Modified Ant Colony Optimization Algorithm Computer Science Essay

Published Date: 02 Nov 2017

Abstract:-

A mobile Adhoc Network is a self-configuring infrastructure-less network of mobile devices connected by wireless. In MANET, nodes are freely move in any direction.So it frequently changing the links. The power consumption and routing are the main issues in MANET. A routing protocol in MANET has to face the challenges of dynamic topology with low power transmission. Nowadays, mobile devices are equipped with low power batteries. It could be impossible for a mobile device to sustain for a long time, if it send and receive data more often like in the wired network. In this paper, an efficient technique with modified ACO is used to solve the power management and routing. Zone Routing Protocol combines the benefits of proactive and reactive routing protocol by maintaining an current topological information of a zone. In intra-zone direct communication is possible. In inter-zone routes are discover by global updating rule.

Keyword:MANET,ZRP,IARP,IERP,ACO.

1.Introduction:-

MANET is a collection of mobile nodes dynamically form temporary network without pre-existing network intra or centralized administration. These nodes can be arbitrarily located and are free to move randomly at any given time, thus allowing network topology and interconnections between nodes to change rapidly and unpredictably.

Majority research has concentrated on proactive or reactive routing protocol for data transmission, improving performance metrics and on the security threats of the protocol by making change in it.

Proactive routing protocol (table driven)provide fast responses to topology change by maintaining routing information for all network destination and react to changes in the network.Reactive routing protocol(on-demand) provide routing information on a need-to-basic. It reduces the signal overhead incurred in maintaining routing tables compared to proactive approaches.

Hybrid is the combination of proactive and reactive protocol. In this project, we use ZRP protocol.ZRP is framework by using it, we can take advantage of both table driven and on-demand protocol according to the application. The local neighbors forms zone, each zone have different size and it overlaps with each other. Selection of zone radius does not depend on the location information. The length of the Radius is α, where α denotes number of hops within the zone. IARP is used to maintain the route information about neighborhood nodes and links for managing the energy resources.IERP is used for route maintenance and route discovery with the help of modified ACO algorithm.

2.Related work:-

This section reviews the some of literature pertaining to routing and energy management in mobile adhoc network.The energy management deals with the process of managing energy resources by means of controlling battery discharge in order to increase the lifetime of the nodes[1].The ACO based routing protocol give the feasible path by continuously checking the routing table. it also reduce the control overhead by limiting the number of control packet[2].A well distribution strategy of initial ants and heuristic parameter updating to avoid the stagnation behavior and premature convergence by updating heuristic parameter[3].Ant colony optimization with power saving technique introduce the energy management and finding the feasible path in MANET. The number of control packet controlled by reducing the bandwidth and minimizing the power consumption .In this stochastic policy is applied for choosing the next node to the destination[4].

The amount of route query traffic is reduced by combining border casting query detection and early termination. It reduces congestion overhead and it increases the reliability and performance based on the ZRP[5]. Detection and prevention of overlapping query message is done achieved by query packet ID. The route discovery process for finding destination through MDVZRP. when establishing a new route,SBZRP reduce the network load by limiting the control packet[6].An extended ZRP to append the service information to its routing message. The EZRP reaches highest service discoverability[7].The secure ZRP is provided to prevent multiple black hole attack in MANET.The performance was evaluated by QualNet simulator V4.It also prevent single malicious and multiple malicious black hole attack[8].

It proposed the MDVZRP for finding the shortest path and reduce the control overhead as well as it does not require periodic update of the routing table[9]. It introduces the scalable routing in MANET through zone routing[10]. It introduces Independent zone routing protocol which permit adaptive and distributed configuration of optimal size of each and every node within the routing zone. A query control mechanism is used for detecting and prevention of overlapping query message[11]. ACO is used to find best shortest path and congestion avoidance by incorporating congestion metrics in route discovery and maintenance[12].The hybrid zone routing protocol analyzing route acquisition delay and quick reconfiguration during link failure[13].

3.Proposed System:-

The major issues in MANET`s are energy management, routing and security.ZRP overcomes the disadvantage of proactive and reactive such as bandwidth constraints, scalability, inadequate service. ZRP divides it’s network in different zones. It consist of two zones routing protocol namely IARP and IERP .IARP maintains route about neighborhood nodes and links for managing energy resources. IERP discover and maintain the route by using ACO algorithm when IARP doesn’t have a path to destination node.

ZRP protocol creates zones with respect to hop count to cover an entire network. But the main issue is power consumption.For example the node uses maximum power for forwarding a packet to the destination. Due to this ,the node will lose its full power in a very short duration to avoid this problem, we use the modified ZRP protocol.In modified ZRP protocol we have to create zones with respect to two power levels .For example 30mw and 50mw.

The main reason for creating a zone with 2 power level is that,if a node is selected as a border node and if the node is moving at particular speed. Due to this approach,the node can avoid the unnecessary routing updates.

3.1 ZRP:

It is the combination of both proactive and reactive protocols.It overcomes the problem of excess bandwidth

utilization in proactive and inefficient flooding in reactive protocol. It also referred as hybrid routing protocol. All the nodes in the zone maintains the topology information. The routing zone as a radius r, where it represents in hop. If the destination is present in an intra-zone then direct communication is possible, else the source node continuously broadcasting the route request until it find out the destination node.

ZRP uses NDP(Neighbor Discovery Protocol) to discover neighbor nodes and detect link failure.NDP announces "HELLO" message at regular interval.The nodes which received this message is comes under the neighbor nodes list.

3.2 BRP:

It stands for Bordercast Resolution Protocol. It is used to direct the query message to peripheral node. The bordercast service provided by BRP with the help of constructing bordercast tree. The BRP act as an interface between peripheral nodes and interior nodes. Query source and destination address, the query ID and previous bordercast address present in BRP packet.

3.3 Query Control Mechanism:

It used to avoid redundant query request/reply messages. ZRP uses 3 query control mechanism such as

1. Query detection,

2. Early termination and

3. Random query processing delay.

Normally the zones are overlap. In the first method, the source node send query to border node. Next this node broadcast to its neighbor nodes. So there is a chance to receive the same message by source node.BRP uses two methods such as QD1 and QD2. In QD1,the nodes have capable to detect the redundant query. In QD2, listen the traffic of other nodes.

In early termination method, a node prevent request from an already entering regions. It requires topology information to extend the size of the region.

A random query processing delay is used to reduce the arrival of same message from different nodes.

3.4 Routing:-

Routing is a major issue in MANET. Proactive routing protocol (table driven)provide fast responses to topology change by maintaining routing information for all network destination and react to changes in the network.

Reactive routing protocol (on-demand) provide routing information on a need-to-basis. It reduces the signal overhead incurred in maintaining routing tables compared to proactive approaches.

Hybrid is the combination of proactive and reactive protocol. In this project, we use ZRP protocol.ZRP is framework by using it, we can take advantage of both table driven and on-demand protocol.

3.4.1 Proposed Algorithm:

Ant Colony Optimization:-

The ACO is based on the definition of the ant’s behavior. ACO adapts the behavior of real ants. Multiple paths are available from food to nest, ants do random walk initially. During their trip , they lay a chemical substance called pheromone, which serves as a route mark that the ants have taken . Subsequently, the newer ants will take a path which has higher pheromone concentration and also will reinforce the path they have taken. As a result of this autocatalytic effect, the solution emerges rapidly. Ant Colony Optimization (ACO) is used to find feasible path from source to destination. A neighbor discovery protocol is added in route maintenance phase to deal with mobility. Heuristic parameter updating method is also used to solve this problem.In initial stage, there are lack of information is present.So that heuristic parameter value is increased.At end stage,the heuristic is set as small value.

The entropy of a random variable is defined as

K

E(X) = -  Pj log Pj

J=1

where pi represents the probability of occurrence of each trails in the pheromone matrix. For a symmetric n cities TSP, there are m(m-1)/2 distinct pheromone trails and k= m(m-1)/2. The maximum entropy is given by

K

Em = -  Pj log Pj

j=1

K

= -  m(m-1)/2 log m(m-1)/2

j=1

Em= log[m(m-1)/2]=log K

using entropy value to get the detail about how much information is gather from pheromone trails and the heuristic parameter can be updated according to the rule is given by

5 threshold X < Ev <= 1

α = 4 threshold Y < Ev <= threshold X

3 threshold Z < Ev <= threshold Y

2 0 < Ev <= threshold X

Ev=1-[(Em-Ecur)/Em]

Ev is the Entropy value for the current pheromone matrix and X,Y,Z are threshold according to the city size.

Algorithm:

The proposed algorithm is combined well distribution strategy of initial ants and dynamic updating of

heuristic parameter. The proposed algorithm is described as follows:

Procedure Proposed ACO algorithm for TSP

Set parameters, initialize pheromone trails

Calculate the maximum entropy

Loop /* at this level each loop is called iteration*/

Each ant is positioned on a starting node according to distribution strategy (each node has at least one ant)

For k=1 to m do /*at this level each loop is called a step */

At the first step moves each ant at different route

Repeat

Select node j to be visited next (the next node must not be visited by the ant) according to

A local updating rule [7] is applied

Until ant k has completed a tour

End for

Local search (2-opt, 2.5 opt) apply to improve tour

A global updating rule [8] is applied

Compute entropy value of current pheromone trails

Update the heuristic parameter

Until End-condition

End

3.5 Route discovery:

In this process new routes are created.There are two ants FAnts and BAnts.The FAnt establishes the pheromone track to source node.The BAnt establishes the pheromone track to destination node.The FAnt contains packet with unique sequence number, which is used to detect duplicate packet and address of source node. This ant is broadcast by source and it is forward by neighbor nodes. A node which receives FAnt message then it creates record in routing table. This record contains destination address, next hop, pheromone value. The node interprets the source address of the forward ants as the destination address,the previous node address as the next hop, the number of hops the FAnt needed to reach the destination node as pheromone value. After receiving the packet by destination ,it extract the information of FAnt and destroy it.Subsequently it creates the BAnt and send it to the source node. When process is complete, a route is established and data packets are sent.

Route maintenance:

It is important in MANET.When there is a broken link along an active path between Source (A) and Destination (B) within radio coverage, a local path repair procedure is initiated.

Repairing is the process of establishing a new path from source to destination.

Since each node has its own routing zone, the routing zones of neighbouring nodes overlap heavily.

Since each peripheral node of a zone Fant request message, the message can reach the same node multiple times without proper control.

Each node may forward the same request multiple times.

Route failure handling:

It is a last phase in ACO. It de-activate the route when a route error message is received.

Energy Management in MANET:

It increases lifetime of the nodes.In this paper, it achieves by using ZRP protocol. There are two power values used for intra(30mW) and inter(50mW) based on the radius. Each node consumes power to transfer the data.In intra-zone direct communication is possible between source and destination,so we fix low power. For inter-zone, there are several intermediate nodes are present, so it consumes more power compare to intra-zone routing.To obtain shortest path in inter-zone, the ACO algorithm is used. Lifetime of nodes increases compare with existing methods.