Computer Science Technical Report Archive

USC Computer Science Technical Reports, no. 772 (2002)

(USC DC Other)

USC Computer Science Technical Reports, no. 772 (2002)

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content 1 Contact-Based Architecture for Resource Discovery
( CARD ) in Large Scale MANets
Saurabh Garg, Priyatham Pamu, Nitin Nahata, Ahmed Helmy
University of Southern California
{ sgarg, pamu, nnahata, helmy}@usc.edu
Abstract - In this paper we propose a novel architecture, CARD , for resource discovery in large scale Mobile Ad hoc Networks
( MANets) which, may scale up to thousands of nodes and may
span wide geographical regions. Unlike previously proposed
schemes, our architecture avoids expensive mechanisms such as
global flooding as well as complex coordination between nodes to
form a hierarchy. CARD is also independent of any external
source of information such as GPS. In our architecture nodes
within a limited number of hops from each node form the
neighborhood of that node . Resources within the neighborhood
can be readily accessed with the help of a proactive scheme
within the neighborhood. For accessing resources beyond the
neighborhood , each node also maintains a few distant nodes
called contacts . Contacts help in creating a small world in the
network and provide an efficient way to query for resources
beyond the neighborhood. As the number of contacts of a node
increases, the network view ( reachability) of the node increases.
Paths to contacts are validated periodically to adapt to mobility.
We present mechanisms for contact selection and maintenance
that attempt to increase reachability while minimizing overhead.
Our simulation results show a clear trade-off between increase in
reachability on one hand, and contact selection and maintenance
overhead on the other. Our results suggest that CARD can be
configured to provide a desirable reachability distribution for
different network sizes. Comparisons with other schemes for
resource discovery, such as flooding and bordercasting , show our
architecture to be much more efficient and scalable.
I. I NTRODUCTION
Ad hoc networks are wireless networks composed of
mobile devices with limited power and transmission
range. These networks are rapidly deployable as they
neither require a wired infrastructure nor centralized
control. Because of the lack of fixed infrastructure, each
node also acts as a relay to provide communication
throughout the network. Applications of ad hoc networks
include coordination between various units (e.g., in a
battlefield), search and rescue missions, rapidly
deployable networks, and vehicular networks, among
others.
Lack of a fixed infrastructure, makes resource discovery
in ad hoc networks a challenging problem. Unlike wired
networks, mobility induces frequent route changes.
Traditional protocols proposed for resource discovery in
such environments either involve global flooding or are
based on complex hierarchy formation. While flooding
is inefficient and therefore does not scale well, hierarchy
formation involves complex coordination between nodes
and therefore may suffer significant performance
degradation due to frequent, mobility induced, changes
in network connectivity.
To overcome these limitations we propose a new
architecture based on the concept of small worlds [10]
[11] . In our architecture we adopt a hybrid approach in
which a node uses periodic updates to reach its
neighborhood within a limited number of hops, R, and
reactive querying beyond the neighborhood via contacts . Contacts act as short cuts that attempt to transform the
network into a small world by reducing the degrees of
separation. They help in providing a view of the network
beyond the neighborhood during resource discovery.
Each node maintains state for a few contacts beyond its
neighborhood. Contacts are polled periodically to
validate their presence and routes. For discovering
resources efficiently, queries are sent to the contacts that
leverage the knowledge of their neighborhood. As the
number of contacts increases, the network view
( reachability) increases. However, at the same time the
overhead involved in contact maintenance also increases.
Our simulation results show this trade-off.
Our architecture has been designed to satisfy
requirements for an efficient resource discovery scheme.
Applications like sensor networks may comprise of
thousands of nodes. Therefore the resource discovery
mechanism should be scalable. Nodes in an ad hoc
network comprise of portable devices with limited
battery power. Therefore to save power the resource
discovery mechanism should be efficient in terms of
messages transmitted. Simulation based comparisons
with flooding and bordercasting [8] [9] show our
architecture to be more efficient. Simulation results also
show that our protocol is scalable and can be configured
to provide good performance for various network sizes.
The rest of this document is organized as follows.
Section II discusses related work. Section III describes
our design requirements and provides an overview of our
2 proposed architecture, CARD , and introduces the contact
selection and maintenance algorithms. Section IV
presents thorough analysis of CARD , in terms of
reachability and overhead, and compares it to other
schemes such as flooding and bordercasting. We
conclude in Section V.
II. R ELATED WORK
Due to lack of an existing infrastructure in ad hoc
networks, resource discovery (which includes route
discovery) is a challenging problem. Most of the
protocols proposed so far can be broadly classified as:
(1) Proactive (table driven), reactive (on - demand) and
hybrid, or (2) flat and hierarchical.
Proactive schemes such as DSDV [1] , WRP [3] and
GSR [2] involve periodic updates to be flooded
throughout the network. This is a resource consuming
process especially for large-scale networks that may
scale up to thousands of nodes. Reactive schemes such
as AODV [5] and DSR [4] try to reduce the overhead
due to periodic updates by maintaining state only for the
active resources. In these schemes a search is initiated
for any new resource required. However, the search
procedure generally involves flooding through the entire
network. This is again inefficient and also leads to
increased response delays for resource discovery. The
above architectures are flat (i.e., non-hierarchical), in
which each node has an equal responsibility of relaying
traffic from other nodes. Such schemes are expensive in
terms of resources and do not scale well.
Hybrid schemes such as ZRP [9] try to combine the
benefits of both the proactive and reactive schemes.
ZRP limits the overhead of periodic updates to a limited
number of hops (zone). Resources beyond the zone are
searched in a reactive manner by sending queries
through the nodes at the edge of the zone
( bordercasting). The zone concept is similar to the
neighborhood concept in our study. However, instead of
bordercasting we use contacts. In our study, we will
show that the contact-based approach is more efficient
than bordercasting.
Hierarchical schemes, on the other hand, such as CGSR
[6] and [15] , involve election of a cluster-head, which
has greater responsibilities than other nodes. The cluster-
head is responsible for routing traffic in and out of the
cluster. Cluster-based hierarchies rely on complex
coordination and thus are susceptible to major re-
configuration due to mobility. This could lead to serious
performance degradation. Also, a cluster head may be a
single point of failure and a potential bottleneck. In our
architecture there is very simple coordination between
various nodes. This enables our architecture to adapt
gracefully to network dynamics. Each node has its own
view of the network and hence does not require major
re-configuration with mobility.
Architectures such as GLS [7] -that require knowledge
by all the nodes of a global grid mapping the whole
network- have also been proposed. GLS requires
knowledge of geographic location (obtained from an
external source such as GPS). Our architecture is
independent of any external information such as GPS.
In [12] a framework was proposed for multicast in large-
scale ad hoc networks that introduced the concept of
contacts. However, only high level concepts were
proposed and no detailed algorithms or evaluations were
presented for contact selection and maintenance. Our
work here fits nicely in that framework. Also, in [13] the
relationship between small worlds and wireless networks
was established. In this paper, we build upon that
relationship with small worlds. Furthermore, a mobility-
assisted scheme was proposed for contact selection in
that study. Although such scheme could be very
efficient, it depends heavily on mobility, and may not be
suitable for static sensor networks. The algorithms
presented in this paper may complement those proposed
in the above work, and may be integrated with them.
III. C ARD ARCHITECTURAL OVERVIEW
A. Design Requirements
The requirements for a resource discovery mechanism
for large-scale Ad hoc networks that motivate the design
of our architecture include:
(a) Scalability : Applications of large-scale ad hoc
networks involve military and sensor network
environments that may span wide geographical areas and
may comprise of thousands of nodes. Therefore the
resource discovery mechanism should be scalable in
terms of control overhead with increase in network size.
(b) Efficiency : Nodes in an ad hoc network comprise of
portable devices with limited battery power. Therefore,
resource discovery mechanisms should be efficient.
(c) Robustness : The mechanism should be robust to
handle frequent link failures due to mobility.
(d) Decentralized operation : For the network to be
rapidly deployable, it should not require any centralized
control.
(e) No location (geographic) information : Rapid
deployability and self-configurability require that the
3 network mechanism should be independent of any
information that requires an external source.
B. Definitions
• Neighborhood: All nodes within a particular number of
hops ( R ) from the source node. R is the radius of the
neighborhood.
• Edge nodes: All nodes at a distance of R hops from the
source node.
• Maximum contact distance (r): This is the maximum
distance (in hops) from the source within which a
contact is selected.
• Overlap: Overlap between nodes represents number of
common nodes between their neighborhoods.
• Number of Contacts ( NoC): NoC specifies the value of
the maximum number of contacts to be searched for each
source node. The actual number of contacts chosen is
usually less than this value. This is due to the fact that
for a particular value of R and r , there is only a limited
region available for choosing contacts. Once this region
has been covered by neighborhoods of the chosen
contacts, choosing more contacts in the same region is
not possible, as their neighborhoods would overlap with
the neighborhoods of the already chosen contacts. This
is according to our policy to minimize overlap.
• Depth of search (D): D specifies the levels of contacts
(i.e., contacts of contacts) queried by a source.
• Reachability: The number of nodes that can be reached
by a source node is termed as the reachability of the
source node. This includes the nodes within the
neighborhood that can be reached directly and the nodes
that lie in the contacts neighborhood and are therefore
reachable through the contact(s), etc.
• Overhead: Overhead is defined in terms of the control
messages generated by a mechanism. For example,
contact selection, maintenance or query. Procedures for
these mechanisms are described in the next sections.
C. Mechanism Description
Our mechanism employs a hybrid of proactive and
reactive approaches for resource discovery. All nodes
within R hops from a node form the node’s
neighborhood. Each node proactively (using a protocol
such as DSDV [1] ) maintains state for all the nodes in its
neighborhood. Therefore a node has complete
knowledge of all the nodes (resources) within its
neighborhood. Apart from the neighborhood knowledge,
each node also maintains state for a few nodes which lie
outside the neighborhood. These nodes serve as contacts
for accessing resources beyond the neighborhood.
Contacts are selected and maintained using the
mechanisms described below.
1. Contact Selection Procedure
(1) Each node (source node s ) sends a Contact Selection
Query ( CSQ ) through each of its edge node, one at a
time.
(2) An edge node receiving a contact selection query
forwards CSQ to a randomly chosen neighbor ( X ).
(3) A node receiving a CSQ decides whether or not to be
a contact for the particular source. This decision is made
using either a probabilistic method ( PM ) or edge method
( EM ). These methods are described later in this section.
(4) After using either procedure PM or EM for deciding
whether to be a contact, if the node receiving the CSQ
does not choose to be the contact, it forwards the query
to one of its randomly chosen neighbor (excluding the
one from which CSQ was received).
(5) The CSQ traverses in a depth-first manner until a
contact is chosen or it reaches a node at a distance of r hops from the source. If a contact is still not chosen (due
to overlap), the query backtracks to the previous node,
which forwards it to another randomly chosen neighbor.
(6) When a contact is selected, the path to the contact is
returned and stored at the source node.
2. Contact Selection Methods
We discuss and compare two different methods or
protocols for contact selection: (a) the probabilistic
method ( PM ), and (b) the edge method ( EM ).
(a) Probabilistic Method (PM):
Contacts help to increase a node’s view (or reachability)
of the network beyond its own neighborhood. To
achieve maximum increase in reachability, the
neighborhoods of any source and its contacts should be
disjoint/separated, i.e. , there should be no overlap
between the neighborhood of the source and the
neighborhood of any of its contacts. The neighborhoods
of different contacts of the same source should also be
non-overlapping, to ensure maximum increase in
reachability. To achieve this, the CSQ contains the
following information: ( i ) ID of the source node, ( ii ) a
list of already chosen contacts of the source node
( Contact_List ; typically small of ~5 IDs ) , and ( iii) the
hop count, d.
This information is used as follows. When a node X receives a CSQ , it first checks if the source lies within its
neighborhood. This check is easily performed since each
4 node has complete knowledge of its neighborhood. So a
node knows the IDs of all the other nodes in its
neighborhood. X also checks if its neighborhood
contains any of the node IDs contained in the
Contact_List.
If neither the source nor any of its already selected
contacts lie in the neighborhood of X , X probabilistically
chooses itself as the contact. This probability ( P ) of
choosing to be a contact is defined as follows:
P = (d – R)/(r – R) -- (1)
Here d is the distance of X from the source node. d is
included in the CSQ as hop count. From the above
equation, when d = R, P = 0. When d = r, P = 1. This
aims to select the contacts between R and r hops from
the source. This has been formulated to provide
maximum increase in reachablility with the addition of
each new contact. However there can still be a case
where equation ( 1 ) does not provide the maximum
benefit of adding a contact. This case is shown in the
fig.1 where s is the source node and c is the contact.
c s c s R
R
R
R
(a) overlap (b) non-overlap
Fig. 1 Overlap in (a) due to the use of P In this figure although the distance between the source
and contact is greater than R hops, there is still an
overlap between the two neighborhoods. Such a situation
will arise whenever a node within R hops from the edge
node becomes the contact. To prevent this situation,
equation ( 1 ) is modified to:
P = (d –2R)/(r –2R) --(2)
In this equation P=0 when d=2R and P=1 when d=r,
i.e., the contact are chosen only between 2R and r hops
from the source.
Fig. 2 explains the contact selection procedure with an
example. In the above figure R =3 and r =6. Nodes a, b, c,
d are the edge nodes for source node s . Node s sends a
Contact Selection Query ( CSQ ) through a . Node a randomly chooses one of its neighbors e and forwards
the query to that node. Node e calculates the probability
P according to equation (1) . If the probability of being
the contact failed at e , it forwards the query to one of its
neighbors, f (chosen randomly). Node f again forwards
the query to g . As g is at r hops from s , the probability P at g is 1. However, g still cannot become a contact for s as there already exists another contact h (which was
selected through a previous query through another edge
node d ) in the neighborhood of g . So g returns the query
to f ( backtracking ). Node f then forwards the query to
another neighbor.
f e a s d c h g R backtracking
b path to h
( previously selected contact)
Fig. 2 Selecting contacts in the P method
(b) Edge Method (EM):
Even with equation ( 2 ) the probabilistic method can
result in a situation where there is some overlap between
the contact and the source neighborhoods. This is
possible due to the fact that the nodes do not have any
sense of direction. Therefore, it is possible that a contact
may be selected at a location where the CSQ has traveled
more than 2R hops in one direction, but the contact may
in fact be closer than 2R hops from the source. More
seriously, the probabilistic method for contact selection
can be expensive in terms of the amount of traffic
generated by the CSQ . This is due to the extra traffic
generated due to backtracking, and lost opportunities
when the probability fails, even when there is no
overlap. To reduce the possibility of such a situation,
probability equations ( 1 ) and ( 2 ) are eliminated. The
probability equations were formulated to have a higher
possibility of choosing the contact that lies either
between R and r hops (equation 1 ) or between 2R and r hops (equation 2 ). To maintain this non-overlapping
property without the probability equations, the contact
selection procedure is modified as follows.
5 The list of all edge nodes ( Edge_List ) of the source is
added to the CSQ . Also, the query and source IDs are
included to prevent looping. On receiving a CSQ , apart
from checking for overlap with the source neighborhood
and the neighborhoods of all the already selected
contacts ( Contact_List ), the receiving node also checks
for overlap with the neighborhoods of any of the nodes
on the Edge_List as well. Since any node that lies at a
distance of less than R hops from the edge will have an
overlapping neighborhood with the source’s
neighborhood, checking for non-overlap with the edges
ensures that a contact is chosen between 2R and r hops
only. This eliminates the possibility of an overlap due to
the lack of direction. Fig. 3 and Fig. 4 show a
comparison of the probabilistic and edge methods. As
can be seen from Fig. 3 the reachability saturates in both
PM and EM. However the saturation occurs much earlier
in the case of probabilistic method. Also as compared to
EM , the reachability achieved is less for PM , for the
same values of NoC . Fig. 4 shows the backtracking
overhead for PM and EM . Due to the reasons explained
earlier, the overhead is significantly reduced for EM . 0 10
20
30
40
50
60
1 2 3 4 5 6 7 8 9 Number of Contacts
Reachability (%) ( 2 ) E M
( 1 ) P M
Fig. 3. Reachability for (1) PM and (2) EM
0 500
1000
1500
2000
2500
3000
3500
4000
4500
5000
1 2 3 4 5 Number of Contacts
Back-tracking/node ( 1 ) P M
( 2 ) E M
Fig. 4. Overhead for (1) PM and (2) EM
(Shown for 500 nodes, 710mx710m, tx range=50m, R=3, r=20, D=1 ) 3. Contact Maintenance Procedure
Node mobility may cause the path to a contact to
change. Therefore a node needs to keep track of its
contacts and their paths. This is done using periodic
polling of the contacts as follows.
(1) Each node periodically sends validation messages to
each of its contacts. These validation messages contain
the source path to the contact.
(2) Each node on the path that receives the validation
message checks if the next hop in the source path is a of
directly-connected neighbors. If so, it forwards the
validation message to the next hop node. If the next hop
is missing, the node tries to salvage the path using local recovery , discussed later in this subsection.
(3) If a path cannot be salvaged using local recovery, the
contact is considered to be lost.
(4) If the path to a contact is validated but the number of
hops to the contact does not lie between 2R and r , the
contact is considered to be lost.
(5) After validating all the contacts, if the number of
contacts left is less than the specified NoC , new contact
selection is initiated as described earlier.
The local recovery mechanism is performed as follows.
Assuming reasonable values of node velocities and
validation frequency, there is a high probability that if a
node has moved out of a contact path, it is still within
the neighborhood of the previous hop in the path. Even
in the case when a node is completely lost (because it
has moved out of the neighborhood of the previous hop),
some other node further down the path might have
moved into the neighborhood of the previous node.
Local recovery takes advantage of these cases to recover
from changes in the path, without having to initiate new
searches from the source. Thus local recovery provides
an efficient mechanism for validating contacts and
recovering from changes in the contact paths. If the next
hop in a source path is missing, the node that received
the validation message looks for the next hop in its
neighborhood routing table. If the next hop is in the
neighborhood, the source path is updated and the
validation message is forwarded to the next hop. If the
lookup for the next hop fails, lookup is done for the
subsequent nodes in the source path.
4. Query Mechanism
When a source node, s , needs to reach a destination or
target resource, T , it first checks its neighborhood
routing table to see if T exists in its own neighborhood.
If T is not found in the neighborhood, s sends a
Destination Search Query ( DSQ ) to its contacts. The
DSQ contains the following information: (1) depth of
search ( D ), and (2) target resource ID ( T ). Upon
receiving a DSQ, each contact checks the value of D . If
6 D is equal to 1, the contact performs a lookup for T in its
own neighborhood. If T exists, then the path to T is
returned to s , and the query is considered successful.
Otherwise, if D>1 , the contact receiving the DSQ
decrements D by 1 and forwards the DSQ to each of its
contact, one at a time. In this way the DSQ travels
through multiple levels of contacts until D reduces to 1.
The source s , first sends a DSQ with D=1 to its contacts,
one at a time. So only the first level contacts are queried
with this DSQ . After querying all its contacts if the
source does not receive a path to the target within a
specified time, it creates a new DSQ with D=2 and sends
it again to its contacts, one at a time. Each contact
observes that D=2 and recognizes that this query is not
meant for itself. So it reduces the value of D in the DSQ
by 1 and forwards it to its contacts one at a time. These
contacts serve as second level contacts for the source. A
second level contact on receiving the DSQ observes that
D=1 and it does a lookup for the target T in its own
neighborhood and returns the path to T , if found. In this
way the value of D is used to query multiple levels of
contacts in a manner similar to the expanding ring
search. However, querying in CARD is much more
efficient than the expanding ring search as the queries
are not flooded at with different TTLs but are directed to
indiviual nodes (the contacts). Contacts leverage
knowledge of their neighborhood topology (gained
through the proactive scheme operating within the
neighborhood) to provide an efficient querying
mechanism.
IV. E VALUATION AND ANAYLSIS
In this section we present detailed simulation based
evaluation and analysis of our architecture. NS-2 [14]
along with our CARD extensions and other utilities were
used to generate various scenarios of ad hoc networks.
Mobility model for these simulations was random way-
point model. Our simulations so far did not consider
MAC-layer issues.
First we try to understand the effect of various
parameters such as the neighborhood radius ( R ), the
maximum contact distance ( r ), the number of contacts
( NoC ), the depth of search ( D ) and the network size ( N ) on ( A ) the reachability and ( B ) overhead. Reachability
here is defined as the percentage of nodes that are
reachable from a source node. For overhead we consider
the number of control messages. We consider overhead
due to contact selection and contact maintenance.
Having developed a deeper understanding of the various
parameter in our architecture, we then compare it with
other schemes such as flooding and bordercasting in
terms of querying overhead and query success rate.
Table 1 shows the scenarios used in our simulations.
These scenarios vary in number of nodes, network size,
and propagation range. The variation is considered to
capture the effect of these factors on CARD . As shown
in Fig. 3 and Fig. 4, the edge method outperforms the
probabilistic method. (We obtained similar results for
other scenarios.) Therefore, we present only the results
for the edge method.
TABLE 1 Description of various scenarios used for simulating CARD
No. Nodes Area Tx Range No. of Links Node Degree
Network
Diameter
Av. Hops
1 250 500*500 50 837 6.75 23 9.378 2 250 710*710 50 632 5.223 25 9.614 3 250 1000*1000 50 284 2.57 13 3.76 4 500 710*710 30 702 4.32 20 5.8744 5 500 710*710 50 1854 7.416 29 11.641 6 500 710*710 70 3564 14.184 17 7.06 7 1000 710*710 50 8019 16.038 24 8.75 8 1000 1000*1000 50 4062 8.156 37 14.33 A. Reachability Analysis
Reachability Analysis was conducted to understand
how contacts help in increasing the view of the network.
Here we present results for a topology of 500 nodes
spread over area of 710m by 710m. The details can be
seen from Table 1, scenario number 5. Similar results
were observed for other scenarios.
1. Varying Neighborhood Size (R)
Fig. 5 shows the effect of increasing the
neighborhood size ( R ) on reachability. As the value
of R is increased, the reachability distribution shifts
towards the right side i.e., more number of nodes
achieve higher percentage of reachability. This
increase in reachability with the increase in R is due
to increase in the number of nodes within the
neighborhood. However as the value 2R approaches
the maximum contact distance ( r ), the region
available for contact selection (between 2R and r ) reduces. This results in less number of contacts
being chosen. In the Fig 5, when R=7 , contacts can
only be selected between 2R=14 and r=16 hops
from the source. This small region for contact
selection significantly reduces the number of
contact and hence the reachability distribution shifts
back to the left side. At this point most of the
reachability is due to the neighborhood of the
source only.
7 0 50
100
150
200
250
300
350
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Reachability (%)
N = 500, Area = 710m* 710m, Propagation range = 50m, r = 16, NoC = 10, D = 1
Number of Nodes R = 1
R = 2
R = 3
R = 4
R = 5
R = 6
R = 7
Fig. 5 Effect of Neighborhood Radius ( R ) on Reachability
2. Varying Maximum Contact Distance (r)
Fig. 6 shows the effect of increasing maximum contact
distance on reachability. Since contacts are selected
between 2R and r hops from the source, higher values of
r provide a wider region for contact selection. The
mechanisms for contact selection described earlier
prevent selection of contacts that have overlapping
neighborhoods. This implies, as r increases a larger
number of contacts can be selected before their
neighborhoods start to overlap. Therefore the
reachability increases with increase in r . Larger values of
r also mean that the average contact path length would
increase (as more and more contacts are chosen at larger
distances from the source). However, once the
neighborhoods of the contact and the source are
completely non-overlapping, in a uniformly distributed
network a far away contact is as good as a near by
contact with respect to increase in reachability.
Therefore as can be seen from the figure, for r > (2R +8 ) there is not a very significant increase in reachability.
0 50
100
150
200
250
300
350
400
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Reachability (%)
N = 500, Area = 710m* 710m, Propagation range = 50m, R = 3, noc = 10, D = 1
Number of Nodes r=2R
r=2R+2
r=2R+4
r=2R+6
r=2R+8
r=2R+10
r=2R+12
Fig. 6 Effect of Maximum Contact Distance ( r ) on reachability
3. Varying Number Of Contacts ( NoC)
As described earlier, NoC specifies the value of the
maximum number of contacts to be selected for
each source node. The actual number of contacts
chosen is usually less than this value. This is due to
the fact that for a particular value of R and r , there
is only a limited region available for choosing
contacts. Once this region has been covered by
neighborhoods of the chosen contacts, choosing
more contacts in the same region is not possible as
their neighborhoods would overlap with the
neighborhoods of the already chosen contacts.
Therefore contact selection mechanism prevents
selection of more contacts. This characteristic can be
seen in the Fig. 7, in which the reachability initially
increases sharply as more and more contacts are chosen.
However, the increase in reachability saturates beyond
NoC=6 as the actual number of contacts chosen saturates
due to the effect of overlapping neighborhoods described
above.
4. Varying Depth Of Search (D)
D specifies the levels of contacts that are queried in a
breadth first manner. When D=1 , a source node looking
for a particular destination beyond its neighborhood,
queries its first level contacts only. When D=2 , if none
of the first level contacts contain the destination in its
neighborhood, second level contacts (contacts of the first
level contacts) are queried through the first level
contacts. As can be seen from the Fig 8, reachability
increases sharply as the depth of search D is increased.
Hence, depth of search helps in making a tree-like
structure of contacts, which is helpful in making CARD
scalable.
5. Varying Network Size
Fig. 9 shows a variation of reachability distribution for
three different network sizes, N . The area of the three
networks has been chosen so that the node density is
almost same across the three networks. Fig. 9 shows that
for any given network (specified by the values of N and
the area), the values of R and r can be configured to
provide a desirable reachability distribution in which
most of the nodes have a high value of the percentage
reachability.
B. Overhead Analysis
Overhead analysis is done in terms of number of control
messages required for contact selection and
maintenance. The total overhead considered includes:
8 1. Contact selection overhead: This is the amount of
CSQ traffic generated for selecting new contacts.
This includes overhead due to Backtracking as
described earlier.
2. Contact maintenance overhead: This is the traffic
generated by the contact path validation messages.
Local recovery , as described earlier, helps in
reducing this part of the total overhead.
Results are shown for scenario number 5 in Table 1.
Similar results were obtained for other scenarios.
0 50
100
150
200
250
300
350
400
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Reachability (%)
N = 500, Area = 710m* 710m, Propagation range = 50m, R = 3, r = 10, D = 1
Number of Nodes NoC = 0
NoC = 2
NoC = 4
NoC = 6
NoC = 8
NoC = 10
NoC = 12
Fig. 7 Effect of Number of Contacts ( NoC Reachability
0 20
40
60
80
100
120
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
Reachability (%) [N = 500, Area = 710m* 710m, Tx range = 50m, R = 3, noc = 10, r = 10]
Number of Nodes D = 1
D = 2
D = 3
Fig 8. Effect of Depth of Search ( D ) on Reachability
0 20
40
60
80
100
120
140
5 15 25 35 45 55 65 75 85 95
Reachability (%)
Number of Nodes N=250, Area=500m * 500m,
NoC = 10, R=3, r=14
N=500, Area=710m * 710m,
NoC = 12, R=5, r=17
N=1000, Area=1000m *
1000m, NoC = 15, R=6, r=24
Fig. 9 Reachability for different network sizes.
1. Varying Number Of Contacts ( NoC)
As shown in Fig. 10, as the number of contacts increases
the maintenance overhead increases sharply as more and
more nodes have to be validated.
2. Varying Maximum Contact Distance (r)
As r increases the number of contacts increases as
shown in Fig. 11, the increase in the number of contacts
is due to the availability of a wider area for choosing
contacts. Moreover, with higher values of r , contacts
may lie at greater distances from the source. That is, the
contact path length is expected to be higher for larger
values of r . This suggests that the maintenance overhead
should increase with increase in r . However as shown in
Fig. 11, the overhead actually decreases with increase in
r . Fig.12 explains this decrease in maintenance
overhead. Fig.12 shows that as the value of r increases
the backtracking overhead decreases significantly.
Recall that backtracking occurs when a node receiving a
CSQ cannot become a contact due to overlap with
already existing contacts. As r increases, the possibility
of this overlap decreases due to availability of a wider
area for contact selection. This decrease in back-tracking
overhead is significantly more than the increase in
overhead due to increased number of contacts and
contact path length. Therefore, the total maintenance
overhead decreases.
0 100
200
300
400
500
600
700
800
2 4 6 8 10
Time (sec) [N = 500, Area = 710m* 710m, Tx range = 50m, R = 3, r = 10, D = 1]
Overhead (Control Messages) per Node NoC = 3
NoC = 4
NoC = 5
NoC = 7
Fig. 10 Effect of Number of Contacts ( NoC) on Overhead
9 0 100
200
300
400
500
600
700
800
2 4 6 8 10
Time (sec) [N=500,Area=710mx710m,Tx range=50m,NoC=5,R=3,D=1]
Overhead (control messages) per node r = 8
r = 9
r = 10
r = 12
r = 15
Fig. 11 Effect of Maximum Contact Distance ( r ) on Total Overhead
3. Maintainance Overhead Over Time
Fig. 13 shows the maintenance overhead per node over a
20sec period. The maintenance overhead decreases
steadily with time. However, the number of contacts
increases slightly. This suggests that the source nodes
find more and more stable contacts. Stable contacts may
be defined as those nodes that have low velocity relative
to the source node. Therefore, a node moving in the
same direction as source node with similar velocity
could prove to be a stable contact. Hence, CARD leads to
source nodes finding more such nodes in the vicinity
1 . C. Trade-off Between Overhead and Reachability
As shown in earlier results, higher reachability maybe
obtained by increasing the number of contacts. However,
increase in the number of contacts also increases the
contact selection and maintenance overhead. Fig. 14
shows this trade-off between reachability and overhead.
The figure shows that there exists a desirable region in
which good reachability (greater than or equal to 50 %)
can be achieved at a reasonable amount of selection and
maintenance overhead.
D. Comparison with Other Approaches
Here, we present comparison of CARD with flooding
and bordercasting [8] . Bordercasting was implemented
with query detection ( QD1 and QD2 ) as described in
[8] . 1 This may also be due to random way-point (RWP) model. We plan to
investigate this problem further and expect that different mobility models may
have different effects on performance of CARD . 0 100
200
300
400
500
600
700
800
2 4 6 8 10
Time (sec)
N = 500, Area = 710m* 710m, Propagation range = 50m, Noc = 5, R = 3, D = 1
overhead (control messages) per node r=8
r=9
r=10
r=12
r=15
Fig. 12 Effect of maximum contact distance ( r ) on backtracking
overhead
0 100
200
300
400
500
600
700
2 4 6 8 10 12 14 16 18 20
Time [N = 250, Area = 710m* 710m, Tx range = 50m, Noc = 6, R = 4, r = 16, D = 1]
Total Contacts Selected
Maintainance Overhead per
Node
Fig. 13 Variation of overhead with time
0 0.2
0.4
0.6
0.8
1 0 2 4 6 8 10
Number of Contacts
Normalized Overhead and Reachability Reachability
Overhead
Fig. 14 Trade-off between reachability vs. contact selection and
maintenance overhead
Fig. 15 shows the average traffic generated per node for
querying 50 randomly selected destinations from 50
random sources. CARD generated far less querying
traffic than the others . Fig. 15 also shows contact
selection and maintenance overhead in CARD . As can be
seen from the figure, CARD is still more efficient than
the other two approaches. Whereas flooding and
bordercasting result in 100% success in queries, CARD
showed a 95% success rate with D=3. CARD’s success
rate can be increased by increasing D .
10 0 1000
2000
3000
4000
5000
6000
7000
8000
250 500 1000
Number of nodes
Querying Traffic Flooding
Bordercasting
CARD
CARD Overhead
Fig. 15 Comparison of CARD with flooding and bordercasting
V. C ONCLUSIONS
In this paper we presented a novel architecture for
resource discovery in large-scale ad hoc and sensor
networks. Salient features of our architecture include its
ability to operate without requiring any location
information or any complex coordination. In our
architecture, each node knows routes to nodes within its
neighborhood. Based on small world concepts, we have
introduced the notion of contacts to serve as short cuts
that extend the view of the network per node and
increase reachability beyond the neighborhood. Two
protocols for contact selection were introduced and
evaluated: (a) probabilistic method and (b) edge method.
We evaluated the protocols for reachability and contacts
selection and maintenance overhead under mobility
conditions. The latter approach was found to experience
less overhead during selection due to reduced
backtracking, and was thoroughly analyzed over the
various dimensions of the parameter space (including R,
r, D, NoC, and network size). We further compared our
approach to other resource discovery techniques ; flooding and bordercasting. The overall overhead
experienced by the contact-based architecture was found
to be significantly lower than the rest. These results
show a lot of promise for the contact-based approach and
we are encouraged to further investigate this direction.
We plan to further evaluate our protocols under various
scenarios of mobility patterns and resource distributions
in the network. We shall also pursue other heuristics for
contact selection mechanisms.
R EFERENCES
[1] C.E. Perkins and P. Bhagwat, "Highly Dynamic Destination-Sequenced
Distance-Vector Routing (DSDV) for Mobile Computers", Comp. Comm.
Rev., Oct. 1994, pp.234-244.
[2] Tsu-Wei Chen and Mario Gerla, "Global State Routing: A New Routing
Scheme for Ad-hoc Wireless Networks" Proc. IEEE ICC'98.
[3] S. Murthy and J.J. Garcia-Luna-Aceves, "An Efficient Routing Protocol
for Wireless Networks", ACM Mobile Networks and App. J., Special Issue on
Routing in Mobile Communication Networks, Oct. 1996, pp. 183-97.
[4] David B. Johnson, Davis A. Maltz, "The Dynamic Source Routing
Protocol for Mobile Ad Hoc Networks" October 1999 IETF Draft.
[5] Charles E. Perkins, Elizabeth M. Royer, Samir R. Das, "Ad Hoc On-
demand Distance Vector Routing", October 99 IETF Draft.
[6] C.-C. Chiang, "Routing in Clustered Multihop, Mobile Wireless
Networks with Fading Channel" Proc. IEEE SICON'97, Apr.1997.
[7] J. Li, J. Jannotti, D. Couto, D. Karger, R. Morris, " A Scalable Location
Service for Geographic Ad Hoc Routing (GLS/Grid)", ACM Mobicom 2000.
[8] M. Pearlman, Z. Haas, " Determining the optimal configuration for the
zone routing protocol ", IEEE JSAC, p. 1395-1414, 8, Aug 1999.
[9] Z. Haas, M. Pearlman, " The Zone Routing Protocol (ZRP) for Ad Hoc
Networks ", IETF Internet draft for the Manet group, June '99.
[10] D. Watts, S. Strogatz, "Collective dynamics of 'small-world'
networks" ,Nature, Vol. 393, June 4, 1998.
[11] D.J.Watts. In Small Worlds, The dynamics of networks between order
and randomness. Princeton University Press, 1999.
[12] A. Helmy, "Architectural Framework for Large-Scale Multicast in
Mobile Ad Hoc Networks", IEEE ICC ’02.
[13] A. Helmy, “ Small Large-Large Scale Wireless Networks: Mobility-
Assisted Resource Discovery”, USC-TR, July 2002.
[14] L. Breslau, D. Estrin, K. Fall, S. Floyd, J. Heidemann, A. Helmy, P.
Huang, S. McCanne, K. Varadhan, Y. Xu, H. Yu, "Advances in Network
Simulation", IEEE Computer, vol. 33, No. 5, p. 59-67, May 2000
[15] J. Liu, Q. Zhang, W. Zhu, J. Zhang, B. Li, “A Novel Framework for
QoS-Aware Resource Discovery in Mobile Ad Hoc Networks”, IEEE ICC
’02.

Linked assets

Computer Science Technical Report Archive

Conceptually similar

PDF

USC Computer Science Technical Reports, no. 790 (2003)

PDF

USC Computer Science Technical Reports, no. 770 (2002)

PDF

USC Computer Science Technical Reports, no. 780 (2002)

PDF

USC Computer Science Technical Reports, no. 781 (2002)

PDF

USC Computer Science Technical Reports, no. 778 (2002)

PDF

USC Computer Science Technical Reports, no. 749 (2001)

PDF

USC Computer Science Technical Reports, no. 674 (1998)

PDF

USC Computer Science Technical Reports, no. 803 (2003)

PDF

USC Computer Science Technical Reports, no. 752 (2002)

PDF

USC Computer Science Technical Reports, no. 755 (2002)

PDF

USC Computer Science Technical Reports, no. 788 (2003)

PDF

USC Computer Science Technical Reports, no. 837 (2004)

PDF

USC Computer Science Technical Reports, no. 716 (1999)

PDF

USC Computer Science Technical Reports, no. 804 (2003)

PDF

USC Computer Science Technical Reports, no. 673 (1998)

PDF

USC Computer Science Technical Reports, no. 811 (2003)

PDF

USC Computer Science Technical Reports, no. 657 (1997)

PDF

USC Computer Science Technical Reports, no. 814 (2004)

PDF

USC Computer Science Technical Reports, no. 753 (2002)

PDF

USC Computer Science Technical Reports, no. 734 (2000)

Description

Saurabh Garg, Priyatham Pamu, Nitin Nahata, Ahmed Helmy. "Contact-based architecture for resource discovery (CARD) in large scale MANets." Computer Science Technical Reports (Los Angeles, California, USA: University of Southern California. Department of Computer Science) no. 772 (2002).

Asset Metadata

Creator Garg, Saurabh (author), Helmy, Ahmed (author), Nahata, Nitin (author), Pamu, Priyatham (author)

Core Title USC Computer Science Technical Reports, no. 772 (2002)

Alternative Title Contact-based architecture for resource discovery (CARD) in large scale MANets (title)

Publisher Department of Computer Science,USC Viterbi School of Engineering, University of Southern California, 3650 McClintock Avenue, Los Angeles, California, 90089, USA (publisher)

Tag OAI-PMH Harvest

Format 10 pages (extent), technical reports (aat)

Language English

Unique identifier UC16269193

Identifier 02-772 Contact-Based Architecture for Resource Discovery (CARD) in Large Scale MANets (filename)

Legacy Identifier usc-cstr-02-772

Format 10 pages (extent),technical reports (aat)

Rights Department of Computer Science (University of Southern California) and the author(s).

Internet Media Type application/pdf

Source 20180426-rozan-cstechreports-shoaf (batch), Computer Science Technical Report Archive (collection), University of Southern California. Department of Computer Science. Technical Reports (series)

Access Conditions The author(s) retain rights to their work according to U.S. copyright law. Electronic access is being provided by the USC Libraries, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name USC Viterbi School of Engineering Department of Computer Science

Repository Location Department of Computer Science. USC Viterbi School of Engineering. Los Angeles\, CA\, 90089

Repository Email csdept@usc.edu

Inherited Values

Title Computer Science Technical Report Archive

Description Archive of computer science technical reports published by the USC Department of Computer Science from 1991 - 2017.

Coverage Temporal 1991/2017

Repository Email csdept@usc.edu

Repository Name USC Viterbi School of Engineering Department of Computer Science

Repository Location Department of Computer Science. USC Viterbi School of Engineering. Los Angeles\, CA\, 90089

Publisher Department of Computer Science,USC Viterbi School of Engineering, University of Southern California, 3650 McClintock Avenue, Los Angeles, California, 90089, USA (publisher)