Ant with Artificial Bee Colony Techniques in Vehicular Ad-hoc Networks

ABSTRACT


INTRODUCTION
The node in the vehicular network desires to send the packet to another node in the network, initially make sure the available routes in the cache.If routes found, then check the shortest path and send the packet through the routes.If no routes found, then broadcast the packet through all the routes and then find the best path [1].This way is similar to the ant colony optimization technique.Just as biological ants initially disperse from their colony in various directions to find food and then return to deposit pheromones on their path to inform others, our routing approach involves Request ant packets dispersal throughout the network via a novel route discovery system.These packets collect information about the route, such as its total length, congestion levels, and end path reliability, as they make their way to the end user node.Upon receiving a Request ant packet, the target node send back a Request ant (Route Reply control packet) carrying the corresponding route details to the source node, following the same route [2] [3].
When Reply Ant packets are received from different routes, the source node acquires knowledge about these routes.Analogous to the ant colony approach, the most favorable route is decided by the concentration of pheromones along that route.Similarly, we gauge the pheromone level of a route by considering elements such as the number of hops, congestion on the path, and the route's end-to-end reliability [4].The route with the greatest pheromone accumulation is selected for the transmission of data packets.Many researchers have been developed various techniques to find the best route from source to the destination but produces less Quality of Service due to the dynamic changing of the network in VANET.The proposed techniques used Ns2 simulator to check the performance level.Due to the failure of dynamically changing vehicles movement, battery collapse, VANETs require an optimal routing techniques tree being constructed as fast as possible, and runtime becomes an important measure [11].

LITERATURE REVIEW
Ant Colony Techniques (ACT) proves to be an exceptional approach for identifying efficient pathways for packet delivery in vehicular ad-hoc networks.ACT employs a method where pheromones are laid on the ground to create trails, attracting other "ants."Pheromones evaporate more swiftly on longer paths, while shorter paths become the preferred route for most other ants in search of sustenance.The applications of Ant Colony Techniques are widespread, encompassing areas such as network modeling problems, security concerns, and vehicular routing.The ultimate objective of ACO (Ant Colony Optimization) is to determine the best path-shortest path from source to destination [8] [9].
ACO-ABC techniques for various application areas such as sales man problem, vehicular ad-hoc network, mobile ad-hoc network, scheduling techniques and image processing methods etc., and the parameters consider in to an account as number of iterations, number of ants and Pheromone evaporation rate [12].This Algorithm produces high execution speed and accuracy in both local and global search modes but produces more residual energy.The hybrid algorithm called as Ant colony optimization and artificial bee colony optimization (ABC-ACO).The proposed algorithm is mainly deliberate for transmitting the information from one place to another place that is routing process 13].This algorithm minimizes the traffic and failure of link through the routing process by using the simulator ns2.The hybrid algorithm proves that produce better result in terms of packet delivery ratio and delay but lack in throughput and residual energy [14].The framework fuzzy logic ABC based algorithm is especially designed for VANETs.The proposed method proves the quality of services efficiency when compare to another existing method in achieves higher optimization accuracy and minimum delay but with the highest energy consumption [15].
During each phase, the ant's shift from one location to another, adhering to specific guidelines, including: • Each ant is required to explore every path precisely once.
• Paths that are farther away have a reduced probability of selection.
• As an illustration, ants leave pheromone traces on paths connecting the source and destination.
• Upon concluding its expedition, an ant leaves additional pheromone marks on all paths it traveled, particularly • if the journey was brief.
• Following each cycle, the pheromone trails gradually dissipate.
ABC algorithm is the best popular method for optimization algorithm and also they solved many of real time scenarios in vehicular ad-hoc networks but the scheduling problems remains in the VANET.According to the survey, many researchers lack in the performance of energy consumption.The proposed AABC shows that the results are outperforms when compare to the other conventional methods, ACT and solve the problem of energy consumption.

METHOD
In comparison to existing evolutionary algorithms, one notable advantage is that the Ant with Artificial Bee Colony (AABC) algorithm incorporates a more effective local search mechanism, enhancing the quality of solutions.The Ant colony algorithm has proven its competence in addressing intricate issues such as digital filter design [7] minimum spanning tree problems [6], and capacitated vehicle routing challenges [5] [10].
The inception of the ant colony algorithm was introduced by Karaboga (2005) with the aim of tackling complex optimization problems.It is a relatively recent approach inspired by the foraging behavior of honey bees.The main objective involves minimizing the weighted sum of transfer counts and the collective travel time of passengers through the reconfiguration of vehicles and the determination of new frequencies.The figure 1 shows the AABC techniques.The OB-onlooker bees are tasked with collecting information from the worker bees regarding the food source, including details like the distance from the hive and the path to the source.They then make decisions about the chosen source of food based on this gathered information.The worker bees are the ones responsible for discovering and exploiting available food sources.Once a worker bee has finished exploiting a food source, it transitions into a scout bee role, autonomously seeking out new food sources in a random manner.
The Ant with Artificial Bee Colony (AABC) algorithm is founded on two essential techniques such as two types of process are Variation and Selection.The variation process assesses the availability of diverse search space regions, while the selection process guarantees the utilization of past experiences.The AABC algorithm follows four different phases of cycle: initialization phase, employed bees phase, onlooker bee's phase, and scout bee phase.

Initialization phase
The food source is generated as X s n = x s min + rand (0,1) (x s max -x s min) x= food source n= no of food source s= direction of food source rand (0, 1) = Once a worker bee has finished exploiting a food source, it transforms into a scout bee, embarking on a random search for new food sources along all boundary directions, spanning from x s min to x s max.

Employed bees phase
The worker bees are capable of retaining the location of the most recent food source while also computing the fitness value of the current food source.This fitness value is established based on the quantity of nectar.In the event that the fitness value of the new food source surpasses the fitness value of the previous source, the bee abandons the old position and substitutes it with the new position of the food source.This process involves determining the updated position.

Onlooker bees phase
The onlooker bee's phase commences once the employed bees phase is concluded.During this stage, all employed bees communicate the updated position details to the onlooker bees within the hive.The onlooker bees assess the food sources using the information gathered from the employed bees and subsequently determine the optimal position to search for food sources.
The process involves identifying the most favorable food source position.
FiGP is a fitness value of Gathered position BP -Best Position ∑ N i=1 -No of gathered position from all employed bees.
The onlooker bees evaluate the fitness of all collected positions.Should a position exhibit greater fitness than the preceding one, the bee stores the new position in its memory while discarding the old one.Consequently, the onlooker bees make adjustments to the stored position in their memory.

Scout bees phase
Once the employed bee process transitions into the scout bee phase, the sequence initiates anew, commencing with the initialization phase.This entails the generation of a fresh food source through random selection across all directions.X s n = x s min + rand (0,1) (x s max -x s min) The onlooker bee employs the same route to locate the food source.When the food source is depleted, the scout bee initiates a random search for a new solution.

SIMULATION SETUP
The simulation is performed in a simulation area 5500x5500m with 30 nodes.The results are produced under the simulation study of the network using NS2 simulator.Each input parameter for the simulation as illustrated in the table

Routing method AABC
In this scenario each simulation was performed for 499.00 seconds.The AABC Method using IEEE 802.11p is implemented using the NS2.35 simulators.100 nodes were selected as the participants of the network and each node movement was highly mobile.Each node equipped with 802.11p wireless module and packet size of each packet is 160kb.Distances among the vehicles are randomly selected.

RESULTS AND DISCUSSION
In this section, average delay time, energy consumption and packet delivery ratio are investigated with respect to certain factors such as packet sending rate, distance between the source and the destination when the source generates the first packet etc. simulation settings and the simulations are carried out using NS2.Simulated results of the existing conventional algorithm, Ant Colony Algorithm and the proposed AABC are depicted.Three metrics which were used to evaluate the performance of AABC are as follows: Figure 2 illustrates the average packet delivery ratio generated by AABC Technique and the performance is evaluated by existing ACT, CONV.The packet delivery ratio of AABC is high when compared to the existing schemes such as ACT and CONV.

Average End to End Delay
Average end to end delay is the average time taken by a data packet to arrive at the destination.The lower value of end-to-end delay indicates better performance of the protocol.The selected techniques to check the performance are ABC, CONV and AABC.

CONCLUSION
The existing techniques use the concepts of ant colony algorithm for finding the best optimal route and also produce more improvement for all parameters taken into the consideration are packet delivery ratio, throughput, and delay and energy consumption.The proposed AABC algorithm produces more improvement especially the excepted parameters is energy consumption than another ant bee colony algorithm.Vehicular ad-hoc network finds the best path scheduling techniques through the Ant with Artificial Bee Colony Techniques.The parameters are taken in this techniques are channel utilization, throughput, delay and processing time.The simulation results shows that the proposed techniques will assure best quality of service when compare to the existing techniques.The future research work is compared with other best techniques and experiments the work in real world environment.

Figure 1 .
Figure 1.Overview of AABC 3.1.Behaviour of Honey Bee in Nature The Ant with Artificial Bee Colony Techniques encompass three distinct bee types • OB-onlooker bees • EB-employed bees • SB-scout bee

Figure 3 .
Figure 3. Average end to End Delay Figure 3 illustrates the average end-to-end delay generated by AABC Method and the performance made by existing ACT, CONV and proposed AABC.The end-to-end delay of AABC is low when compared to the existing schemes 5.3.Energy Consumption

Figure 4 .
Figure 4. Energy Consumption Figure 4 illustrates the Energy consumption generated by means of AABC Method and the overall performance made by existing CONV, ACT and proposed AABC.The energy consumption of AABC is low when compared to the existing schemes CONV and ACT.

Table 1 .
Simulation Environment for AABC