Improving best route using intelligent Ad Hoc system

ABSTRACT


Introduction
A wireless net is a new technique that makes the users connect and provided the data between them anywhere.Wireless networks are classified to infrastructure less (ad hoc) and infrastructure network [1].The infrastructure network is wired and fixed gateways.The portable mobile connects with the network.When the mobile moved to another area, it will disconnect from previous and connect with another network that in its range [2], which called handoff.Bluetooth is a new wireless network and helps the mobiles to connect and transform the data called ad hoc networks.Mobile networks don't need infrastructure.The mobile network is a horizontal and quick network; it could be utilized in all conditions [3].An ad hoc network is established for military purposes after 1990 combined with Bluetooth and wireless LAN [4].At 1997 several interesting internet groups are using Mobile Ad hoc Network with the work regulate protocols because they thought that the protocols doing the work more efficiency [5].Nowadays, more than (50) protocols are using in the wireless network such as ad hoc network that related with mobile nodes.Ad hoc network is several portable computers can connect outside the direct wireless.Ad hoc networks didn't need centralized management or infrastructure and characterized by such as inexpensive, easy access, quick [6].

Ad Hoc
In computer network, Ad Hoc is means is a wireless network and don't need an infrastructure, sometimes called spontaneous network.Reverse some of the other networks that required infrastructure such as Wireless Local Area Network (WLAN) and cellular network [7].In the network that required infrastructure it could send the packets to another node using the access point.The access point can provide local network and make the distributed nodes in its area can be connected.If there any dropin connection, the nodes will lose the network.Many reasons making access point's services are lost such as install access point required a long time and cost factor, therefore it is necessary to build a network by using these nodes, called the ad hoc network [8].It only needs transceiver and equipped without infrastructure, and it can do its communication with other nodes.The nodes can connect with other nodes that exist in another area in the infrastructure based wireless network and transferring the data [9].The ad hoc networks are not effective if compared with a large network such as infrastructure network wherever; the node has a limited range if the nodes ranged are combined, that would produce large data transmission area [10].

Genetic Algorithms (GA)
The cell is the primary unit of living organisms.The entire cell has the same chromosomes.The chromosomes carry the genetic material and encoding the body, wherever it consists of genes [11].The gene encoded the protein.For example, a particular gene is encoding the colour of eyes (blue, brown).Each gene has a site (locus) in the chromosome [12].

Concepts of GA
The Genetic Algorithms are means natural selection and natural genetics.GAs simulated the processes in natural evolution.The process was included operates on chromosomes (the element can encode the living being structure) [13].The GA is searching among a population of points, and it differs from other search methods.Also, it uses data of the objective function without gradient data.The traditional ways use gradient data; wherever the transition of the GA is probabilistic [14].It was used as a general optimization of the algorithm.Furthermore, it provides methods for searching in the irregular area and could Appling it for optimization of the machine learning applications, optimization of the function, parameter estimation [15].

Fitness Function
Genetic algorithms are useful for solving large or maximization problems.The maximization problems are converted to minimization problems.The fitness function as shown in equation 1 is derived from the objective function and used in successive genetic operations.For example, fitness pheromone is used in determining reproductive properties.The fitness function could represent an objective function for minimization problems; the following transformation is possible.

1-Elitism
Our study has great chance, if try to form new population by mutation or crossover, that resulting losing the chromosomes.Elitism is new method creating copies the best chromosome [17].Elitism prevents losing the best-found solution therefore, it very rapidly in the performance of GA [18].

2-Selection
Chromosomes are elements selected and moved among the generations from parents to a new generation.Theory of Darwin's evolution is explaining selection by the individual that can survive and create a new generation.Many of methods of the selection including the most common methods are roulette wheel method [18].

 Roulette Wheel Selection
The fitness is determined and chooses the parents wherever it was done selecting the best chromosomes.Figure 1 is showed roulette wheel methods depend on the percentage of the chromosomes and it placed accordingly to its fitness.Fig. 1: Roulette Wheel Selection [18] Then a small ball is thrown there to select the chromosome.A chromosome with bigger fitness will have a bigger chance to be selected more times.

3-The crossover
It is formed from mixing two strings to form another better string.The recombination creates new ones in new generations by mixing genetic material from two ones.The good strings in a population have larger of the copies.Exchanging the data among strings will create new strings.The parent is participating in produce new strings (children).Exchanging all bits randomly is performed as Figure 2 [19].Fig. 2: operation of crossover in one site crossover operation [7] The process is done randomly.Right portions are chosen for exchanging the strings to produce new strings.The new two places are different if compared with old strings as Figure 3.The information between the strings is transfered better than the parents [19].

4-The mutation
The mutation will create new data randomly.The mutation is a factor producing new diversity in the population to create new individual.The mutation makes the chromosomes changed Figure 4. Fig. 4: The basic GA operations [7] Mutation is any randomly change in genetic data.It works at a bit level; if the coping of bits from string to another string that represents changing and called mutation.The probability of mutation is called P m .If the number is limited between (0 and 1), [20] then the bit become inverted.Therefore the zero converts to one and one convert to zero.That will cause bit diversity by scattering the occasional points.The mutation is used for creating point changing.For example, see Figure 5

Implementation
The performance of the Genetically-trained Ad hoc Network to identify different nonlinear dynamical systems is evaluated here.As the performance index to be minimized by the GA, the MSE criterion was employed.For the simulation tests performed here, the parameters of the real-coded GA were set to the following values.Population size is 1000 for the first group test and 200 for the second group test Maximum number of generations is 1000, mutation probability (P M ) 0.08 and area size 10 and it was implemented in the Matlab environment.From several simulation tests, the above GA parameters settings were found to be the best values that provide the best training for all Ad hoc networks considered in this work.

Discussion and Conclusions
Through our work, this study have been able to obtain several nodes that can perform a high-performance system in communication ways to take advantage of time and short track.And create secondary towers that can perform the functions of the emergency node and take advantage of the genetic algorithm and continue to work and reduce the errors resulting from the collapse of the network regardless of the error that occurs during the failure of the node and this will make the system can maintain the path.

Recommendations and Future Work
It is possible to develop the system by integrating it with the network of military aircraft, where the error rate is very few.In addition to reducing the time and cost and choosing the most appropriate way compared to the rest of the systems.The system can link the boundaries of thirty-five towers where it can't move because the system will have a coefficient So our study suggest a way to get more by improving genes using techniques that can count between valid chromosomes.

Figure ( 5 )
Figure (5): Mutation operation [8] is: Number of barges: 45 Number of generation: 1000 Population size: 1000 Mutation Probability: 0.08 The red barge is the actual barge that selects the best course