The mobile devices have path to the internet technology that reduces an outlay with consenting us to utilize Mobile Ad Hoc Wireless Network. Mobile devices are free to move anywhere and they detect the presence of other nodes in the network and they communicate with each other in the absence of a central administration. The field of MANET has experienced an exceptional growth in many applications that embraces military troop communication, disaster recovery, information exchange, e-commerce, multiuser games, etc. The fundamental performance limits like throughput capacity and packet delay are major distresses to settle on since the topology is unreliable. The mutual interference of concurrent transmission between nodes is the major hindrance to achieve the reduced packet delivery delay and optimized throughput with the available two hop relay protocols in MANET environment [1] [2] .

The capacity theory for the wireless network is an exigent obstruction that stunts the development of MANET. Using the efficient two-hop routing scheme and its variants, the throughput capacity has been improved appreciably [3] . In two-hop routing scheme, initially the source no deal ways transmits the specified packet to a destination node if it is within the hearing range. When the destination node is not within one hop region, the source transmits the packet to the nearest neighbour node (relay) and the relay node in turn transmits it to the destination as soon as it comes in contact with the destination. They suggest that rather than fixed nodes mobile nodes increases the throughput capacity of the wireless network. Their work on two hop relay protocols supports applications with loose delay constraints such as electronic mail, database synchronization but does not hold up applications like video streaming, real time monitoring that are sensitive to delay. Figure 1 illustrates the Mobile Ad Hoc Network model.

Two hop communication protocols do not reflect on the packet redundancy. The exposed order of two hop relay and acknowledged redundancy as suggested by [4] shows that the packets have more than one (multiple) copies while transmitting the packets from source to destination through relay nodes. In case of out of order reception, the reception occasion takes full advantage but the mobile nodes have to carry in buffer with contains very big used to accommodate all the arrived packets, which is not practically possible in MANET [5] . Another disadvantage with this case of reception incorporates the early arrival of packets, which becomes expired after the meticulous interval of time [6] .

The main purpose of our proposed work compared to existing work is used to attain the exact optimum throughput for every node capacity with reduced delay in accordance with packet redundancy and power control [8] . To accomplish this throughput for every node capacity with delay control a general theoretical framework based on Routine Response Control and the Modified Markov chain model is initially developed [9] . The generalized group based routing scheme for two-hop with r-cast (2HR-(r, g)) algorithm is worn where limited number of copies for each packet is dispatched to different relay nodes (i.e., each packet has a restricted redundancy r) and all packets are inward bound in order in the destination if it is among the group, the destination is requesting. The optimum setting of the transmission range and packet redundancy limit is initially determined. Once the above factors are decided the next step is to determine the group size of the packets [10] - [13] .

SR State: (Source-to relay state) When the Source delivers the i^th copy to the relay node I the destination does not receive the fresh packet from any other relay node.

RD State: (Relay-to-Destination) Some relay node delivers the fresh copy to the destination when S fails to deliver the i^th copy to a new relay node.

SR + RD State: (Source-to-relay and relay-to-destination) Both the transmission turns out simultaneously.

SD State: (Source-to-Destination) S directly transmits a fresh packet to the destination.

Algorithm 1 2HR-(r, g) Algorithm

if the node S gets a transmission opportunity then

if the node D is among the one-hop neighbours of node S then

S executes Procedure 1 with D;

{

source-to-destination transmission

}

else

S randomly selects one node (say K) from its one-hop neighbors;

S flips an unbiased coin;

if it is the head then

S executes Procedure 2 with K;

{

source-to-relay transmission

}

else

S executes Procedure 3 with K;

{

relay-to-destination transmission

}

end if

end if

end if

Process 1 SD

S initiates a handshake to obtain the 𝑅G(D) and IN(D) from node D;

if SG ≥ 𝑅G(D) then

if S𝑁(Pt) > I𝑁(D) then {Pℎ is the head-of-line packet at the local-queue of S}

S retrieves from its already-sent-queue the packet P with (P) = I𝑁(D);

S sends the P to node D;

else if S𝑁(Pt) == I𝑁(D) then

S sends Pt directly to node D;

else

{

I(D) = S𝑁(Pt) + 1

}

S sends to node D the packet waiting right behind Pt in the local-queue;

end if

S deletes all packets with SG < 𝑅G(D) inside the already-sent-queue and local-queue;

S moves ahead remaining packets waiting at its local-queue.

Process 2 SR

S initiates a handshake with node K;

if K has one copy of Pt then

S remains idle;

else

S sends a copy of packet Pt to K;

if f copies have been distributed for packet Pt then

S puts Pt to the end of its already-sent-queue;

S moves ahead the remaining packets in its local-queue;

end if

K puts Pt at the end of its relay-queue dedicated to node D;

end if

Process 3 RD

S initiates a handshake to obtain the 𝑅G(K) and IN(K) from node K;

if S has a packet P in the relay-queue dedicated to K with SN(P) = IN(K) then

S sends packet P to node K;

else

S remains idle;

end if

S deletes all packets with SG ≤ 𝑅G(K) from its relay-queuededicated to K;

Every time node S wins a transmission opportunity, it operates as follows.

Step 1: (“Source-to-destination”) If S transmits directly to D a new packet of group RG(D). A packet is called a new packet if it has not been received yet by its destination.

Step 2: Otherwise, S randomly chooses to perform one of the following operations with equal probability:

・ Source-to-relay: It randomly selects one node, say R, from the one-hop neighbors, and checks whether R is a new node. If not, it delivers to R a copy of the head-of-line packet Ph at its local-queue;

・ Relay-to-destination: The node S acts as a relay and randomly selects a node (say K) from the one-hop neighbors. After obtaining the RG(K) and IN(K) from K, S checks whether there exists a fresh packet of group RG(K) in its relay-queue specified for K. If so, it transmits this packet to K and removes all packets with SG ≤ RG(K) from its relay-queue for K; otherwise, it remains idle for this time slot.

The whole process of the proposed algorithm is illustrated in Figure 3.

In Figure 4 the blue dotted line indicates the existing system which shows the exact throughput for every node capacity ω(m, r) with minimum transmission power.

where

XS(k): The time at the local queue.

XD(k): The time at the virtual queue.

m: Transmission power, r: Redundancy limit.

The red line indicates the proposed system which indicates the increased throughput (packets/slot). Using the above throughput result, the optimal throughput capacity is obtained for any r and a

fixed m. From this result the maximum throughput capacity for any r and m of a network is calculated using Formula (2).

where

ω^* does not increase any more when υ increases beyond some threshold because the node is able to cover the whole network region and destination receives each packet. The Transmission power of each node is increased which increases the throughput capacity.

Average packet delivery delay of one packet T_p is calculated using the formula

Expected packet delivery delay E{T_p} is determined using the below formula

where, g: group size; k: D has received k of g packets; N: Markov chain matrix.

The packet delay can be reduced using Markov Chain Framework. This can be achieved through the 4 states with the increasing Throughput in Figure 5. Table 1 describes the Energy vs Delay in detail (Figure 6).

Figure 7 describes the Number of Packets Sensed vs Residual Energy for the Existing System and the Pro-

posed system and Figure 8 describes the Routing Overhead vs Number of nodes for the Network with the specific Lifetime.

The throughput for every node capacity of MANET is determined along with delay control by adjusting the transmission range, limiting the packet redundancy and grouping the packets. A generalized group based two- hop routing scheme with f-cast (2HR-(f, g)) algorithm is used where each packet waiting at the source is delivered to at most f distinct relay nodes (i.e., each packet has a limited redundancy f) and can be received in order at its destination if it is a fresh packet and is among g packets of the group, and the destination is requesting. Transmission power of each node is manipulated to amend to specified transmission range and limited packet redundancy limit. Increasing the transmission power increases the throughput capacity and minimizes the delay. This in turn increases the performance of the network. Tuning parameters like f, v and g carefully, the optimized throughput capacity of each node along with minimized delay is achieved which helps in delivering the packets efficiently in the dynamic network topology.

Future research work is to derive the throughput for every node capacity with minimum delay in which each node has constrained buffer space, and further explore the possible thresholds of buffer size.

Dr. K. Saravanan,Dr. S. Silas Sargunam,Dr. M. Rajaram, (2016) Accomplishing Consummate Throughput with Delay and Power Control in MANET. Circuits and Systems,07,1834-1844. doi: 10.4236/cs.2016.78158