D-ETBMRPFS: Development of an Efficient Traffic Balanced Multicast Routing Protocol for Wireless Mobile Ad-hoc Network Using Fibonacci Sequence Approach

Mobile ad hoc network is a decentralized wireless networks being developed using a different kinds of digital devices without any fixed infrastructure. This is because resource conservation is a very important task in wireless networks. Therefore multicast Routing is an effective technique for maintaining the same state. However, multicasting in mobile ad-hoc networks is not an easy task due to the dynamic nature of network topologies, large fluctuations in signal strength, lack of battery power, and network traffic. Traffic was considered a network issue in this document. To solve the aforementioned problem, an efficient multicast routing protocol for traffic balancing using Fibonacci sequence is proposed. This mechanism allows the sender to send data along a specific routing path using Fibonacci sequence numbers to improve network performance.


Introduction
MANETs is a decentralized wireless networks.Mobile devices can move in any direction within this network.Wireless channels are used to establish communication between mobile devices.The special goal of such a network is to manage the data on each node that may used to transfer data from one node to other at once [1].Ad-hoc networks (MANET) [2][3][4][5] is best suited for various application where fast deployment and dynamic reconfiguration is required.This includes military bleed edges, emergency search and rescue operations, military communications, and automated battlefields [4,6].
The random mobility of the device may hamper the network topology that leads to loss the communication among the devices.To send the data packet to the target node, it is being redirected to another path that may leads to congestion issue.Therefore, we need a much dynamic and robust routing algorithm [1,7].The role of multicast routing is very important in self-organizing wireless networks that can transmit the data from a source to a set of receivers at the same time.The main difficulty to implement the multicast routing protocols for wireless mobile networks is to control the mobility speed of devices in a network.Hence, in order to implement this, it includes many limitation as "dynamic topology, excessive node mobility, limited network assets like bandwidth, energy and low channel bandwidth" [8].A lot of research has been done in recent years for MANET's [2] and different kinds of multicast routing protocols have been designed for MANET [9,10].
When wireless networks are heavily stacked, leads to congestion problems result is loss of data packet.Hence to solve the congestion issue in multicasting in a MANET, multipath routing path comes in to the picture.However, a set of rules have encountered to discover the entire possible multicast routing path in order to transmit the data optimally, reduces the network traffic.

Constraints of a Wireless Mobile Ad-hoc Networks
When we are going to design a routing protocols for a wireless mobile Ad-hoc network, first we have to need explore the various physical constraints [11].
• Every device in the network have limited processing capacity.
• Normally, memory capacity of a mobile devices are bounded.
• Each devices in mobile ad-hoc network have limited battery power.
• currently available bandwidth is shared by various mobile devices.
• Each mobile devices have limited transmission range since they depends upon the radio waves.• The network topology changes very frequently due to device mobility.

Motivation
Referring to the traffic issue, we have motivated to reduce the traffic with proposal of load balancing multipath multicast routing protocols based on Fibonacci sequence approach.The proposed method consider multi-routing paths and distribute the transmitted information based on "Fibonacci sequence approach" over various multicast routing paths that leads to minimize the traffic and renovate the network performance.
Mathematically, the Fibonacci series is described as the arrangement of numbers that begins with 0 and 1, and every quantity is the sum of the preceding numbers, as shown in equation 1.

Organization of the Paper
The rest of part of the paper is structured as: Sect.2, shows the brief explanation of multicast routing protocols.Section 3, describes the mobility model for wireless ad-hoc mobile networks.Section 4, includes the details of the proposed protocol (D-ETBMRPFS). (1) Section 5, compares the overall evaluation, and experimental results with other multicast routing protocols such as "Ad-hoc On-demand Distance Vector (AODV)" and "Zone assisted mobility aware multipath routing (ZM2R) ".The abbreviations that can be used are indicated by the notation Table 1.

Taxonomy for Multicast Routing Protocols
This section gives a brief discussion about multicast routing for mobile ad-hoc networks (MANETs).

Overview of Multicast Routing Protocols
Numerous multicast routing protocols with specific characteristics have been developed for MANET.Basically, the multicast routing mechanism have been divided into several classes based on their topology and topology services.Here, topology based classification includes "Mesh-Based, Tree-Based and, Zone Based" routing protocols.However, some of multicast routing is also categorized based on their services that mainly focus on "Reliability, Bandwidth, Latency, and Bandwidth-Delay product."The best-known classification of multi-cast protocols is discussed in the literature [12].
The mesh-based multicast routing" protocol, is defined as where multi-routes are exist between a pair of hubs.This protocols can be applied for highly dynamic network.The data packets are scattered throughout a mesh structure that acts as a set of interconnected hubs.Compared to tree-based protocols, the packet transmission speed of mesh-based multicast routing protocols is usually very good.
In case of tree-based routing protocols there is single route exist between a pair of hubs, the performance of these protocols are better than mesh-based routing protocol in terms of data packet transmissions because of having the number of connection among the nodes.The forwarding efficiency and robustness of such routing protocols are very high because of the same.Further, the tree-based approach have been categorized into two subcategories: (i) source-rooted tree and (ii) shared-rooted tree approach.In source-rooted methodologies each source creates its single tree.Here, we have incorporates number of protocols that fall into the this category say: "Bandwidth-Efficient Multicast Routing Protocol (BEMRP)" [13], "Multicast Routing Protocol Based on Zone Routing (MZRP)" [14], "A location prediction based routing protocol (LPBR)" [15], and "Design of Load Balanced Multicast Routing Protocol for Wireless Mobile Ad-hoc Network (DLBMRP)" [16]".In shared-rooted approach all source share only a single tree that is governed by one or more special hubs.Many routing protocols follow this approach which includes "Multicast Ad-hoc On-Demand Distance Vector Routing Protocol (MAODV)" [17], "Entropy-based MAODV (EMAODV)" [18], "Ad-hoc Multicast Routing Protocol utilizing Increasing ID Numbers (AMRIS)" [10].
In addition, multicast routing has also been classified based on the nature of routing information flow like: "Active or global, and Passive or demand-driven"."Table-Driven" routing protocols, also known as proactive routing protocols that have a routing table on each node, "AMRoute (Adhoc Multicast Routing Protocol)" [7]. Figure 1 describes a complete scientific classification of the multicast routing protocols.In the active routing protocol, the nodes in a ad hoc mobile network continuously evaluate the routes to all accessible nodes and try to keep the routing information consistent and up-to-date.In a reactive protocol, as long as the source needs to be routed through the route discovery process, the route is determined.Typical reactive routing protocols are "On Demand Multicast Routing Protocol (ODMRP )" [9] and "On Demand Adhoc Multicast Distance Vector (MAODV)" [17]".For a complete review, readers refer to [19,20].
Anika Mansura [21] tries to enhance the network life by proposing the "Energy balanced and nodes aware (EBNA) routing protocol" for a wireless network claiming that most of the techniques do not consider the remaining energy and harvested energy, nor investigated the weightages of energies in order to select the cluster head (CH).It takes into account both residual and harvested energy and number of active nodes when selecting CH to improve throughput.
Maryam Ataei Kachooei [22] has been developed a routing protocol say "OLSR" based on Geocast for Vehicular Ad Hoc Networks in order to send the data from a source to a group of nodes located in a specific region.Author claiming that the traditional routing protocols are based on fooding mechanism that leads to high overheads when unicast routing been taken into account for data transmission.Therefore, to overcome these problems, recent georouting protocols exploit on-demand unicast routing methods, such as "Ad-hoc On-demand Distance Vector (AODV)" for routing the data to receiver and then transmit them to that region.
Michael Lescisin [23] has proposed a new quality of experience based routing protocol for the delivery of video data in a special peer-to-peer (P2P) network mode.P2P video data delivery is built as a resource allocation problem with the aim of maximizing the quality of Fig. 1 Categorization of ad-hoc multicast routing protocols the end-user experience.Resource allocation problem is a form of linear optimization problem where we intend to use bandwidth and energy at each mobile node.In addition, a scalable "video encoding" technique was used, which maximizes the quality of the experience by delivering video in layers across multiple sources using multiple paths to minimize power consumption.and download bandwidth at each node.
Kim et al. [24] has been proposed a protocol for mobile ad-hoc network referring the congestion as a major issue that affect over the network performance.Various existing routing protocols give the solutions for load balancing.A novel protocol has been proposed for"congestion adaptive routing along with load balancing, that is, load balanced congestion adaptive routing (LBCAR)".This method look over the nodes which are less busy and transfer the traffic load from congested nodes.

Mobility Model
This section describe mobility model that show the co-ordinate position, velocity and acceleration modifications of devices against time.A "random way point" mobility model [2] becomes a benchmark due to its simplicity as well as extensive availability of the same.
Let underlying the co-ordinate position of a device m and n are (p 1 , q 1 ) and (p 2 , q 2 ) sepa- rately at a time.The device m and n moves with variable speeds in a specific direction making an angle and with respect to positive x-axis.The Device m moves l 1 and n moves a l 2 distance in length T = 0 to t.At time t, m and n move to new positions (p 6 , q 6 ) and (p 3 , q 3 ) individually as demonstrated in Fig. 2.
The Euclidean distance ( l (mn,0) ) between the node m(p 1 , q 1 ) and n(p 2 , q 2 ) at time T = 0 given by Consider the node m and n moves with the variable speed v 1 and v 2 making an angle and with x-aixs, the distance l 1 and l 2 navigated by the hub in a specific time t is given by ( 2) Fig. 2 Co-ordinate position of the node before and after mobility At a time T = t , when m(p 1 , q 1 ) move at distance l 1 getting angle with x-axis then the new co-ordinate of the node will be y(p 6 , q 6 ) as demonstrated in Fig. 3 the estimation of p 6 and q 6 is given regarding speed and time t is given as Likewise, when n(p 2 , q 2 ) move at distance l 2 getting angle with x-axis the new position of the hub will be o(p 3 , q 3 ) the co-ordinate estimation of p 3 and q 3 is given with respect to speed and time t is shown as: When m and n occupy new position as y(p 6 , q 6 ) and o(p 3 , q 3 ) , separately.The new devia- tion between y(p 6 , q 6 ) and r(p 3 , q 3 ) at a time t is shown as:

Proposed Work
This section describes the detail of the proposed protocol "Development of an Efficient Traffic Balanced Multicast Routing Protocol using Fibonacci Sequence (D-ETBMRPFS)".

Routing Mechanism
The proposed mechanism includes two phases i.e. route discovery phase and route reply phase.Each node in a mobile ad-hoc network consist of its won routing table (RT) having (3) numerous relevant information in its header about their adjacent nodes such as S_add , MD_add , U id and NH_add etc.When the source node S want to transmits the data packets to a group of nodes say P and Q, first, it initiates the route discovery phase.It broadcast the "Route Request Packet (RREQ)" across the network to gain the better routing path.Let the nodes A, B, C, and D receive the "RREQ" packets.After receiving the "RREQ" packet by the node A, B, C, and D they compare their address with a multicast address which is carried out in the "RREQ" packet, if the address match with the multicast address stops broadcasting otherwise continue the same until getting the group of receivers nodes.When the receivers the "RREQ" packet, they create the route reply packet and communicate to the sender following the respective intermediate nodes involved in route discovering.
Subsequently, when the reply packets received by the source S, all the routing paths are arrange in increasing order of intermediate hops and assigned a weight using Fibonacci Series.The data packets are distributed over different routing paths based on equation 1.

Basic Idea for Traffic Balancing
Let us consider the source node have n data packets is to be transmitted to a group of nodes and there is P number of routing paths are exist.In our proposed work it has been supposed that if the network has P number of routes then all the routers are put in the increasing order of intermediate hops.The available data have been distributed over different routing path based on "Fibonacci Number".The proposed idea allocates Fib(P), Fib(P − 1), Fib(P − 2)......Fib(2)andFib(1) number of data packets over a specific routing path, where P is number of routing paths.We calculate the value of F(P) using equation 1, gives number of data packets which are transmitted along a specific routing path having minimum number of intermediate hop.The next Fib(P-1) number of data packets will be allocated over another routing path having just immediate higher number of intermediate hope, and so on, finally the routing path having maximum number of intermediate nodes has been allocated Fib (1) data packet.After covering all routing path the proposed protocol repeat the same mechanism.The proposed idea has been explained using following algorithm.

Algorithm 1 Data Packet Distribution
1: Begin 2: The source node S have n data packet to send over a P routing path.3: Find the number of intermediate hopes in the routing path using route discovery mechanism.4: Available routing paths have been arranged in increasing order based on their intermediate hope.5: Calculate the value of Fib(P) using equation (1), gives the number of data packets being transmitted over a routing paths having less number of intermediate hopes.6: Again calculate the value of Fib(p-1), gives the number of data packets that being transmitted over next routing path having just immediate more number of intermediate hope.7: Finally, find the value of Fib(1) that gives the number of data packets being transmitted over routing path having maximum number of intermediate hops.8: If all the available routing paths are consumed and some of the data are not delivered , go to step 4 and repeat all the steps until rest of the data packets are being transmitted 9: End For a better understanding, we have taken a network as shown in Fig. 4. When the source node S wants to transmit the data packet to the receiver nodes {P, Q} .There are four possible routes are available from the source node S to the destination node {P, Q} as {S, A, E, I, L, P, Q} , {S, B, F, P, Q} , {S, C, G, J, P, Q} and {S, D, H, K, P, Q} .According to the principal of proposed work, all the routes are put in increasing order of the intermediate hops.
Let the source node S have 9 data packets to transmit from the source node to receivers nodes.The first three packets are transmitted through the routing path SBFPQ because the total routing path is four, and the value of Fib(4) is three (3), the next two packets have been transmitted through SCGJPQ as the value of Fib(3) is two, and the next one data packet has been sent over SDHKPQ and SAEILPQ respectively because the value of Fib(2) and Fib (1) is one.Finally, the rest of the data packet i.e. (two packets) has transmitted over SBFPQ again.In this way, we can reduce network congestion using a proposed method that leads to higher network performance.

Performance Assessment
This section shows the simulation of the proposed idea and performance analysis has been shown around various network parameters.

Experimental Setting
The proposed protocol has been tested over well known simulation (NS-3).Simulation is done nearly 800 m × 800 m square meter having 15 to 300 movable nodes.In our experi- ments, "random waypoint" and "free-space propagation" models were considered.When implementing the proposed idea, we took into account many parameters as shown in Table 2.The overall network outcomes of the proposed system is QoS using several available multicast routing protocols including "AODV" and "Z2MR" in terms of "PDR", "E2E" delay and 'routing overhead'.

Packet Delivery Ratio (PDR):
It is a kinds of network metric used to measure the network performance and defined as: "The ratio between the number of data packets Fig. 4 Route discovery process successfully delivered to the destination node and the number of packets transmitted by the source node." In equation (10), P Received represents the total number of packets received by the receiver node, P generated is the total number of packets generated by the source nodes and n represents the number of nodes.

Packet Delivery Delay (PDD):
It is also a network metric, used to measure the network performance is defined as: "The average sum of the difference delay of each data packet received by the destination node and the time a data packet is sent by source nodes.
In equation (11) T Received represents the time when data packet received by the receiver nodes, T Transmission is the time when data packets generated by the source node.

Simulation Results and Analysis
The simulation shows the impact of transmission rate and mobility speed on the "PDR", "E2E" delay, and "Routing Overhead" of "AODV", "Z2MR" and the proposed protocol "D-ETBMRPFS".The Fig. 5, shows the "(PDR)" for AODV, ZM2R and the proposed protocol �� D − ETBMRPFS �� with respect to mobility speed.As the speed of the nodes increases the "Packet Delivery Ratio" gets decreases because there is more chances to lose the data packets due the to high speed of the node.The performance of Z2MR is better than AODV because Z2MR supports a multicasting routing approach, and less number of retransmissions take place.The proposed protocol " D − ETBMRPFS "further renovate the perfor- mance of Z2MR with minimizing the traffic load using "Fibonacci Sequence" approach.The proposed method reduces the traffic load by distributing the data over different routing path based on fibonacci number that leads to better "Packet Delivery Ratio (PDR)".Figure 6 shows the "E2E" delay of AODV, Z2MR and the proposed protocol D − ETBMRPFS with respect to mobility speed.As the speed of the nodes increases the "Packet Delivery Delay" also increases because there are more chances to disconnect the connection due to high speed of the node.The performance of Z2MR is better than AODV because Z2MR make the routing decisions by selecting the optimal nodes with less energy cost; hence the "Packet Delivery Delay" of ZM2R is less in compare to AODV.The proposed protocol enhances the performance of ZM2R by distributing the data over different routing path based on "Fibonacci Sequence" approach that leads to minimize the "E2E" delay.Figure 8 shows the "E2E" delay of AODV, Z2MR and the proposed protocol D − ETBMRPFS with respect to transmission rate.As the transmission rate of the data packet is growing the "End-to-End" delay is also growing.This is only because of high speed of the data transmission rate, leads to disconnect the communication gap among the nodes in the network.The performance of Z2MR in terms of E2E delay is less than AODV because Z2MR minimize the number of route reconstructions and packet retransmissions.Hence, the packet delivery ratio (PDR) in Z2MR increases as compared to AODV.The proposed protocol D − ETBMRPFS enhances the performance of Z2MR in terms of PDR by distributing the data over different routing path based on fibonacci number that improve the network performance.
Figure 9 explore the routing overhead of AODV, ZM2R, and D-ETBMRPFS with respect to transmission rate.As the transmission rate increases the routing overhead is also increases due to dynamic nature of the network.The performance of the ZM2R in terms of routing overhead is less than AODV because ZM2R routing protocol uses multipath approach in order to transmit the data packet, hence the routing overhead of ZM2R is less in compare to AODV.Further, the proposed protocol D-ETBMRPFS improve the performance of ZM2R due to consideration of multi-path as well as Fibonacci Sequence approach by distributing the data packets over different routing path that minimize the overhead.Due to the consideration of this technique, D-ETBMRPFS finds a better path for data transmission, leads to less overhead.Fig 10. explore the routing overhead of AODV, ZM2R, and D-ETBMRPFS with respect to speed.As the speed increases the routing overhead is also increases.Due to the high speed of the devices the communication links break down among the nodes the result is to determine the better path once again for retransmission of data that leads to routing overhead.

Conclusions
Various multicast routing protocols have been proposed for data transmission to a specific group of nodes.This paper proposes a "Efficient Traffic Balanced Multicast Routing Protocol using Fibonacci Sequence (D-ETBMRPFS)" in order to send the packet to a group of nodes in contrast to AODV and Z2MR method.Here, the data-gram has been transmitted over a specific routing path based on "Fibonacci Sequence" approach in order to minimize the network traffic.Our proposed protocol reduces the traffic overhead and perform better when the availability of the multicast route is high.The proposed protocol has been simulated using well-known simulator NS-3 tool.According to the simulation results, the "D-ETBMRPFS " outperforms in terms of the "PDR", "E2E" delay, and routing overhead than those of the AODV and Z2MR.Also, in order to enhance the reliability of paths to a Fig. 9 Route overhead against packets delivery ratio group of nodes, we will try to find the impact of potential of the nodes involves in creating the routing path as a convenient future work that can be done.

Fig. 3
Fig. 3 New position of the nodes after mobility

Figure 7
Figure7shows the "(PDR)" of AODV, Z2MR and the proposed protocol D − ETBMRPFS with respect to transmission rate.As the transmission rate of the data is increasing the "Packet Delivery Ratio" is decreasing because of communication gap among the nodes.The performance of Z2MR in terms of PDR is better than AODV because Z2MR makes routing decisions by determining the nodes having less energy cost.Further, the proposed protocol D − ETBMRPFS enhances the performance of Z2MR in terms of PDR by distributing the data over different routing path based on fibonacci sequence approach.Figure8shows the "E2E" delay of AODV, Z2MR and the proposed protocol D − ETBMRPFS with respect to transmission rate.As the transmission rate of the data packet is growing the "End-to-End" delay is also growing.This is only because of high speed of the data transmission rate, leads to disconnect the communication gap among the nodes in the network.The performance of Z2MR in terms of E2E delay is less than AODV because Z2MR minimize the number of route reconstructions and packet retransmissions.Hence, the packet delivery ratio (PDR) in Z2MR increases as compared to AODV.The

Fig. 7
Fig. 7 Packets delivery ratio against transmission rate

Table 2
Parameters used in the