Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


Mobile Ad Hoc Network (MANET) routing is challenged by power and bandwidth constraints as well as frequent topology changes to which it must adapt to and converge quickly. Energy efficient routing protocols play an important role in MANET. This paper presents an innovative energy aware routing protocol for wireless ad hoc network called Energy Efficient Ad hoc On Demand Multipath Distance Vector (E2AOMDV) routing protocol. Ad hoc On Demand Multipath Distance Vector (AOMDV) routing protocol presents the advantage of achieving faster and efficient recovery from node failures. We propose the inclusion of Energy Aware (EA) selection mechanism and the maximal nodal surplus energy estimation in the AOMDV routing protocol to improve its energy performance in mobile ad hoc networks. The 'readiness' concept of EA Selection mechanism prolongs the network lifetime and the maximal nodal surplus energy estimation concept increases the energy efficiency of E2AOMDV routing protocol. A comparison of E2AOMDV and AOMDV routing protocols can be done using the performance metrics like packet delivery ratio and network lifetime to show that the former outperforms the latter and confirms to be a better performing routing protocol in saving battery energy in a dense mobile network with high traffic loads.

Keywords: Ad hoc network, energy efficiency, Surplus energy, network lifetime, battery capacity.


Mobile Ad hoc Network (MANET) [1] is a dynamically reconfigurable wireless network with no fixed infrastructure. Each node acts as a router and host and it moves in an arbitrary manner. MANET routing [2] is recently the topic of extensive research. The interest in such network stems from their ability to provide temporary and instant wireless networking solutions in situations where cellular infrastructures are lacking and are expensive or infeasible to deploy. Due to their inherently distributed nature, MANETs are more robust than their cellular counterparts against single-point failures and have flexibility to reroute around congested nodes. In many ad hoc networks, each node is powered by a battery and has limited energy supply [3, 4]. Over time, various nodes will deplete their energy supplies and drop out of the network. Unless nodes are replaced or recharged, the network will eventually become partitioned. In a large network, relatively few nodes may be able to communicate directly with their intended destinations. Instead most nodes must rely on other nodes to forward their packets. Some nodes may be especially critical for forwarding these packets because they provide the only path between certain pair of nodes. Associated with each node that depletes its battery and stops operating, there may be number of other nodes that no longer communicate. Energy is scarce by the fact that the devices are mobile i.e. they must be small and therefore cannot be fitted with large battery packs. For these reasons a number of researchers have focused on design of energy efficient routing protocols [5, 6, 7].

This paper is distributed as follows - In section II we have discussed about previous such works. Section III gives details about Energy Efficient Ad hoc On Demand Multipath Distance Vector (E2AOMDV) routing protocol. The analyzed performance of the proposed routing protocol is given in Section IV. Finally, Section V summarizes this paper.


Ad hoc On Demand Distance Vector (AODV)

The AODV routing protocol [8] provides on demand route discovery in mobile ad hoc networks. Route finding is based on a route discovery cycle involving a broadcast network search and a uni-cast reply containing discovered paths. AODV relies on per node sequence number for loop freedom and for ensuring selection of most recent routing path. When a source node has data packets to send to some destination, it first checks its routing table to determine whether it already has a route to the destination. If such a route exists, it uses that route for data packet transmission. Otherwise it initiates a route discovery procedure to find a route. The source node creates a route request (RREQ) packet and broadcast it to its neighbors. When a neighbor node receives the RREQ packet it first creates a reverse route to the source node. Each intermediate node receiving the RREQ packet checks its local routing table to find if there is a valid route to the destination. If a route is available then RREP packet is generated; otherwise the RREQ packet is propagated in broadcast. Eventually if the destination node receives the RREQ packet, it produces the RREP message, which is routed back to the source by using the unicast reverse path established by the corresponding node.

Ad hoc ON Demand Multipath Distance Vector (AOMDV) routing protocol

It is an extension of AODV routing protocol. It is used to compute multiple paths during route discovery. It is designed primarily for dynamic ad hoc networks. The AOMDV protocol [9] applies a route update rule to establish and maintain multiple loop free routes at each node. A distributed protocol is used to find link disjoint paths. In AOMDV, different instances of RREQs are not discarded by intermediate nodes. If a new RREQ instance preserves the loop free condition and comes from a different last hop node, then a new reverse route towards the source node is logged in the intermediate node. If the intermediate node knows one or more valid forward routes to the destination, a RREP packet is produced and forwarded back to the source along the reverse path. When the destination receives more RREQ instances, in order to get multiple link disjoint routes, it replies with multiple RREP messages.

Maximal Minimal Residual Energy Ad hoc ON Demand Multipath Distance Vector (MMRE-AOMDV) routing protocol

MMRE-AOMDV [10] proposed by Yumie et al. exploits maximal minimal nodal residual energy concept. It balances the nodal energy consumption. This protocol finds the minimal nodal residual energy of each route in the route discovery process, then sorts the multi route by descending nodal residual energy and uses the route with maximal residual energy to forward the data packets. Each RREP and RREQ message carries an additional field called min_ re_energy to indicate the minimum nodal residual energy. At the source node, the min_re_energy is assigned the maximum value. For the intermediate nodes, the MMRE-AOMDV route update rule is applied. When an intermediate node receives a RREQ message, if its sequence number is less than the sequence number in the RREQ message, its residual energy is compared with the min_re_energy of RREQ. If it less, then the min _re_energy value in the RREQ message is updated. At the destination, reverse routes are set up. The node with route list sorted by the descending value of min_re_energy can send data packets using the route with maximal nodal residual energy.


We have proposed a multipath routing protocol that can be designed mainly for highly dynamic energy deficient ad hoc networks where link failure and route break occurs frequently. Energy management is a critical issue for the deployment of these networks because the nodes are small battery powered devices. When a single path on demand routing protocol is used in such networks, a route rediscovery is needed in response to every route break. Each route discovery is associated with high overhead and latency. This inefficiency can be avoided by using multipath routing protocol. Maintaining a routing table for multipath makes the protocol complex. This can be avoided by selecting few of the best paths from the multipath and routing data packets through them. The main idea in E2AOMDV is to select few best paths from the multipath available using energy aware selection technique and to balance the nodal energy consumption to prevent one or more critical nodes from depleting their energy supplies. The AOMDV route discovery procedure is modified to enable the selection of the best paths and for the computation of maximal nodal surplus energy. The RREQ and RREP message includes two additional fields namely readiness factor and max- surplus energy. If the readiness factor denotes 'Discard' a RERR message is propagated in the reverse path instead of RREP. The RREQ message contains the sequence number, advertised hop count, route list, readiness factor and max-surplus energy.

Four important steps are involved in designing this protocol:

1. Energy Aware selection mechanism

2. Finding the maximal nodal surplus energy along the best paths

3. Sorting the multipath in descending order using the nodal surplus energy

4. Forwarding the data packets through the path with maximal nodal surplus energy.

The source node assumes the node's initial energy as its surplus energy. At the intermediate node energy aware selection mechanism is applied.


Sequence Number

Advertised- hop count

Route list

{ next hop1, hopcount1}

{ next hop2, hopcount2}



{ next hop n, hop count n}

Expiration time


Sequence Number

Advertised-hop count

Route list

{ next hop1, hopcount1, readiness( ) , max-surplus energy1}

{ next hop2, hopcount2, readiness( ) , max-surplus energy2}



{ next hop n, hop count n, readiness n ( ) , max-surplus energy n}

Expiration time


Figure 1: Structure of routing table entries for AOMDV and E2AOMDV

Energy Aware Selection Mechanism

It is a mechanism to involve energetic considerations in best path selection. The E2AOMDV specification has a variable 'readiness' representing the availability of that node to act as an intermediate node. Each node calculates its own energetic status and declares an appropriate readiness. The readiness selection is based on battery capacity and predicted lifetime of a node. The heuristic used to associate a readiness ('Discard', 'Moderate', 'High') to a pair(battery capacity, lifetime) is shown in table 1.

Table1. Energy based Readiness Selection

Lifetime ->








Low [<0.1]




High [>0.1]


EA-Accept moderate

EA-Accept high

For example, in condition of high battery capacity, if the predicted lifetime is short a node declares EA-Discard. On the other hand, if a longer node life time is predicted, the node declares EA- Accept high readiness. In the same way, if the battery charge is low, a node is less available to become intermediate node and declares EA-Discard. This mechanism permits better load balancing to be obtained.

In our implementation, we decided to use the ratio between actual and initial energy of a node to measure its battery capacity and drain rate [16] to measure the predicted life time. We chose to consider less than 10% of residual capacity as low battery values, less than 10 seconds as short lifetime and greater than 100 seconds as long life time.

Finding the maximal nodal surplus energy

When the intermediate node receives a RREQ message, it checks if the sequence number specified in the RREQ message is greater than the node's sequence number. If so, it compares the surplus energy in the RREQ message and the surplus energy of the node. In case the node's surplus energy is greater than that specified in the RREQ message, the surplus energy variable in the RREQ message is updated with the nodal surplus energy. By this method we are able to achieve the value of maximum surplus energy among all nodes in the specified path. The reverse paths are set up just as in AOMDV. The pseudo code to implement E2AOMDV in the AOMDV is shown in figure 2.

Sorting multipath and forwarding data packets

The paths in the route list are sorted by the descending value of surplus energy. The path with the maximum surplus energy is chosen to forward the data packets. Once the source node receives the RREP message containing the new path with maximum surplus energy, it forwards the data packets through this path.


Simulation Environment

To evaluate E2AOMDV, we are using the ns-2 network simulator [12]. A dense wireless network of fifty nodes is simulated in a field with 870m*870m area. Each simulation has duration of 400 seconds. During each simulation, 12 constant bit rates (CBR) connection are generated, producing 4 packets per second with a packet size of 512 bytes. The 'Random Waypoint' model is used to simulate node movement. Each node starts moving from its initial

s=0; n=i; for i=1to d

if n !=d then

Battery capacity=actual energy/initial energy;


if (drain rate( )!=0) then

Lifetime=actual energy/ drain rate ( );

if lifetime<10s or battery capacity <0.1joule then

Readiness ( ) = 'EA-Discard';

else if 10s <lifetime <100s and battery capacity >0.1joule then

Readiness ( ) = 'EA-Accept Moderate';

else if lifetime >100s or battery capacity >0.1joule then

Readiness ( ) = 'EA-Accept High';

endif; endif; endif;

for i=n; j=n+1;

if readiness (i) and readiness(j)='EA-Accept Moderate' or 'EA-Accept High' then

if (sequence numberid<sequence numberjd) then

sequence numberid<=sequence numberjd;

if (i!=d) then

if(surplus energyi>max-surplus energyjd) then

max-surplus energyjd = surplus energyi

advertised -hopcountid = ∞;

route-listid= NULL;

insert(j, advertised -hopcountjd+ 1, max-surplus energyjd) into route-listid;


advertised -hopcountid = 0;


else if (sequence numberid= sequence numberjd) and ((advertised -hopcountid, i) >(advertised -hopcountjd,j)) then

if(surplus energyi>max-surplus energyjd) then

max-surplus energyjd = surplus energyi;

insert(j, advertised -hopcountjd+ 1, max-surplus energyjd) into route-listid;

end if; endif; endif;

Figure2. E2AOMDV route update rules

position to a random target position selected inside the simulation area. When a node reaches the target position it waits for a pause time, then selects another random target location and moves again.

We adopt the energy model as implemented in the ns-2.1b8 version. In this model, each node has an initial energy level and a given energy usage on every packet it receives and transmits. The distributed coordination function (DCF) of IEEE 802.11 [13] for wireless LANs is used as the MAC layer. The radio model uses characteristics similar to a commercial radio interface, Lucent's Wave LAN [14]. Wave LAN is a shared-media radio with a nominal bit-rate of 2 Mb/sec and a nominal radio range of 250 meters. The two-ray-ground reflection model is used to propagation model. The default value for the transmitting power as well as the receiving power is 281.8 mW. The energy consumption during the idle time is not considered in this model. In our simulations, we set the initial energy as 60.0 joules per node.

Performance Analysis

To show how EA Selection Mechanism improves the performance of a MANET, we plotted the expiration time of connections [6] as shown in figure 3.

Figure3. Expiration Time of Connections

We assess two key performance metrics:

i) Packet delivery fraction is defined as ratio of the data packets delivered to the destination to those generated by the CBR sources.

ii) The Network Lifetime is defined as the duration from the beginning of the simulation to the first time a node runs out of energy.

Figure 4 shows the performance of MMRE-AOMDV routing protocol and AOMDV routing protocol. We are expecting E2AOMDV routing protocol to guarantee a longer lifetime than shown in figure 4.a for every node in the network because nodes with lower residual energy are not stressed in this routing method. In figure 4.b, we plotted the packet delivery ratio over the max speed for AOMDV and MMRE-AOMDV. Under all the max speed conditions, we are expecting E2AOMDV to lose fewer packets and outperform the classical AOMDV protocol because E2AOMDV prevents the critical nodes from acting as the intermediate node.

a) Network's Lifetime

b) Packet Delivery Ratio

Figure4. Network's performance


In this paper we have proposed a balanced energy efficient routing protocol for mobile ad hoc networks. E2AOMDV routing protocol is an extension of the existing multipath routing protocol AOMDV. The routing model uses the energy aware selection mechanism and the maximal nodal surplus energy concept to prolong node's lifetime and preserve node's battery capacity. It prevents the critical nodes from depleting their energy earlier and avoids route rediscovery for every route break. The performance analysis shows that E2AOMDV can perform better than the existing routing protocols.