Adaptive QoS Multicast Routing with Mobility Prediction in MANETs

International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC) Vol.1, No.3, September 2010
DOI : 10.5121/ijasuc.2010.1305 59
Adaptive QoS Multicast Routing with Mobility
Prediction in MANETs
G. Santhi1
and Dr.Alamelu Nachiappan2
1
Assistant Professor, Department of IT, Pondicherry Engineering College.
2
Associate Professor, Department of EEE, Pondicherry Engineering College.
shanthikarthikeyan@yahoo.co.in nalam63@yahoo.com
ABSTRACT
A Mobile Ad hoc NETwork (MANET) is a collection of wireless mobile hosts that form a temporary network
without a centralized administration or wired infrastructure. Due to the high mobility of nodes, the network
topology of MANETs changes very fast, making it more difficult to find the routes that message packets use.
Network control with Quality of Service (QoS) support is a key issue for multimedia applications in MANET.
Most of the real time applications have stringent requirements on bandwidth, delay, delay-jitter, packet loss
ratio, cost and other QoS metrics. This paper proposes a Multi-constrained QoS routing with mobility
prediction protocol. If the node has enough resources to transmit data packets, it uses the Global
Positioning System (GPS) to get the location information of the mobile nodes and selects the routing path
with the maximum Route Expiration Time (RET). A set of static and mobile agents are used to find the
multicast routes and transmit the packets. Extensive simulations have been conducted to evaluate the
performance of MC_MAODV using Network Simulator (NS-2). The simulation results show that the
proposed protocol achieves good performance in terms of improving packet delivery ratio and minimizing
end-to-end delay.
KEYWORDS: MANETs, Multi-constrained routing multicasting, clusters, end-to-end delay, packet loss
ratio.
1. INTRODUCTION
A MANET is a dynamically re-configurable wireless network with no fixed infrastructure. In such
a network, each mobile node operates not only as a host but also as a router and it can move in
any direction. Such networks have recently drawn significant research attention since they offer
unique benefits and versatility with respect to bandwidth spatial reuse, intrinsic fault tolerance and
low-cost rapid deployment. In order to facilitate communication within the network, a routing
protocol is used to discover routes between nodes [1]. The primary goal of such an ad hoc
network routing protocol is to provide an efficient route establishment between a pair of nodes so
that messages may be delivered in a timely manner. Route construction should be done with a
minimum of overhead and bandwidth consumption.
The increasing popularity of collaborative multimedia applications is enabling the support for
multicast communication. Multicasting plays an important role in ad hoc networks when
applications must send the same data to more than one destination. Multicasting facilitates
bandwidth saving, reduced delays and high scalability. Most of conventional multicast protocols
were designed for best-effort data traffic [15]. They construct multicast trees primarily based on
connectivity. Such trees may be unsatisfactory when Quality of Service is considered due to the
lack of resources. QoS is more difficult to guarantee in ad hoc networks than in other type of
networks, because the wireless bandwidth is shared among adjacent nodes and the network

60
topology changes as the nodes move. With the extensive applications of MANETs in many
domains, the appropriate QoS metrics should be used, such as bandwidth, delay, packet loss rate
and cost for multicast routing [16]. Mobility prediction has also been applied to determine the
duration of life time between two connected mobile nodes using GPS [11]. Using mobile
prediction mechanism the proposed scheme chooses the most stable links which satisfy the
multiple QoS constraints in MANETs .
This paper presents a QoS multicast routing protocol with mobility prediction to meet the
requirement of a single call under multiple QoS constraints. The rest of the paper is organized as
follows. Related work is presented in the next section. Section 3 explains the network model,
mobility prediction mechanism and agent model. In Section 4, the proposed QoS aware multicast
routing is explained in detail. Some simulation results are presented in section 5. The paper
concludes with section 6.
2. RELATED WORKS
Multicast routing protocols in ad hoc networks must deal with typical limitations of these
networks, which include high power consumption, low bandwidth, and high error rates. Most of
the conventional multicast protocols are designed for maximizing the throughput or for
minimizing the end-to-end delay. When QoS is considered some protocols may be unsatisfactory
or impractical due to the lack of the resource and the excessive computation overhead [1]. Some
algorithms [2] provide heuristic solutions to the NP-complete constrained Steiner tree problem,
which is to find the delay-constrained least-cost multicast trees. These algorithms however are not
practical in the Internet environment because they have excessive computation overhead, require
knowledge about the global network state, and do not handle dynamic group membership.
Jia’s distributed algorithm [3] does not compute any path or assume the unicast routing table can
provide it. However, this algorithm requires excessive message processing overhead. The
spanning join protocol [4] handles dynamic membership and does not require any global network
state. However, it has excessive communication and message processing overhead because it
relies on full flooding to find a feasible tree branch to connect a new member. QoS MIC, proposed
by Faloutsos et al. [5] alleviates but does not eliminate the flooding behavior. In addition, an extra
control element, Manager router, is introduced to handle the join requests of new members.
QMRPCAH proposed by Layuan et al.[6] provides multiple guarantees for satisfying multiple
constraints but it does not maintain any global network state. The Source Routing-based
Multicast Protocol (SRMP) is an on-demand multicast routing protocol proposed in [7]. Route
selection takes place through establishing a multicast mesh, started at the multicast receivers, for
each multicast session. In SRMP, two important issues in solving the routing problems are
addressed: a) path availability concept and higher battery life concept. The former allows the
protocol to distinguish between available and unavailable paths.
A new scalable QoS multicast routing protocol (SoMR) that has very small communication
overhead and requires no state outside the multicast tree is proposed in [8]. A QoS Multicast
Routing protocol (QMR) with a flexible hybrid scheme for QoS multicast routing is proposed in
[9]. QMR is a mesh-based protocol which is established on-demand to connect group members
and provides QoS paths for multicast groups. The QMR protocol integrates bandwidth
reservation function into a multicast routing protocol with the assumption that available

61
bandwidth is constant and equal to the raw channel bandwidth. Admission control mechanism is
used to prevent intermediate node from being overloaded and reject requests of new sources if
there is no available bandwidth. An agent based multicast routing scheme (ABMRS) in MANETs,
which uses a set of static and mobile agents for route discovery and maintenance is proposed in
[10]. But it doesn’t consider the multiple QoS constraints.
In MANETs, all the nodes are free to move around randomly and network topoly changes
dynamically. Due to these characteristic, the routing path is often invalidation. So, it is important
that a stable routing is to be found [11]. Thus, predicting the mobility of nodes is an effective and
feasible method. At present, the mobility prediction methods are path availability models [12],
prediction-based link availability estimation[13], link expiration time model [14] and prediction
strategy. Link expiration time model predicts link stability between two nodes according to the
node’s information such as location, velocity, and direction etc with the aid of Global Position
System (GPS).
It is observed from the literature that still multicast routing with QoS needs much more attention
so that the routing scheme must be robust, maximize packet delivery ratio and adapt dynamically
to changes un MANET topology and environment. This paper proposes a QoS aware multicast
routing scheme with mobility prediction that ensures QoS guarantees such as bandwidth
reservation, delay constraint, delay-jitter and minimum cost to multicast sessions
3. PROPOSED WORK
This section describes the functioning of the proposed multicast routing scheme. It includes the
following steps.
3.1. QoS multicast model
As far as routing is concerned, a network is usually represented as a weighted digraph G = (V, E),
where V denotes the set of nodes and E denotes the set of communication links connecting the
nodes. Let Φ denotes the set of multicast nodes, Φ V. |V| and |E| denote the number of nodes and
links in the MANET, respectively. Let s Є Φ be source node of a multicast tree, and M {Φ –
{s}} be a set of end nodes of the multicast tree. Let R be the positive weight and R+
be the
nonnegative weight.
• For any Link, e Є E, we can define some QoS. metrics: delay function delay (e): E →R,
packet loss ratio plr(e): E→R, bandwidth function bandwidth(e): E→R and delay jitter
function delay jitter(e): E→R.
• Similarly, for any node n Є V, one can also define some metrics: delay function
delay(n): V → R, packet-loss function packet-loss(n): V→ R+,
delay jitter function delay
jitter(n): V→ R+
We also use T (s, M) to denote a multicast tree in which the following relationships hold:
Delay (p(s,t)) = ∑ eЄP(s,t) delay(e) + ∑ nЄ P(s,t) delay(n)
Bandwidth (p(s,t)) = min{bandwidth(e), e Є P(s,t)}
Packet-loss (p(s,t)) = 1 – ΠnЄP(s,t) (1-packet-loss(n))
Delay-jitter (p(s,t) = ∑eЄP(s,t) delay – jitter(e) + ∑nЄP(s,t) delay – jitter(n)

62
Cost(T(s,M)) = ∑ eЄT(s,M) cost(e) + ∑ nЄ T(s,M) cost(n).
where p(s, t) denotes the path from source s to end node t of T(s,M).
Delay Constraint : delay(p(s,t))≤ Dt
Bandwidth Constraint : bandwidth(p(s, t))≥B
Packet loss Constraint : packet-loss(p(s,t)) ≤ L
Delay jitter Constraint : delay-jitter(p(s,t)) ≤ J
Meanwhile, the Cost(T(s,M)) should be minimum. Where, D is delay constraint, B is bandwidth
constraint , L is packet loss constraint and J is the delay Jitter constraint.
3.2. Mobile Prediction mechanism
In mobile ad hoc network, the reliability of a path depends on the stability or availability of each
link of this path because of the dynamic topology changes frequently. It supposes a free space
propagation model [9], where the received signal strength solely depends on its distance to the
transmitter. Therefore, using the motion parameters (such as speed, direction, and the
communication distance) of two neighbors, the duration of time can be determined in order to
estimate that two nodes remain connected or not. Suppose two nodes i and j are within the
transmission distance ra between them. Let (xi, yi) and (xj, yj) be the coordinate of mobile host i
and mobile host j. Also let (vi, θi) be the speed and the moving direction of node i, let (vj, θj) be
the speed and the moving direction of node j. The LET (Link Expiration Time) is predicted by [9]:
- (ab+cd) + (a2
+c2
) r2
– (ad – bc)2
LET(i,j) =
a2
+c2
------------------- (1)
Note that a = v1cosθ1 − v2cosθ 2 , b= x1 −x2 , c = v1sinθ1 − v2Sinθ2 , and d = y1−y2 . Note also
that the equation cannot be applied when v1 = v2 and θ 1= θ 2 , and when LET is ∞. In order to
get and utilize the information from GPS, the packets must include extra fields. When a source
node sends a request packet, the packet appends its location, direction, and speed. The next hop
neighbor of the source node receives the request packet to predict the duration of time between
itself and the source node. If node B is the next hop of the packet for node A, node A will insert its
location information in the packet so node B will be able to compute the duration of time between
node A and node B.
Assume that l is a routing path and (l1, l2, … lk) is the set of all the links along each hop l. The
route expiration time of RET(l) is the minimum of the LETs along path l and can be written as the
following equation:
RET(l) = min (LET(li)) --------------------- (2)
iє(1,2,...k)

63
3.3. Agent model
The proposed scheme employs a set of static and mobile agents. Agents use their own
knowledgebase to achieve the specified goals without disturbing the activities of the host. The
primary goal of an agent is to deliver information of one node to others in the network. Here, the
agents are used to find the multicast routes and to create the backbone for reliable multicasting. It
comprises a set of following agents and maintains a knowledgebase for inter agent
communication.
• Knowledge Base: The knowledgebase maintains a set of network state variables such as
status of the node (cluster head, child, others) available power, bandwidth, number of
movements made in recent interval, group id, neighbor ids and their status.
• Route Discovery Agent (RDA): It is a static agent that runs on nodes, creates agents and
knowledge base, controls and coordinates the activities of the multicast routing agency.
• Node Manager Agent (NMA): It maintains the multicast tree and provides group ID to all
the members of the packet-forwarding nodes in the multicast tree.
• Link Management Agent (LMA): When a node moves out of the communication range or
dies, LMA initiate route error (RERR) message to the upstream node and again initiates
the route request.
•
Figure 1. Routing agent model
To initiate multicasting, the mobile agent which resides in the source node provides group ID and
distributes multicast key to all the group members. The source node examines all the QoS metric
values of next intermediate node using the knowledgebase of agents. If it satisfies all constraints
data transmission takes place among all multicast members. When an intermediate node either
moves out of the range or fails, the static agent resides at the node that monitored such a situation
will find out the new alternate path with minimum cost between the nodes. The alternate path and
its connectivity to the network are broadcast, so that the new forwarding table is generated at
every node. The agents update their knowledgebase with more recent values.

64
3. QoS aware Multicast Routing Protocol with Mobility Prediction
(MPQMRP)
In this section we propose a QoS aware multicast routing protocol with mobility prediction. In
MPQMRP we create a multicast tree for transmitting data packets. The proposed protocol
includes the route discovery process and route maintenance process.
4.1. Route Discovery Process
In MPQRMP, the route discovery process is initiated whenever a source node needs to
communicate with another node for which it has no routing information in its table. Here the RDA
takes care of the routing process. It provides the one hop information to the source node. By using
this information, the source node initiates the route discovery by broadcasting a route request
(RREQ) packet to its neighbors. The route request packet includes (sequence_id, Xi,Yi, θi, Vi,
Source ID, Destination ID, D,J,B,C ) fields. When a node receives a new RREQ, it looks in its
route table for the destination. If it doesn’t match, then
it checks the available bandwidth between them. If the available bandwidth is above the
constrained bandwidth then it will utilize the location information to get the LET between the
nodes. Now, the intermediate node forwards the RREQ to its own neighbors after updating its
own information. In the
Figure 3. Route discovery process
above Fig. 3, node ‘s’ is the multicast source. D1 and D2 are the destinations. The network’s edge
are described by four tuples (B,D,J,C) where B is the Bandwidth constraint, D is the delay
constraint, J is the delay-jitter constraint and C is Cost. In this example, B, D, J, C constraints are
specified as 45, 15, 30 and 60 respectively. The source node ‘s’ broadcast the RREQ packet to all
of its one hop neighbors and it is received by n1, n5 and n6 respectively. All the receiving nodes
check the destination id , but it doesn’t match with them. In this example the bandwidth
constraint 45 is satisfied by the links between the nodes n5 n7 and n1 n2 are satisfied and
hence they forward the RREQ packets to its neighbors. At the same time the link n5 n6 does not
satisfy the bandwidth constraint and it discards the route request. This process is continued until
the RREQ packet reaches the destination. After receiving RREQ packet, each receiving node
calculates the LET between the nodes.
n1
s
n6
n5
n2
n3
n7
n8
d1
d2
50,5,4,10
45,2,7,12
50,2,1,4
502,5,1
2
45,2,6,12
50,4,4,10
(45,3,8,20
24,3,8,20
45,2,6,12
25,2,6,12
45,2,6,10

65
Figure 3. Route relay process
Each destination node sends it Route Reply (RREP) packet back to the source node. Source node
receives all the destination nodes reply packets and selects the path with highest RET. The
Network Management Agent (NMA) takes care of the multicast data transmission. In the above
example, from source ‘s’ to destination node ‘d1’ there are fours paths exist.
S n1 n2 n3 d1, S n1 n2 n8 d1, S n5 n7 n8 d1, and S n6 n7 n8 d1.
Among these four paths S n1 n2 n8 d1, S n6 n7 n8 d1 do not satisfy the delay
constraints. The paths S n2 n3 d1 and S n5 n7 n8 d1 satisfy delay, delay-jitter, and
bandwidth constraints.
Route path R(S,n1,n2,n3,d1) R(S,n5,n7,n8,d1)
RET 8 6
Table 1. Paths and their RET values
The RET in these paths are calculated and the route with maximum RET and least cost is selected
for data transmission. The RET of the above example is depicted in Table 1.
Figure 4. Reliable multicast tree
In the above figure 5, the route R(S,n1,n2,n3,d1) which has the maximum RET and selected for
data transmission
45,2,7,12
n1
s
n6
n5
n2
n3
n7
n8
d1
d2
50,5,4,1050,2,1,14
502,5,12
45,2,6,12
50,4,4,10
(45,3,8,20)
45,3,8,20)
45,2,6,12
50,2,6,12
45,2,6,10
45,2,7,12
45,2,6,12
n1
s
n6
n5
n2
n3
n7
n8
d1
d2
50,5,4,1050,2,1,4
502,5,1
2
45,2,6,12
50,4,4,10
(45,3,8,20
24,3,8,20
25,2,6,12
45,2,6,10

66
4.2. MPQMRP routing Algorithm
The routing algorithm used in MPQMRP protocol is as follows.
Input: a network G = (V,E) and a QoS reuest R = (Source ID, Destiation IDs, B,J,D,C)
Output : a routing path P that satisfies the QoS request R.
For each destination Do
Step 1. Source node broadcasts the RREQ packet to all of its one-hop neighbors.
Step 2: If the packet is new, and node is destination and the link of the node pair can
satisfy the QoS requirements such as Bij ≥ B, Dij ≤ D and Jij ≤ J then directly send the
message.
Step 3: Else, if the packet is new and the node is not destination and Bij ≥ B then
forward the packet to all one-hop neighbors. Calculate LET between them using
the equation (1).
Step 4: Repeat step 3 until the RREQ packet reaches the destination.
Step 5: Find the collection of routing path from the source to destination.
Let them be R1, R2, …Rn.
Step 6: For each routing path Ri Do
Calculate ∑ Dij , ∑Jij and ∑ Cij
{if ∑ Dij ≤ D and ∑Jij ≤ J then calculate RET using formula ( )
Else
Delete the routing path from the collection.
Step 7: Select the routing path with maximum RET. If more than path exists with equal
RET select the path with minimum cost ∑ Cij.
Step 8: End routing.
4.3. Route Maintenance Process
Due to the dynamic nature of the network topology and restricted resources, the established route
often gets invalid. When the link is disconnected, the LMA informs it to the upstream node. This
node removes the corresponding entry from its multicast routing table and sends route reconstruct
packet to the source. The source starts to discovery the route again. Similarly, when a node wants
to join a multicast tree, it broadcast a join request across the networks. Only a node that is a
member of a multicast tree may respond. It creates a reverse route entry to the new node and
broadcasts the join request packet to its neighbors. Each member node of the multicast tree
receives join request and make its entry in the routing table and send back the join reply packet to
the new node.

67
5. Simulation
The proposed scheme has been simulated in various network scenarios using NS-2 simulator. A
discrete event simulation is done to test operation effectiveness of the scheme. In this section we
describe the simulation model and the simulation procedure.
5.1. Simulation model
A mobile ad hoc network consisting of ‘n’ nodes is generated by using a random placement of the
nodes and allowed for the free movement within the area of ‘l x b’m2
. Each node starts from a
random location and moves in all directions. A maximum number of movements allowed per node
every period ‘per’ is ‘move_max’. The communication range for each node is selected as ‘C_ran’.
All nodes are considered to be non-malicious and are included in the clustering scheme. All nodes
must support an agent server, interpreter and transport mechanism. Every node has enough
memory to support the agent’s knowledge database. Every mobile agent is only allowed three
hops from the parent node to avoid network congestion.
Table 2. Simulation parameters
Parameters Value
MAC Layer IEEE 802.11
Simulation area(m) 1000*1000
Simulation Time 60 secs
Number of nodes 25
Node mobility speed 0-60m/s
Mobility pattern Random way point
Traffic flow CBR
Packet size 512 bytes
Transmission range 250m
5.2. Simulation Procedure
To illustrate some results of the simulation, we have taken n = 50, l = 1000 m and b = 1000 m, per
= 100 s, move_max = 2, C_ran = 225 m. D=15, J=30, B=45, C=60 are given as user input for
various scenarios. The proposed routing scheme is evaluated in terms of packet delivery ratio,
end-to-end delay and control overhead.
• Packet delivery ratio: The ratio of the average number of data packets received by the
destination node to the number of data packets transmitted by the multicast source.
• End-to-end delay: The time when a data packet is sent by the source to the time the data
packet is received at the destination node.
• Control overhead: The total number of control received by the destination node.

68
5.3. Analysis of results
5.3.1. Packet Loss ratio: Figure 5. shows the performance of the average packet loss ratio under
various mobility speeds. The packet loss ratio is increased with increasing mobility due to more
link breaks.
Figure 5. Packet loss ratio vs. mobility speed
When the mobility is low, the multicast tree structure was mostly static and therefore the
packet loss ratio is low. In MPQMRP, the required resources are reserved in advance and the
most stable path is also identified with maximum RET. Hence the packet loss ratio of
MPQMRP is higher than that of MAODV. Packet loss ratio of the proposed system is less
than MAODV for all group size values. This is because of finding more reliable path and
managing node breakage thereby avoiding the recomputation of route. As the group size
increases packet loss ratio also increases since the entire nodes share the common wireless
medium for transmission of packets.
5.3.2. End-to-End delay: Figure 6.a. shows the performance of the end-to-end delay under
various mobility speeds. As the mobility speed increases average end to end delay also
increases.
Figure 6 a) End-to-End delay Vs group size Figure 6 b) End-to-End delay Vs mobility speed

69
Packet delivery latency is significantly less than that for MPQMRP compared to MAODV
(Figure 6.b) even at higher node mobility for constant group size. This is due to the fact,
forwarding nodes chosen are reliable nodes and are less prone to scarcity of resource. Also once
the transmission path breaks, the intermediate node chooses another backup path immediately for
transmitting the data packets.
5.3.3. Control overhead: Figure 7. shows the performance of the control overhead under
various mobility speeds. As was expected the control overhead increased as the number of nodes
increases. The reason is that the inclusion of extra fields in RREQ packer header. Also, the
number of route broadcasts increased when the group size increases.
Figure 7. Control overhead vs. Group size
6. Conclusions
This paper proposes a QoS aware multicast routing algorithm with mobility prediction based on
mobile agents. It effectively routes data packets to group members even in case of high mobility
and frequent link failures. A set of static and mobile agents are used to carry out route discovery
and route maintenance process. This routing path satisfies multiple QoS constraints according to
the QoS demand. It has higher packet delivery ratio as compared to MAODV and reduces end-
to-end delay. This work can further be extended to include the mobile prediction scheme for
multiple multicast trees.
7. References
[1] Luo Junhai and Ye Danxia et al., “A survey of multicast routing protocols for Mobile ad-hoc networks”
IEEE Communication surveys & Tutorials, vol.11, no.1, pp.78-90, First Quarter 2009.
[2] Luo Junhai, Xue Liu, Ye Danxia, Research on multicast routing protocols for mobile ad-hoc networks.
Computer Networks 52(5): 988-997, 2008.
[3] X.Jia A distributed algorithm of delay-bounded multicast routing protocol for multimedia applications
in wide area networks, IEEE/ACM Transactions on Networking 6(6), 1998, 828-837.
[4] L.Layuan et al., “A QoS guaranteed multicast routing protocol”, Computer communication, vol.27/1,
2004, 56-69.
[5] M. Faloutsos et al., QoSMIC: Quality of service sensitive multicast internet protocol. SIGCOMM’ in:
Proceedings of 98, September 1998, 56-63.
[6] Li Layuan et al., “A QoS multicast routing protocol for clustering mobilie ad hoc network” Computer
Communication 30(2007), 1641-1654.

70
[7] M. Shree, J.J. Garcia-Luna-Aceves, An efficient routing protocol for wireless networks, ACM Mobile
Networks and Applilcation Journal 1 (1996) 183-197.
[8] J. H. Cui and L.Lao et al. AQoSM: Scalable QoS multicast provisioning in Diff-Serv networks,
Computer Networks 50 (2005) 80-105.
[9] Huayi Wu and Xiaohua Jia QoS multicast routing by using multiple paths/tress in wireless ad hoc
networks, Ad Hoc Networks 5 (2007), 600-612.
[10] S. S. Manvi, M. S. Kakkasageri, Multicast routing in mobile ad hoc networks by using a multiagent
system. Inf. Sci. 178(6): 1611-1628, 2008.
[11] Xiaoyan ZHU and Jin LIAN A QoS multicast routing protocol with mobile prediction based on
MAODV in MANETs, International Conference on Computer Science and Software Engineering, 355-357,
2008.
[12] McDonald et al., “Path availability model for wireless ad hoc networks” Proceeding of IEEE wireless
communications, 1999, (WCNC ’99).
[13] Jiang Shengming et al., “A prediction based link availability estimation for mobile ad-hoc networks”
IEEE Infocom, 2001.
[14] Su W.Lee Sung-Ju et al., “ Mobility predictin and routing in ad-hoc wireless networks”» International
Journal of Network Management, 2001.
[15] E.M. Royer and C.E.Perkins, “Multicast operation of the ad-hoc on-demand distance vector routing
protocol”, in Proc. ACM MOBICOM, Aug. 1999, pp. 207-218.
[16] M. Saghir. T.C Wan, R. Budiarto, A new cross-layer framework for QoS multicast applications in
mobile ad hoc networks, International Journal of Computer Science and Network Security 6 (2006) 142-
151.

Adaptive QoS Multicast Routing with Mobility Prediction in MANETs

Recommended

More Related Content

What's hot (20)

Similar to Adaptive QoS Multicast Routing with Mobility Prediction in MANETs (20)

Recently uploaded (20)

Adaptive QoS Multicast Routing with Mobility Prediction in MANETs