantial content.
Although ad hoc networks can certainly be
constructed with smartphones since they include
various multilevel interfaces, such as Bluetooth as
well as Wi-Fi, the on the web connectivity between
smartphones is expected to be intermittent a result of
the movement styles of carriers and also the signal
propagation phenomena. To overcome this challenge,
researchers possess proposed various store carry-
forward routing schemes. In these types of schemes,
a node stores an email and carries it for just a certain
length of time until a communication possibility
arises. Local forwarding decisions are at home made
making use of utility capabilities, and multiple copies
on the same information are disseminated in parallel
to raise the delivery probability. Therefore, Delay-
Tolerant System (DTN) routing protocols achieve
better overall performance than classic mobile ad hoc
network (MANET) routing protocols. The advantage
of both DTN as well as MANET routing protocols
may be the absence on the requirement of the central
server.
For this reason, contents usually are distributed as
well as stored for the smartphones. Among your
proposed DTN routing protocols, Epidemic routing is
a basic DTN routing solution. Within Epidemic
routing, messages usually are forwarded in order to
every encountered node that will not have a copy on
the same information. This solution demonstrates the
most effective performance regarding delivery price
and latency, but it requires sufficient resources, such
as storage, bandwidth, as well as energy. The works,
that happen to be proposed pursuing Epidemic
routing, are categorised into 3 categories: learning
resource based, possibility based, as well as
prediction structured. In the 1st category, methods
employ “data mules” as message ferries that directly
produce messages in order to destinations.
Opportunity-based routing protocols work with
unscheduled as well as random encounter
RESEARCH ARTICLE OPEN ACCESS

opportunities to change messages. Within prediction-
based schemes, sophisticated electric functions were
created using a brief history of mobility, encounter
instances, and encounter rates. Each node maintains a
computer program value for any other node, this also
value continues to be updated while using time
involving contacts. A node forwards an email copy
simply to nodes which has a higher utility to the
message vacation spot. Prediction-based schemes
outperform opportunity-based routing protocols both
regarding delivery rate and delivery latency. In
addition, the schemes reduce learning resource
consumption by simply adequately choosing how
many message identical. However, a lot of the
existing prediction-based schemes focus on the
prediction of whether or not two nodes could
encounter the other down the road and not
necessarily on once they would encounter the other
again. On the other hand, Yuan et 's. use the time of
incurs to properly predict encounter opportunities.
Nevertheless, the prediction of encounter
opportunities using the encounter record of nodes is
just not optimal on account of missing connection
opportunities.
DTN routing protocols are made to deliver an
email to the destination, whereas this content sharing
device first discovers content previous to delivering it
towards destination. This two-step functioning is
challenging a result of the absence connected with
central servers. In in search of contents, prior works
employed two techniques: flooding as well as random
stroll. Most flooding-based schemes work with a
time-to-live (TTL)-based limit to regulate the spread
on the search requests. By compare, search schemes
depending on random walks avoid the massive
distributed of messages that racing creates but
nonetheless achieve trustworthiness using
probabilistic paths to realize the responder.
II. Previous Work
Wireless ad hoc networks have been traditionally
modelled as connected graphs with stable end-to-end
paths. However, for emerging wireless applications,
such as sensor networks for wildlife tracking and
MANETs operating in challenging environments,
wireless links are short-lived and end-to- end
connectivity turns out to be sporadic. Such
phenomena are prevalent in disruption tolerant
networks (DTNs), where the connection between
nodes in the network changes over time, and the
communication suffers from frequent disruptions,
making the network partially connected.
The intermittent end-to-end paths and the changing
topology make conventional MANET routing
protocols fail, as they are designed with the
assumption that the network stays connected. Routing
in DTNs is an especially challenging problem
because of the temporal scheduling element in a
dynamic topology, which is not present in traditional
MANETs. Nodes have to decide who the next hop is,
but also when to forward, as they route packets to
destinations in a store-and-carry way.
Researchers have proposed a number of broad
methods to solve the above issue. In general,
previous related works fall into three categories:
mobile resource-based, opportunity- based, and
prediction-based. In the first category, systems
employ mobile resources (data mules and mobile
agents) as message ferries. These can be directed to
pick up, move towards the next hop, and deliver
messages to implement end to end store-and-carry
message delivery. In opportunity-based schemes,
nodes forward messages during contacts that are
unscheduled or random.
For the prediction-based schemes inter-node
contacts and mobility behavior are predicted,
generally using prior contact history. The next hop
and the contact in which a message is forwarded are
selected using the predictions such that a quality of
service (QoS) metric (e.g. delay or delivery ratio) is
maximized. Most of the existing prediction-based
routing protocols focus on the prediction of whether
two nodes would have a contact in the future without
considering when the contact happens. We believe
that lack of contact timing information undermines
the contact prediction accuracy, and consequently
reduces routing performance. [1]
The most expensive routing protocol, epidemic,
forwards copies of a message to any possible node
and guarantees a maximized delivery rate. Effectively
flooding the network with every message, epidemic
is impractical because of poor scalability in large
networks. Recently, much effort has been focused on
probabilistic forwarding (or opportunistic
forwarding), which tries to reduce the number of
copies of each message while retaining a high routing
performance, i.e., a high delivery rate. Since only a
small fraction of the nodes can get copies of a
message, it is desired that these copies are forwarded
by the nodes which have a higher delivery
probability than the other nodes.
The first drawback is that a forwarding decision
based on comparing the direct forwarding qualities of
nodes i and j cannot guarantee a good forwarding. (1)
The forwarding quality of j being better than i does
not necessarily mean that j is a good forwarder. (2)
Even though the quality of j is high, i might
encounter better nodes in the near future. (3)
Similarly, even though the quality of j is lower than i,
j might still be the best forwarder that i could

encounter in the future. To rectify this drawback, we
use a comprehensive metric which reflexes not only
the direct delivery probability of a particular message
copy but also its indirect delivery probability when
the node can forward the message to other
intermediate nodes.
The second drawback is that the quality of a node
is a constant regardless of two important states of the
copy: remaining hop-count and residual lifetime.
Remaining hop-count is an important factor: a node
can be a bad 1-hop forwarder for having a low direct
delivery probability, but it can still be an excellent 2-
hop forwarder if it has a frequent contacting node
which has a high direct delivery probability. On the
other hand, residual lifetime is important because it
affects a node's direct delivery probability and also
the message's chance of being forwarded to other
high quality nodes.[2]
A different approach to significantly reduce the
overhead of epidemic routing, while still maintaining
good performance, is to distribute only a bounded
number of copies. In a manner similar to the 2-hop
scheme, a copy is handed over to a fixed number of
relays, each of which can then deliver it only directly
to the destination. Nevertheless, in many situations
where node movement is strongly correlated or
predominantly local, the performance of this scheme
deteriorates.
Despite the variety of existing approaches, the
majority of them are multi-copy ones. Furthermore,
the minority that deals with single-copy techniques
only studies direct transmission or some form of
utility-based schemes in relatively different contexts.
In this work, we perform a detailed inquiry into the
problem space of single-copy routing, and show how
to achieve competitive performance without using
multiple copies. We look into how utility functions
can be designed to fully take advantage of the “field”
of past encounters, and propose a function that is
shown to achieve up to an order of magnitude
improvement in ICMNs over existing utility
functions. Finally, we propose a novel, hybrid routing
scheme, which uses randomization when necessary to
overcome some inherent shortcomings of utility-
based forwarding. [3]
The traditional approach to map search queries to
resources is implemented by maintaining an index
about the resources. The index contains keywords
and maps them to sets of related resources. This can
be used to map a query consisting of several
keywords to a list of suggested search results.
Examples of centralized search engines using such
indices are Google, MSN, and Yahoo. The problem
with such centralized services is that it is expensive
to keep the index up-to date and the assumption is
made that the clients can reach both the index and the
locations where the contents reside. There are
approaches for managing decentralized indexes in
peer-to-peer environments, but these models often
assume good connectivity among the nodes.
Otherwise, unreliable connections may lead to a high
maintenance cost of the index.
Furthermore, not all contents of interest resides on
servers connected to the infrastructure network:
mobile users are not just consumers, but also prime
producers of contents and often eager to share their
impressions with others. Particularly in disconnected
environments, this “mobile” contents cannot easily be
indexed by centralized search engines unless
uploaded to a some server. Yet, such contents may be
especially interesting to other mobile users in
proximity, which could access it by means of ad-hoc
networking. [5]
Communities and social relationships are formed
when a set of nodes are doing the same activity in the
same location. For example, nodes with the same
home are family members, while nodes with the same
location are colleagues from work. Nodes are doing
the activities on a daily basis starting from home in
the morning. Each node is assigned a wakeup time,
which determines when the node should start from
home. This value is drawn from a normal distribution
with mean 0 and configurable standard deviation.
The node uses the same wakeup time every morning
during the whole simulation. The variance in the
wakeup time models the differences in rhythms in
real life.
At the wakeup time, nodes leave their homes, and
use different transport methods to travel to work.
Nodes travel between activities either by car or by
bus, which are both different submodels. The
working time is configurable. After the working
hours, the nodes decide, by drawing, whether they go
out for the evening activity, or return home. Again,
different submodels are used for transitions between
the locations. Different user groups have different
locations where the activities take place.[8]
The recent advances in WSNs are rapidly
expanding the range of applications they can support:
from “traditional” environmental monitoring, where a
number of nodes is scattered across an area collecting
data for a single sink, to mobile scenarios involving
multiple sinks. This happens when the entities to
monitor are animals (e.g., in farming scenarios),
people (e.g., in elderly care scenarios), or things
moving around (e.g., in logistics), while several
mobile devices (e.g., PDAs) are used as sinks, or
actuators, also acting as sinks, are involved.
Unfortunately, mobility and the presence of multiple
sinks is something that has been largely neglected by

research on WSNs so far, especially if we consider
the case of data-aware routing protocols. Indeed, one
of the main peculiarities of WSNs is the data-centric
form of communication that they usually adopt: a few
sinks (a single one in the simplest scenarios) are
interested in receiving only some specific data among
those collected by sensors, e.g., the temperature
readings that exceed some threshold. This suggests
abandoning traditional, address-based routing
protocols, to adopt a Content-Based Routing (CBR)
protocol, in which messages do not carry any explicit
address, while they are routed based on their content
and on the interests specified by nodes. This is the
solution taken by popular protocols for WSNs like
Directed Diffusion and TinyCOPS, without however
considering mobility as a key aspect. Moreover, if we
look at the typical scenarios of usage of a WSN, we
may notice that communication is not only data-
centric but it is also often context-aware. As an
example, a farmer could be interested in knowing the
activity level of “young” cattle only, while, in a
logistics application, different temperature thresholds
are critical for different goods. Encoding such
context-awareness as part of the message content and
using standard CBR to route messages is possible,
but can be inefficient, increasing message size and
communication overhead. [9]
Due to DTN dynamics, deterministic data
forwarding, either unicast and multicast, is only
guaranteed in two cases: (i) the network is flooded,
and (ii) the data forwarding process does not have
time constraint. Neither of the two cases are practical
in DTNs due to the inevitably high forwarding cost.
Thus, a more practical solution is to maximize the
data forwarding probability with a given time
constraint. From such a probabilistic perspective, the
essential difference between multicast and unicast in
DTNs is that, a selected relay for multicast is
expected to forward data to as many destinations as
possible. The cumulative probability for a relay to
forward data to multiple destinations therefore needs
to be calculated, and such calculation may require
global knowledge of social relations among
nodes.[10]
III. Proposed System
A. Content Discovery
In content discovery, most systems focus on how
to formulate queries, which depends on assumptions
about the format of the content to be discovered. A
general protocol should support many different types
of queries and content, but we abstract from the
actual matching process in order to focus on
discovering content over the network. The simplest
strategy to discover and deliver contents is Epidemic
routing. However, due to resource limitations,
Epidemic routing is often wasteful, so we need to
consider methods that limit the system resources
spent on both content discovery and delivery. Ideally,
a query should only be forwarded to neighbors that
hold matching contents or that is on the path to other
nodes having matching content. Different nodes
should return non overlapping responses to the
requester. As global knowledge or active
coordination is not an option in our scenario, each
node can only make independent forwarding
decisions. These forwarding decisions should achieve
a good tradeoff between discovery efficiency and
required resources. Similar limitations apply to
content delivery.
B. Content Delivery
When query matching content is discovered, the
content carrying node should send only a subset of
the results. This requirement is necessary to limit the
amount of resources used both locally and globally
for transmitting and storing the responses, and to
remove potential duplicates. The query originator
assigns a limit for both the number of replications
and the volume of content that should be generated.
When nodes need to forward a query message, the
limitations included in the query message are used to
make a forwarding decision. If the volume of the
content exceeds the response limit, the node needs to
select which ones to forward. For example, when AP
hotspots are huge in number, the set of AP hotspots
with the most reliable Internet speed is chosen and
forwarded to the query originator.
C. Movement Tracking
In proposed system the Activity Manager monitors
the acceleration vector of a three-axis accelerometer
and detects the motion of the user.
The motion detector function of the Activity
Manager is basically a classifier M that has two
outputs: moving or stationary. When the user is
walking, running, or moving in a vehicle, the motion
is classified as moving, whereas when the user stays
at a certain location, the motion is classified as
stationary.

D. Mobility Prediction
As location information to estimate if a node
approaches the destination of the content or diverges
from the destination, the prediction of nodes’
mobility information is essential.
In computing we first need to estimate the
predicted mobility information from time for the
duration of query lifetime per time instance. The
prediction of the required mobility information is
formulated as a posterior distribution where the past
mobility observation is used. The direct estimation of
this probability distribution is not feasible as the
complexity is huge for smartphones. The first-order
Markov chain assumption is made, with which the
probability of observing location at time is only
conditionally dependent on the location at t. Further,
we apply temporal feature extraction to reduce
observation size, that is, we extract the locations
where the user has mostly visited in the past from
time t for the duration T.
IV. Results
The concept of this paper is implemented and
different results are shown below, The proposed
paper is implemented in Java technology on a
Pentium-IV PC with 20 GB hard-disk and 256 MB
RAM with Java Environment. The propose paper’s
concepts shows efficient results and has been
efficiently tested on different Datasets. The Fig 1, Fig
2, Fig 3 and Fig 4 shows the real time results
compared.
Fig. 1 Time taken by Node to initialize by objects.
Time Instance Objects
1 17632
2 1800
3 2659
4 3137
5 2629
6 2892
Fig. 1 Time taken by Node to compute Memory
Used
Time Instance Memory Used
1 523752
2 86660
3 127072
4 146904
5 141544
6 154596

Fig. 4 Time taken by Node Communicating and Free
MEmory
Time Instance Memory Free
1 1573400
2 2010492
3 1970080
4 1950248
5 1955608
6 1942556
V. Conclusions
In this particular paper, we proposed an effective
content discussing mechanism inside smartphone-
based DTNs. We experimented with utilize why
people love today’s touch screen phones (i. e.,
availability of varied localization in addition to
communication technologies) in addition to
appropriately developed the protocol. In creating a
articles sharing criteria, we aimed at two factors: 1)
people maneuver meaningful places, and 2) this
mobility of folks is expected. Based with this
proposition, we developed any mobility mastering
and conjecture algorithm in order to compute this
utility purpose. Thus, as opposed to conventional
procedures, the proposed sharing mechanism won't
require make contact with history. We learned that
contents without a doubt have physical and
temporary validity, in addition to we proposed a
scheme by contemplating these features of articles.
VI. Future Scope
For future extent, we believe our system still
possesses room regarding improvement. Specifically,
the use of asymmetric multicore processors and
efficient sensor scheduling is usually reduce the
energy consumption connected with smartphones’
detectors. Further, since location may be the key
portion of the suggested solution, user privacy needs
to be carefully considered. We plan to address most
of these issues within our future performs.
References
[1] Q. Yuan, I. Cardei, and J. Wu, “Predict and
Relay: An Efficient Routing in Disruption-
Tolerant Networks,” Proc. 10th ACM
MobiHoc, pp. 95-104, 2009.
[2] C. Liu and J. Wu, “An Optimal Probabilistic
Forwarding Protocol in Delay Tolerant
Networks,” Proc. ACM MobiHoc, pp. 14,
2009.
[3] T. Spyropoulos, K. Psounis, and C.
Raghavendra, “Efficient Routing in
Intermittently Connected Mobile Networks:
The Single-Copy Case,” IEEE/ACM Trans.
Networking, vol. 16, no. 1, pp. 63-76, Feb.
2008.
[4] I. Cardei, C. Liu, J. Wu, and Q. Yuan, “DTN
Routing with Probabilistic Trajectory
Prediction,” Proc. Third Int’l Conf. Wireless
Algorithms, Systems, and Applications
(WASA ’08,), pp. 40-51, 2008.
[5] M. Pitkanen, T. Karkkainen, J. Greifenberg,
and J. Ott, “Searching for Content in Mobile
DTNs,” Proc. IEEE Int’l Conf. Pervasive
Computing and Comm. (PERCOM ’09), pp. 1-
10, 2009.
[6] M.C. Gonzalez, C.A. Hidalgo, and A.-L.
Barabasi, “Understanding Individual Human
Mobility Patterns,” Nature, vol. 453, pp. 779-
782, 2008.
[7] Y. Chon and H. Cha, “LifeMap: Smartphone-
Based Context Provider for Location-Based
Service,” IEEE Pervasive Computing
Magazine, vol. 10, no. 2, pp. 58-67, Apr.-June
2011.
[8] F. Ekman, A. Kera¨nen, J. Karvo, and J. Ott,
“Working Day Movement Model,” Proc. First
ACM SIGMOBILE Workshop Mobility
Models, pp. 33-40, 2008.
[9] G. Cugola and M. Migliavacca, “A Context
and Content-Based Routing Protocol for
Mobile Sensor Networks,” Proc. Sixth
European Conf. Wireless Sensor Networks
(EWSN ’09), pp. 69-85, 2009.
[10]W. Gao, Q. Li, B. Zhao, and G. Cao,
“Multicasting in Delay Tolerant Networks: A
Social Network Perspective,” Proc. 10th
ACM
MobiHoc, 2009.
[11]S. Hong, I. Rhee, S.J. Kim, K. Lee, and S.
Chong, “Routing Performance Analysis of
Human-driven Delay Tolerant Networks Using
the Truncated Lvy Walk Model,” Proc. First
ACM SIGMOBILE Workshop Mobility
Models (Mobility Models ’08), pp. 25-32,
2008.
[12]D.H. Kim, Y. Kim, D. Estrin, and M.B.
Srivastava, “SensLoc: Sensing Everyday
Places and Paths Using Less Energy,” Proc.

Eighth ACM Conf. Embedded Networked
Sensor Systems (SenSys ’10), pp. 43-56, 2010.
[13]M.B. Kjaergaard, J. Langdal, T. Godsk, and T.
Toftkjaer, “Entracked: Energy-Efficient
Robust Position Tracking for Mobile
Devices,” Proc. ACM MobiSys, pp. 221-234,
2009.
[14]W. Gao and G. Cao, “Fine-Grained Mobility
Characterization: Steady and Transient State
Behaviors,” Proc. Eleventh ACM Int’l Symp.
Mobile Ad Hoc Networking and Computing,
pp. 61-70, 2010.
[15]A.J. Nicholson and B.D. Noble,
“BreadCrumbs: Forecasting Mobile
Connectivity,” Proc. ACM MobiCom, pp. 46-
57, 2008.
Author’s Information
1. L. Ramya Rekha is M.Tech Student at Vidya
Jyothi Institute of Technology (VJIT).
2. G. Surekha is a post –graduate specialized in
computer science from Nalanda institute of engg and
technology(affliated to JNTUK),she is working as
assistant professor in CSE dept at VJIT.
3.Ravi Mathey is a post-graduate specialized in
Computer Science from BIT -Ranchi and he did
Instrumentation technology in Andhra university. He
has more than 20 years of experience in Research and
Development, and teaching Presently he is working
as a Associate Professor and HOD of CSE
Department at Vidya Jyothi Institute of Technology
(VJIT). His area of research is wireless embedded
application and image compression techniques by
using fractals and wavelets, cloud computing

Content Sharing over Smartphone-Based Delay-Tolerant Networks

More Related Content

What's hot (16)

Viewers also liked (20)

Similar to Content Sharing over Smartphone-Based Delay-Tolerant Networks (20)

Recently uploaded (20)

Content Sharing over Smartphone-Based Delay-Tolerant Networks