- Email: [email protected]
The performance of a hybrid routing intelligent algorithm in a mobile ad hoc network
Computers and Electrical Engineering 40 (2014) 1255–1264
Contents lists available at ScienceDirect
Computers and Electrical Engineering journal homepage: www.elsevier.com/locate/compeleceng
The performance of a hybrid routing intelligent algorithm in a mobile ad hoc network q B. Nancharaiah a,⇑, B. Chandra Mohan b a b
Department of ECE, JNT University, Hyderabad, Andhra Pradesh, India Department of ECE, Bapatla Engineering College, Bapatla 522101, Andhra Pradesh, India
a r t i c l e
i n f o
Article history: Available online 26 February 2014
a b s t r a c t End-to-end delay, power consumption, and communication cost are some of the most important metrics in a mobile ad hoc network (MANET) when routing from a source to a destination. Recent approaches using the swarm intelligence (SI) technique proved that the local interaction of several simple agents to meet a global goal has a significant impact on MANET routing. In this work, a hybrid routing intelligent algorithm that has an ant colony optimisation (ACO) algorithm and particle swarm optimisation (PSO) is used to improve the various metrics in MANET routing. The ACO algorithm uses mobile agents as ants to identify the most feasible and best path in a network. Additionally, the ACO algorithm helps to locate paths between two nodes in a network and provides input to the PSO technique, which is a metaheuristic approach in SI. The PSO finds the best solution for a particle’s position and velocity and minimises cost, power, and end-to-end delay. This hybrid routing intelligent algorithm has an improved performance when compared with the simple ACO algorithm in terms of delay, power consumption, and communication cost. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction A mobile ad hoc network (MANET) is a self-configuring infrastructureless network. Communication in MANETs is typically performed with the aid of temporary multi-hop relays, i.e., a source node uses its neighbouring node as a relay router. Routing is the act of moving information from a source to a destination across an inter-network. Routing protocols employ several important metrics to determine the best path for a packet to travel through. The process of path selection and directing packets from a source node to a destination node in a network is called routing and is an active area of research in ad hoc networks [1]. A metric is a standard for measuring parameters, such as distance, bandwidth, delay, and load for a path, and is used by routing algorithms to determine the optimal path to the destination [2]. An ad hoc network is a collection of nodes that are dynamically and arbitrarily located in such a manner that the interconnections between the nodes can change on a continual basis [3]. Currently, a major problem exists for routing protocols in ad hoc networks with wireless hosts. Numerous routing protocols are presently proposed for ad hoc networks [4]. In ad hoc networks, each node forwards its data to other nodes willingly. Each node communicates with other nodes within its transmission range [5,6]. To send a packet to a destination, the node forwards the packet to its neighbouring node,
q
Reviews processed and approved for publication by Editor-in-Chief Dr. Manu Malek.
⇑ Corresponding author. Tel.: +91 9866250010.
E-mail addresses: [email protected] (B. Nancharaiah), [email protected] (B. Chandra Mohan). http://dx.doi.org/10.1016/j.compeleceng.2014.01.007 0045-7906/Ó 2014 Elsevier Ltd. All rights reserved.
1256
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
which, in turn, forwards the received packet to its neighbour until the packet reaches the destination [6,7]. Therefore, a wireless ad hoc network does not have a clear line of defence, and every node must be prepared for encounters with an adversary directly or indirectly [8]. Most destination/next hop relations inform the router that a particular destination can be reached optimally by sending the packet to a specific node on behalf of the ‘‘next hop’’ end route to the final destination [9]. When the router receives a packet, it checks the address of the destination node on its routing table. Based on the information in the routing table, the router forwards the packet to the next hop or destination. These networks include a combination of fixed wireless services and mobile networking [2]. In community networks lacking in hierarchically organised networks, several challenges arise [10]. The sender uses this route to transmit the packet if a route is identified. Additionally, the sender may attempt to find a route using the route discovery protocol if no route is found. This paper is organised as follows: Section 2 presents a brief overview of other MANET routing strategies proposed in the literature. The basic idea of particle swarm optimisation is introduced in Section 3. Ant colony optimisation is discussed in Section 4. Section 5 explains the proposed hybrid routing intelligent algorithm model. Section 6 offers the performance evaluation and simulation results. Finally, conclusions are summarised in Section 7.
2. Related works Gunes et al. [11] presented an ant colony-based routing algorithm in the routing schemes for MANETs. This algorithm is constructed similarly to the many existing routing approaches and consists of three phases, viz., route discovery, route maintenance, and route failure handling. Ali et al. [12] have proposed a method for the routing protocols in the ad hoc and sensor wireless networks, including genetic programming (GP), neural network, evolutionary programming (EP), particle swarm optimisation (PSO), and ant colony optimisation (ACO). Restraints existed in those protocols and were the result of the mobility and non-infrastructure nature of the ad hoc and sensor networks. This paper presents probabilistic performance evaluation frameworks and swarm intelligence approaches (PSO, ACO) for routing in MANETS. The performance assessment metrics employed for ad hoc network routing algorithms include route optimality, power consumption, and routing overhead. The method proposed provides a critical analysis of the PSO and ACO-based algorithms, and additional approaches are applied for the optimisation of the ad hoc and wireless sensor network routing protocols. Cheng et al. [13] propose the use of a series of dynamic genetic algorithms (GAs) to solve the dynamic load-balanced clustering problem in MANETs. By using dynamic handling procedures, the population changes the topology and produces closely related solutions in good standard. The exploratory results show that these GAs can work well for the dynamic loadbalanced clustering problem and can overcome traditional GAs that do not consider lively network optimisation needs. Mahmood et al. [14] have evaluated the method of MANETs. They introduce a new, adaptive, and dynamic routing algorithm for MANETs based on ant colony optimisation (ACO) algorithms using the network delay analysis. The ACO algorithm assists in finding, if not the shortest, a very good path to connect the colony’s nest with a source of food. This experimental evaluation of MANETs is based on the estimation of the mean end-to-end delay for sending a packet from the source to the destination node through a MANET. The most important performance evaluation metric in computer networks is the mean end-to-end delay. The results prove that the proposed algorithm offers favourable results under certain conditions, such as long pause time and low node density. Limin and Wenbo [15] propose a Grover’s searching algorithm. With this new algorithm, optimal routing is selected by the node probability function. Pi and Sun [16] have proposed fuzzy controllers based multipath routing in MANETs (FMRM). The proposed approach is better and more effective in MANET applications. Ejaz et al. [17] have proposed a routing protocol that provides efficient and reliable communication with minimum overhead. This routing protocol is applicable in the real traffic of ships. Amri et al. [18] have explained different classes of routing protocol characteristics. Most of the current routing protocols assume homogeneous networking conditions in which all nodes have the same capabilities and resources. Guo et al. [19] have proposed a new routing protocol, i.e., hybrid on-demand distance vector multi-path (HODVM), that performs static routing and dynamic routing by dividing the spatial wireless ad hoc networks into the backbone and nonbackbone networks. HODVM can adaptively establish and maintain multiple node-disjoint routes using multi-path routing. Wang et al. [20] proposed a novel cooperative opportunistic routing scheme for mobile ad hoc networks as an extension of the Ex-OR to accommodate node mobility. This scheme uses the proactive routing protocol as a back-end along with increased provisions for effective data transportation. A simulation of the present protocol is performed over the NS-2 and compared with the live routing protocol AODV. Nancharaiah and Mohan [21] have proposed work that designates the routing problem based on the ACO and fuzzy logic performances while developing the routing algorithm. The path details by the ants will be provided to the fuzzy interference system(FIS) to calculate the score values of the convenient path, and based on this score value from the FIS system, the optimal paths will be chosen. Hence, the routing problem can be answered more effectively by achieving a highly successful path transportation rate as opposed to using the ACO routing algorithms.
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
1257
3. Particle swarm optimisation Particle swarm optimisation (PSO) is the mathematical modelling of the food searching activities of a swarm of birds (particles). Each particle in the swarm is moved towards the optimal point by adding a velocity and its position. The velocity of a particle is influenced by three components: inertial, cognitive, and social. The inertial component simulates the inertial behaviour of the bird to fly in the previous direction. The cognitive component models the memory of the bird for its previous best position, and the social component models the memory of the bird for the best position among the particles. Particle swarm has two primary operators: a velocity update and a position update. During each generation, each particle is accelerated towards the particle’s previous best position and the global best position [22]. Eqs. (1) and (2) provide the formula for the velocity and position updates.
v id ¼ w v id þ c1 r1 ðpid xid Þ þ c2 r2 ðpgd xid Þ;
ð1Þ
xidþ1 ¼ xidþ1 þ v id ;
ð2Þ
vid is the velocity of the particle. xid is the current position of the particle. w is the inertia weight. C1 is the cognitive acceleration coefficient. C2 is the social acceleration coefficient. Pid is the own best position of the particle. Pgd is the global best position among the group of particles. r1, r2 are the uniformly distributed random numbers in the range [0 to 1]. xid+1 is the modified position. The flow chart for the PSO model formulation scheme is shown in Fig. 1, where E is the previous best position, E0 is the current position, and F is the optimum solution (minimal particles in this case) from among all of the best solutions of the particle in Fig. 1. The steps involved in particle swarm optimisation are as follows: Step 1: Select the number of particles randomly to start the optimal solution search. Step 2: Initialise the particle position and velocity. Step 3: Select the particle’s individual best value for each generation. Step 4: Select the particle’s global best value, i.e., the particle nearest the target from among all of the particles is obtained by comparing all of the individual best values. Step 5: Select the particle’s individual worst value, i.e., the particle farthest away from the target. Step 6: Update the velocity and position of the particle per Eqs. (1) and (2). Step 7: Find the optimal solution with a minimum value for the updated new velocity and position.
Initialise the population Compute the objective function Yes If E
Pid ) so far
No For each generation
Search is terminated when optimal solution reached
For each particle Current value = new
Pid
Choose the minimum ‘F’ of all particles as the Pgd
Calculate particle velocity
Calculate particle position Update memory of each particle
Fig. 1. Flow chart representation for particle swarm optimisation.
1258
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
Initialisation of population
Evaluation of fitness
Pheromone updation
Discover new paths
If termination reached ?
Condition terminates Fig. 2. Flow chart representation for ant colony optimisation.
4. ANT colony optimisation Ant colony optimisation (ACO) is a paradigm for designing metaheuristic algorithms for combinatorial optimisation problems. The first algorithm that was classified within this framework was presented in 1991. Since then, many diverse variants of the basic principle have been reported in the literature. The essential trait of ACO algorithms is the combination of a priori information regarding the structure of a promising solution with a posteriori information regarding the structure of previously obtained good solutions. An ACO is a famous swarm intelligence approach that has received inspiration from the social behaviour of real world ants. In this algorithm, the best path for routing is identified by the pheromone deposited by ants. Upon finding the food, the ants return back to their nests and simultaneously deposit the pheromone along the paths. Therefore, the ants are likely to move through these paths and strengthen (update) the existing pheromone. Over time, the pheromone starts to evaporate, and its strength is reduced. At regular intervals, several ants are launched toward the destination node to discover the feasible, low cost path from the source node to the destination node. Each ant in an ACO considers two parameters to select its next hop. The first parameter is the amount of pheromone deposited on the path to the next node, and the second parameter is the queue length associated with the link. The flow chart for the ACO model formulation scheme is shown in Fig. 2. 5. Proposed hybrid routing intelligent algorithm model: PSO hybrid with ACO Several heuristic traditional algorithms were used to find a solution to the routing problem in the MANETs, including GA and PSO algorithms. The ACO technique is independent of these routing problems, and the outcomes obtained using the ACO technique can be improved with PSO [23]. Thus, a hybrid model that combines the ACO and PSO techniques can be suggested for the optimisation technique. The flow chart for the proposed hybrid routing intelligent algorithm model is shown in Fig. 3, where E is the previous best position, E0 is the current position and, ‘‘F’’ is the optimum solution (minimal particles in this case) from among all of the best solutions of the particle in Fig. 3. The steps involved in the proposed hybrid model are as follows:
Step Step Step Step
1: 2: 3: 4:
T xy
Initialise the number of particles and generate its value randomly. Initialise ACO parameters. Generate solutions from each ant’s random walk. Update the pheromone intensities using Eq. (3), where q ? Pheromone evaporation coefficient.
ð1 qÞT xy þ RK rKTxy :
ð3Þ
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
1259
Initialise the population Initialise the ACO parameters
Compute the objective function Invoke the hybrid algorithm Perform pheromone update
Yes best solution achieved? No
If E < E’ (
Yes
Pid )
so far No For each generation and each particle
Current value = new
Pid
Choose the minimum ‘F’ of all particles as the
Pgd
Calculate particle velocity and position
Update memory of each particle
End
No If stopping condition reached Yes Condition terminates Fig. 3. Flow chart representation of a PSO hybrid with an ACO.
rKTxy ! Amount of pheromone deposited. K ! Ant that deposits the pheromone. x is the index for the subsystem, and y refers to the components in a subsystem. Step 5: If the solution is not the best, initialise the swarm with random positions and velocities. Step 6: Select each particle’s individual best value for each generation. Step 7: Select the particle’s global best value, i.e., the particle nearest the target is obtained by comparing all of the individual best values.
1260
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
Step 8: Select the particle’s individual worst value, i.e., the particle farthest away from the target. Step 9: Update the velocity and position of the particle per Eqs. (1) and (2). Step 10: Terminate the process if the maximum number of iterations is reached or if an optimal value is obtained. Otherwise, proceed to Step 3. 6. Results and discussion This section discusses the performance evaluation and a comparison of the ACO algorithm and the proposed hybrid algorithm in MATLAB. The proposed hybrid algorithm routing was evaluated using a network of 100 nodes spread over a1000 m 1000 m region. Every node has a maximum transmit range of 250 m. The routing parameters, such as distance, delay, capacity, power consumption and cost, are used for the fitness evaluation using algorithms. The performance of the proposed PSO hybrid with ACO (PSO_ACO) is compared with the ACO algorithm using the routing parameters. The evaluation method for the parameters is given in Eqs. (4)–(7). 1=2
Distance between two nodesðABÞ ¼ ½ðA2 A1 Þ2 þ ðB2 B1 Þ2
ð4Þ
;
A2 and A1 are the latitudes, and B2 and B1 are the longitudes in real-time measurements. Delay can be determined using the formula in Eq. (5).
P Delay ¼
ðpacket arriv al time—packet forwarded timeÞ P ; number of nodes connection
ð5Þ
Capacity can be expressed as the sum of the capacity of the individual nodes, where as power consumption is assumed to be a random value for the nodes. Communication cost is the cost or money spent for the usage time as given in Eq. (6).
tcomm ¼ ts þ tn þ t w ;
ð6Þ
ts ? (Start-up time) Time exhausted for sending and receiving nodes. tn ? (Per-hop time) this time is a function of the number of hops. tw ? (Per-word transfer time) this time includes all of the overheads determined by the length of the message. The cost formula in Eq. (6) is based on time. If the process is performed in a minimum amount of time, the cost is minimal. The communication cost increases as time increases. The power consumption is calculated using Eq. (7).
Power consumption ¼ jReceiv ing power—Transmitting powerj:
ð7Þ
The path delay is measured based on the number of iterations for the individual ACO and hybrid PSO_ACO as shown in Fig. 4. The delay value reaches its minimum for PSO_ACO before the 20th iteration, but the ACO does not reach the optimal delay value even after 100iterations as shown in Fig. 4. The end-to-end delay is measured against the number of nodes using an individual ACO and a hybrid PSO_ACO as shown in Fig. 5, where the total delay varies based on the number of nodes. Therefore, the total delay not only depends upon the number of nodes but also depends on the distance (hops) between the source node and destination node in the simulation environment. The total delay increases and decreases suddenly in Fig. 5 and depends on the configuration of the network. Thus, the delay varies despite the hybrid routing intelligent
25
Best Path Delay (Sec.)
ACO PSO ACO
20
15
10
5
0
0
10
20
30
40
50
60
70
80
90
100
iteration Fig. 4. Path delay comparisons of ACO and PSO_ACO for a number of iterations (100 nodes).
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
1261
Fig. 5. Total delay of link versus the number of nodes using algorithms.
Best Path power value (mW)
60
ACO PSO ACO
50 40 30 20 10 0 0
10
20
30
40
50
60
70
80
90
100
iteration Fig. 6. Power consumption (mW) of ACO and PSO_ACO for a number of iterations (100 nodes).
Fig. 7. Power consumption of ACO, PSO_ACO and AGA_FACO for a different number of nodes.
algorithm’s improved performance compared with ACO. In fact, the delay level of the hybrid routing intelligent algorithm is less when compared with the ACO. Hence, the proposed hybrid routing intelligent algorithm is highly suitable for enormous networks consisting of a large number of nodes. The power consumption between the ACO and PSO_ACO is shown in Fig. 6 for 100 nodes. Therefore, we conclude that the hybrid routing intelligent algorithm requires 23 iterations to achieve the best power value. However, the ACO required 38 iterations to achieve the best value as shown in Fig. 6.
1262
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
Best Path cost value
300
ACO PSO ACO
250 200 150 100 50 0 0
10
20
30
40
50
60
70
80
90
100
iteration Fig. 8. Communication cost (money for the usage time) using ACO and PSO_ACO for a number of iterations (100 nodes).
Fig. 9. Communication cost using ACO, PSO_ACO and AGA_FACO for a different number of nodes.
Best Distance value (m)
800
ACO PSO ACO
700 600 500 400 300 200 100 0 0
10
20
30
40
50
60
70
80
90
100
iteration Fig. 10. Distance comparisons of ACO and PSO_ACO for a number of iterations (100 nodes).
In Fig. 7, a comparison of the power consumption versus the number of the nodes for the ACO is shown. The analysis results for PSO_ACO are improved for nodes up to 350, and the adaptive genetic algorithm combined (AGA) with fuzzy ant colony optimisation (FACO) increases while the number of nodes increases. The power consumption is less for the hybrid routing intelligent algorithm when compared with the ACO.
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
1263
The path cost parameter versus the number of iterations is illustrated in Fig. 8 for the ACO and the proposed hybrid routing intelligent algorithm. In this comparison graph, the PSO_ACO algorithm achieves its best cost value in five iterations, whereas the ACO algorithm requires 38 iterations to achieve its best cost value. The communication cost parameter comparison is shown in Fig. 9 for the ACO, the proposed hybrid routing intelligent algorithm, and the hybrid AGA_FACO. In this comparison graph, the PSO_ACO algorithm has a lower cost when compared with the individual ACO algorithm and the hybrid adaptive genetic algorithm combined with fuzzy ant colony optimisation (AGA_FACO).In the ACO, the optimal path finding process is very slow and leads to a higher cost as shown in Fig. 9. In Fig. 10, distance is used as a parameter to find the shortest distance between the nodes. The PSO_ACO distance is less than the ACO distance. The shortest distance is referred to as distance. The shortest distance between the source and destination for each iteration is shown in Fig. 10. When the iterations increase, the distance decreases, and the optimum distance results are achieved as shown in Fig. 10. Therefore, the shortest distance is achieved using the PSO_ACO algorithm. 7. Conclusion The proposed hybrid routing intelligent algorithm model that combines the PSO approach and the ACO approach for the MANETs was simulated, and the results were presented. Our simulation results predicted that the proposed hybrid routing intelligent algorithm (PSO_ACO) has the ability to cope with enormous networks that contain a large number of nodes. From the performance analysis, we conclude that the path outcome using the hybrid routing intelligent algorithm (PSO_ACO) has the shortest distance, a minimum delay, low power consumption, and low cost when compared with the individual performance of the ACO algorithm. An additional advantage of the proposed hybrid routing intelligent algorithm approach is its tendency to find the optimal route in the MANETs and its ability to perform proficiently when compared with the ACO algorithm. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
Taneja Sunil, Kush Ashwini. A survey of routing protocols in mobile ad hoc networks. Int J Innov Manage Technol 2010;1(3). Munaretto Anelise, Fonseca Mauro. Routing and quality of service support for mobile ad hoc networks. Comput Networks 2007;51(11):3142–56. Gupta Sachin Kumar, Saket. Routing protocols in mobile ad-hoc networks. Special Issue Int J Comput Appl 2011;4(December):24–7. Tan Kun, Zhang Qian, Zhu Wenwu. Shortest path routing in partially connected ad hoc networks. In: IEEE global telecommunications conference, vol. 2; 2003. p. 1038–42 Breed Gary. Wireless ad hoc networks: basic concepts. High Freq Electron 2007:44–6. Baskar Raju, Ahmed Gulfishan. Different approaches on cooperation in wireless ad hoc networks. Int J Comput Appl 2011;28(3):36–41. Johnson David B, Maltz David A, Broch Josh. DSR: the dynamic source routing protocol for multi hop wireless ad hoc networks. In: Ad hoc networking; December, 2000. p. 139–72. Luo Haiyun, Zerfos, Kong Jiejun, Lu Songwu, Zhang Lixia. Self-securing ad hoc wireless networks. In: The Seventh IEEE symposium on computers and communications; 2002. p. 1–17. Lee Seoung-Bum, Ahn Gahng-Seop, Zhang Xiaowei, Campbell Andrew T. INSIGNIA: An IP-based quality of service framework for mobile ad hoc networks. J Parallel Distr Comput 2000;60:374–406. Varshney Upkar, Vetter Ron. Emerging mobile and wireless networks. Mag Commun ACM 2000;43(6):73–81. Günes, Sorges, Bouazizi I. ARA-the ant-colony based routing algorithm for MANETs. In: International conference on parallel processing workshops (ICPPW-02). IEEE Computer Society Press; 2002. p. 79–85. Ali Zulfiqar, Shahzad. Critical analysis of swarm intelligence based routing protocols in ad-hoc and sensor wireless networks. In: International conference on computer networks and information technology; 2011. p. 287–292. Cheng Hui, Yang Shengxiang, Cao Jiannong. Dynamic genetic algorithms for the dynamic load balanced clustering problem in mobile ad hoc networks. Expert Syst Appl 2013;40(4):1381–92. Mahmood Toqeer, Nawaz Tabassam, Ashraf Rehan, Adnan Shah Syed M. Gossip based routing protocol design for ad hoc networks. IJCSI Int J Comput Sci Issues 2012;9(1):177–81. MENG Limin, SONG Wenbo. Routing protocol based on grover’s searching algorithm for mobile ad-hoc networks. China communications; March 2013. p. 145–56. Pi Shangchao, Sun Baolin. Fuzzy controllers based multipath routing algorithm in MANET routing. In: International conference on applied physics and industrial engineering. Part B, vol. 24; 2012. p. 1178–85. Ejaz Waleed, Manzoor Kamran, Kim Hyung Joo, Jang Byung Tae, Jin Gwang-Ja, Kim Hyung Seok. Two-state routing protocol for maritime multi-hop Wireless networks. Comput Electr Eng 2013;39(6):1854–66. Amri Huda Al, Abolhasan Mehran, Wysocki Tadeusz. Scalability of MANET routing protocols for heterogeneous and homogenous networks. Comput Electr Eng 2010;36(4):752–65. Guo Lei, Zhang Lincong, Peng Yuhuai, Jingjing Wu, Zhang Xiaoying, Hou Weigang, et al. Multi-path routing in spatial wireless ad hoc networks. Comput Electr Eng 2012;38(3):473–91. Wang Z, Chen Y, Li C. CORMAN: Anovel cooperative opportunistic routing scheme in mobile ad hoc networks. IEEE J Select Areas Commun 2012;30(2):289–96. Nancharaiah B, Chandra Mohan B. Routing in mobile ad hoc networks (MANETS) using fuzzy aided ant colony optimization (ACO) technique. J Theor Appl Inform Technol 2013;53(2):227–35. Thanushkodi, Deeba K. Hybrid intelligent algorithm [improved particle swarm optimization (PSO) with ant colony optimization (ACO)] for multiprocessor job scheduling. Sci Res Essays 2012;7(20):1935–53. Nancharaiah B, Chandra Mohan B. MANET link performance using ant colony optimization and particle swarm optimization algorithms. In: IEEE international conference on communications and signal processing (ICCSP); April, 2013. p. 767–70.
B. Nancharaiah received a Bachelor’s degree in Electronics and Communications Engineering from the SRKR Engineering College in Bhimavaram in 1999 and a Master’s degree in Electronics and Communications Engineering from the Pondicherry Engineering College at Pondicherry Central University in 2003. He is pursuing a PhD in JNTU, Hyderabad. He is currently working as a faculty member at the NRI Institute of Technology, Guntur. His research interests are wireless communications and networks.
1264
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
B. Chandra Mohan received a Bachelor’s degree in E.C.E. from the Bapatla Engineering College in Bapatla in 1990 and a Master’s degree in Microwave Engineering from the Cochin University of Science and Technology in 1992. He obtained his PhD from JNT University in Hyderabad in 2009. Presently, he works as a professor and the head of the ECE Dept. in BEC in Bapatla. His research interests include image watermarking, image compression, and communications.
Contents lists available at ScienceDirect
Computers and Electrical Engineering journal homepage: www.elsevier.com/locate/compeleceng
The performance of a hybrid routing intelligent algorithm in a mobile ad hoc network q B. Nancharaiah a,⇑, B. Chandra Mohan b a b
Department of ECE, JNT University, Hyderabad, Andhra Pradesh, India Department of ECE, Bapatla Engineering College, Bapatla 522101, Andhra Pradesh, India
a r t i c l e
i n f o
Article history: Available online 26 February 2014
a b s t r a c t End-to-end delay, power consumption, and communication cost are some of the most important metrics in a mobile ad hoc network (MANET) when routing from a source to a destination. Recent approaches using the swarm intelligence (SI) technique proved that the local interaction of several simple agents to meet a global goal has a significant impact on MANET routing. In this work, a hybrid routing intelligent algorithm that has an ant colony optimisation (ACO) algorithm and particle swarm optimisation (PSO) is used to improve the various metrics in MANET routing. The ACO algorithm uses mobile agents as ants to identify the most feasible and best path in a network. Additionally, the ACO algorithm helps to locate paths between two nodes in a network and provides input to the PSO technique, which is a metaheuristic approach in SI. The PSO finds the best solution for a particle’s position and velocity and minimises cost, power, and end-to-end delay. This hybrid routing intelligent algorithm has an improved performance when compared with the simple ACO algorithm in terms of delay, power consumption, and communication cost. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction A mobile ad hoc network (MANET) is a self-configuring infrastructureless network. Communication in MANETs is typically performed with the aid of temporary multi-hop relays, i.e., a source node uses its neighbouring node as a relay router. Routing is the act of moving information from a source to a destination across an inter-network. Routing protocols employ several important metrics to determine the best path for a packet to travel through. The process of path selection and directing packets from a source node to a destination node in a network is called routing and is an active area of research in ad hoc networks [1]. A metric is a standard for measuring parameters, such as distance, bandwidth, delay, and load for a path, and is used by routing algorithms to determine the optimal path to the destination [2]. An ad hoc network is a collection of nodes that are dynamically and arbitrarily located in such a manner that the interconnections between the nodes can change on a continual basis [3]. Currently, a major problem exists for routing protocols in ad hoc networks with wireless hosts. Numerous routing protocols are presently proposed for ad hoc networks [4]. In ad hoc networks, each node forwards its data to other nodes willingly. Each node communicates with other nodes within its transmission range [5,6]. To send a packet to a destination, the node forwards the packet to its neighbouring node,
q
Reviews processed and approved for publication by Editor-in-Chief Dr. Manu Malek.
⇑ Corresponding author. Tel.: +91 9866250010.
E-mail addresses: [email protected] (B. Nancharaiah), [email protected] (B. Chandra Mohan). http://dx.doi.org/10.1016/j.compeleceng.2014.01.007 0045-7906/Ó 2014 Elsevier Ltd. All rights reserved.
1256
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
which, in turn, forwards the received packet to its neighbour until the packet reaches the destination [6,7]. Therefore, a wireless ad hoc network does not have a clear line of defence, and every node must be prepared for encounters with an adversary directly or indirectly [8]. Most destination/next hop relations inform the router that a particular destination can be reached optimally by sending the packet to a specific node on behalf of the ‘‘next hop’’ end route to the final destination [9]. When the router receives a packet, it checks the address of the destination node on its routing table. Based on the information in the routing table, the router forwards the packet to the next hop or destination. These networks include a combination of fixed wireless services and mobile networking [2]. In community networks lacking in hierarchically organised networks, several challenges arise [10]. The sender uses this route to transmit the packet if a route is identified. Additionally, the sender may attempt to find a route using the route discovery protocol if no route is found. This paper is organised as follows: Section 2 presents a brief overview of other MANET routing strategies proposed in the literature. The basic idea of particle swarm optimisation is introduced in Section 3. Ant colony optimisation is discussed in Section 4. Section 5 explains the proposed hybrid routing intelligent algorithm model. Section 6 offers the performance evaluation and simulation results. Finally, conclusions are summarised in Section 7.
2. Related works Gunes et al. [11] presented an ant colony-based routing algorithm in the routing schemes for MANETs. This algorithm is constructed similarly to the many existing routing approaches and consists of three phases, viz., route discovery, route maintenance, and route failure handling. Ali et al. [12] have proposed a method for the routing protocols in the ad hoc and sensor wireless networks, including genetic programming (GP), neural network, evolutionary programming (EP), particle swarm optimisation (PSO), and ant colony optimisation (ACO). Restraints existed in those protocols and were the result of the mobility and non-infrastructure nature of the ad hoc and sensor networks. This paper presents probabilistic performance evaluation frameworks and swarm intelligence approaches (PSO, ACO) for routing in MANETS. The performance assessment metrics employed for ad hoc network routing algorithms include route optimality, power consumption, and routing overhead. The method proposed provides a critical analysis of the PSO and ACO-based algorithms, and additional approaches are applied for the optimisation of the ad hoc and wireless sensor network routing protocols. Cheng et al. [13] propose the use of a series of dynamic genetic algorithms (GAs) to solve the dynamic load-balanced clustering problem in MANETs. By using dynamic handling procedures, the population changes the topology and produces closely related solutions in good standard. The exploratory results show that these GAs can work well for the dynamic loadbalanced clustering problem and can overcome traditional GAs that do not consider lively network optimisation needs. Mahmood et al. [14] have evaluated the method of MANETs. They introduce a new, adaptive, and dynamic routing algorithm for MANETs based on ant colony optimisation (ACO) algorithms using the network delay analysis. The ACO algorithm assists in finding, if not the shortest, a very good path to connect the colony’s nest with a source of food. This experimental evaluation of MANETs is based on the estimation of the mean end-to-end delay for sending a packet from the source to the destination node through a MANET. The most important performance evaluation metric in computer networks is the mean end-to-end delay. The results prove that the proposed algorithm offers favourable results under certain conditions, such as long pause time and low node density. Limin and Wenbo [15] propose a Grover’s searching algorithm. With this new algorithm, optimal routing is selected by the node probability function. Pi and Sun [16] have proposed fuzzy controllers based multipath routing in MANETs (FMRM). The proposed approach is better and more effective in MANET applications. Ejaz et al. [17] have proposed a routing protocol that provides efficient and reliable communication with minimum overhead. This routing protocol is applicable in the real traffic of ships. Amri et al. [18] have explained different classes of routing protocol characteristics. Most of the current routing protocols assume homogeneous networking conditions in which all nodes have the same capabilities and resources. Guo et al. [19] have proposed a new routing protocol, i.e., hybrid on-demand distance vector multi-path (HODVM), that performs static routing and dynamic routing by dividing the spatial wireless ad hoc networks into the backbone and nonbackbone networks. HODVM can adaptively establish and maintain multiple node-disjoint routes using multi-path routing. Wang et al. [20] proposed a novel cooperative opportunistic routing scheme for mobile ad hoc networks as an extension of the Ex-OR to accommodate node mobility. This scheme uses the proactive routing protocol as a back-end along with increased provisions for effective data transportation. A simulation of the present protocol is performed over the NS-2 and compared with the live routing protocol AODV. Nancharaiah and Mohan [21] have proposed work that designates the routing problem based on the ACO and fuzzy logic performances while developing the routing algorithm. The path details by the ants will be provided to the fuzzy interference system(FIS) to calculate the score values of the convenient path, and based on this score value from the FIS system, the optimal paths will be chosen. Hence, the routing problem can be answered more effectively by achieving a highly successful path transportation rate as opposed to using the ACO routing algorithms.
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
1257
3. Particle swarm optimisation Particle swarm optimisation (PSO) is the mathematical modelling of the food searching activities of a swarm of birds (particles). Each particle in the swarm is moved towards the optimal point by adding a velocity and its position. The velocity of a particle is influenced by three components: inertial, cognitive, and social. The inertial component simulates the inertial behaviour of the bird to fly in the previous direction. The cognitive component models the memory of the bird for its previous best position, and the social component models the memory of the bird for the best position among the particles. Particle swarm has two primary operators: a velocity update and a position update. During each generation, each particle is accelerated towards the particle’s previous best position and the global best position [22]. Eqs. (1) and (2) provide the formula for the velocity and position updates.
v id ¼ w v id þ c1 r1 ðpid xid Þ þ c2 r2 ðpgd xid Þ;
ð1Þ
xidþ1 ¼ xidþ1 þ v id ;
ð2Þ
vid is the velocity of the particle. xid is the current position of the particle. w is the inertia weight. C1 is the cognitive acceleration coefficient. C2 is the social acceleration coefficient. Pid is the own best position of the particle. Pgd is the global best position among the group of particles. r1, r2 are the uniformly distributed random numbers in the range [0 to 1]. xid+1 is the modified position. The flow chart for the PSO model formulation scheme is shown in Fig. 1, where E is the previous best position, E0 is the current position, and F is the optimum solution (minimal particles in this case) from among all of the best solutions of the particle in Fig. 1. The steps involved in particle swarm optimisation are as follows: Step 1: Select the number of particles randomly to start the optimal solution search. Step 2: Initialise the particle position and velocity. Step 3: Select the particle’s individual best value for each generation. Step 4: Select the particle’s global best value, i.e., the particle nearest the target from among all of the particles is obtained by comparing all of the individual best values. Step 5: Select the particle’s individual worst value, i.e., the particle farthest away from the target. Step 6: Update the velocity and position of the particle per Eqs. (1) and (2). Step 7: Find the optimal solution with a minimum value for the updated new velocity and position.
Initialise the population Compute the objective function Yes If E
Pid ) so far
No For each generation
Search is terminated when optimal solution reached
For each particle Current value = new
Pid
Choose the minimum ‘F’ of all particles as the Pgd
Calculate particle velocity
Calculate particle position Update memory of each particle
Fig. 1. Flow chart representation for particle swarm optimisation.
1258
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
Initialisation of population
Evaluation of fitness
Pheromone updation
Discover new paths
If termination reached ?
Condition terminates Fig. 2. Flow chart representation for ant colony optimisation.
4. ANT colony optimisation Ant colony optimisation (ACO) is a paradigm for designing metaheuristic algorithms for combinatorial optimisation problems. The first algorithm that was classified within this framework was presented in 1991. Since then, many diverse variants of the basic principle have been reported in the literature. The essential trait of ACO algorithms is the combination of a priori information regarding the structure of a promising solution with a posteriori information regarding the structure of previously obtained good solutions. An ACO is a famous swarm intelligence approach that has received inspiration from the social behaviour of real world ants. In this algorithm, the best path for routing is identified by the pheromone deposited by ants. Upon finding the food, the ants return back to their nests and simultaneously deposit the pheromone along the paths. Therefore, the ants are likely to move through these paths and strengthen (update) the existing pheromone. Over time, the pheromone starts to evaporate, and its strength is reduced. At regular intervals, several ants are launched toward the destination node to discover the feasible, low cost path from the source node to the destination node. Each ant in an ACO considers two parameters to select its next hop. The first parameter is the amount of pheromone deposited on the path to the next node, and the second parameter is the queue length associated with the link. The flow chart for the ACO model formulation scheme is shown in Fig. 2. 5. Proposed hybrid routing intelligent algorithm model: PSO hybrid with ACO Several heuristic traditional algorithms were used to find a solution to the routing problem in the MANETs, including GA and PSO algorithms. The ACO technique is independent of these routing problems, and the outcomes obtained using the ACO technique can be improved with PSO [23]. Thus, a hybrid model that combines the ACO and PSO techniques can be suggested for the optimisation technique. The flow chart for the proposed hybrid routing intelligent algorithm model is shown in Fig. 3, where E is the previous best position, E0 is the current position and, ‘‘F’’ is the optimum solution (minimal particles in this case) from among all of the best solutions of the particle in Fig. 3. The steps involved in the proposed hybrid model are as follows:
Step Step Step Step
1: 2: 3: 4:
T xy
Initialise the number of particles and generate its value randomly. Initialise ACO parameters. Generate solutions from each ant’s random walk. Update the pheromone intensities using Eq. (3), where q ? Pheromone evaporation coefficient.
ð1 qÞT xy þ RK rKTxy :
ð3Þ
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
1259
Initialise the population Initialise the ACO parameters
Compute the objective function Invoke the hybrid algorithm Perform pheromone update
Yes best solution achieved? No
If E < E’ (
Yes
Pid )
so far No For each generation and each particle
Current value = new
Pid
Choose the minimum ‘F’ of all particles as the
Pgd
Calculate particle velocity and position
Update memory of each particle
End
No If stopping condition reached Yes Condition terminates Fig. 3. Flow chart representation of a PSO hybrid with an ACO.
rKTxy ! Amount of pheromone deposited. K ! Ant that deposits the pheromone. x is the index for the subsystem, and y refers to the components in a subsystem. Step 5: If the solution is not the best, initialise the swarm with random positions and velocities. Step 6: Select each particle’s individual best value for each generation. Step 7: Select the particle’s global best value, i.e., the particle nearest the target is obtained by comparing all of the individual best values.
1260
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
Step 8: Select the particle’s individual worst value, i.e., the particle farthest away from the target. Step 9: Update the velocity and position of the particle per Eqs. (1) and (2). Step 10: Terminate the process if the maximum number of iterations is reached or if an optimal value is obtained. Otherwise, proceed to Step 3. 6. Results and discussion This section discusses the performance evaluation and a comparison of the ACO algorithm and the proposed hybrid algorithm in MATLAB. The proposed hybrid algorithm routing was evaluated using a network of 100 nodes spread over a1000 m 1000 m region. Every node has a maximum transmit range of 250 m. The routing parameters, such as distance, delay, capacity, power consumption and cost, are used for the fitness evaluation using algorithms. The performance of the proposed PSO hybrid with ACO (PSO_ACO) is compared with the ACO algorithm using the routing parameters. The evaluation method for the parameters is given in Eqs. (4)–(7). 1=2
Distance between two nodesðABÞ ¼ ½ðA2 A1 Þ2 þ ðB2 B1 Þ2
ð4Þ
;
A2 and A1 are the latitudes, and B2 and B1 are the longitudes in real-time measurements. Delay can be determined using the formula in Eq. (5).
P Delay ¼
ðpacket arriv al time—packet forwarded timeÞ P ; number of nodes connection
ð5Þ
Capacity can be expressed as the sum of the capacity of the individual nodes, where as power consumption is assumed to be a random value for the nodes. Communication cost is the cost or money spent for the usage time as given in Eq. (6).
tcomm ¼ ts þ tn þ t w ;
ð6Þ
ts ? (Start-up time) Time exhausted for sending and receiving nodes. tn ? (Per-hop time) this time is a function of the number of hops. tw ? (Per-word transfer time) this time includes all of the overheads determined by the length of the message. The cost formula in Eq. (6) is based on time. If the process is performed in a minimum amount of time, the cost is minimal. The communication cost increases as time increases. The power consumption is calculated using Eq. (7).
Power consumption ¼ jReceiv ing power—Transmitting powerj:
ð7Þ
The path delay is measured based on the number of iterations for the individual ACO and hybrid PSO_ACO as shown in Fig. 4. The delay value reaches its minimum for PSO_ACO before the 20th iteration, but the ACO does not reach the optimal delay value even after 100iterations as shown in Fig. 4. The end-to-end delay is measured against the number of nodes using an individual ACO and a hybrid PSO_ACO as shown in Fig. 5, where the total delay varies based on the number of nodes. Therefore, the total delay not only depends upon the number of nodes but also depends on the distance (hops) between the source node and destination node in the simulation environment. The total delay increases and decreases suddenly in Fig. 5 and depends on the configuration of the network. Thus, the delay varies despite the hybrid routing intelligent
25
Best Path Delay (Sec.)
ACO PSO ACO
20
15
10
5
0
0
10
20
30
40
50
60
70
80
90
100
iteration Fig. 4. Path delay comparisons of ACO and PSO_ACO for a number of iterations (100 nodes).
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
1261
Fig. 5. Total delay of link versus the number of nodes using algorithms.
Best Path power value (mW)
60
ACO PSO ACO
50 40 30 20 10 0 0
10
20
30
40
50
60
70
80
90
100
iteration Fig. 6. Power consumption (mW) of ACO and PSO_ACO for a number of iterations (100 nodes).
Fig. 7. Power consumption of ACO, PSO_ACO and AGA_FACO for a different number of nodes.
algorithm’s improved performance compared with ACO. In fact, the delay level of the hybrid routing intelligent algorithm is less when compared with the ACO. Hence, the proposed hybrid routing intelligent algorithm is highly suitable for enormous networks consisting of a large number of nodes. The power consumption between the ACO and PSO_ACO is shown in Fig. 6 for 100 nodes. Therefore, we conclude that the hybrid routing intelligent algorithm requires 23 iterations to achieve the best power value. However, the ACO required 38 iterations to achieve the best value as shown in Fig. 6.
1262
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
Best Path cost value
300
ACO PSO ACO
250 200 150 100 50 0 0
10
20
30
40
50
60
70
80
90
100
iteration Fig. 8. Communication cost (money for the usage time) using ACO and PSO_ACO for a number of iterations (100 nodes).
Fig. 9. Communication cost using ACO, PSO_ACO and AGA_FACO for a different number of nodes.
Best Distance value (m)
800
ACO PSO ACO
700 600 500 400 300 200 100 0 0
10
20
30
40
50
60
70
80
90
100
iteration Fig. 10. Distance comparisons of ACO and PSO_ACO for a number of iterations (100 nodes).
In Fig. 7, a comparison of the power consumption versus the number of the nodes for the ACO is shown. The analysis results for PSO_ACO are improved for nodes up to 350, and the adaptive genetic algorithm combined (AGA) with fuzzy ant colony optimisation (FACO) increases while the number of nodes increases. The power consumption is less for the hybrid routing intelligent algorithm when compared with the ACO.
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
1263
The path cost parameter versus the number of iterations is illustrated in Fig. 8 for the ACO and the proposed hybrid routing intelligent algorithm. In this comparison graph, the PSO_ACO algorithm achieves its best cost value in five iterations, whereas the ACO algorithm requires 38 iterations to achieve its best cost value. The communication cost parameter comparison is shown in Fig. 9 for the ACO, the proposed hybrid routing intelligent algorithm, and the hybrid AGA_FACO. In this comparison graph, the PSO_ACO algorithm has a lower cost when compared with the individual ACO algorithm and the hybrid adaptive genetic algorithm combined with fuzzy ant colony optimisation (AGA_FACO).In the ACO, the optimal path finding process is very slow and leads to a higher cost as shown in Fig. 9. In Fig. 10, distance is used as a parameter to find the shortest distance between the nodes. The PSO_ACO distance is less than the ACO distance. The shortest distance is referred to as distance. The shortest distance between the source and destination for each iteration is shown in Fig. 10. When the iterations increase, the distance decreases, and the optimum distance results are achieved as shown in Fig. 10. Therefore, the shortest distance is achieved using the PSO_ACO algorithm. 7. Conclusion The proposed hybrid routing intelligent algorithm model that combines the PSO approach and the ACO approach for the MANETs was simulated, and the results were presented. Our simulation results predicted that the proposed hybrid routing intelligent algorithm (PSO_ACO) has the ability to cope with enormous networks that contain a large number of nodes. From the performance analysis, we conclude that the path outcome using the hybrid routing intelligent algorithm (PSO_ACO) has the shortest distance, a minimum delay, low power consumption, and low cost when compared with the individual performance of the ACO algorithm. An additional advantage of the proposed hybrid routing intelligent algorithm approach is its tendency to find the optimal route in the MANETs and its ability to perform proficiently when compared with the ACO algorithm. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
Taneja Sunil, Kush Ashwini. A survey of routing protocols in mobile ad hoc networks. Int J Innov Manage Technol 2010;1(3). Munaretto Anelise, Fonseca Mauro. Routing and quality of service support for mobile ad hoc networks. Comput Networks 2007;51(11):3142–56. Gupta Sachin Kumar, Saket. Routing protocols in mobile ad-hoc networks. Special Issue Int J Comput Appl 2011;4(December):24–7. Tan Kun, Zhang Qian, Zhu Wenwu. Shortest path routing in partially connected ad hoc networks. In: IEEE global telecommunications conference, vol. 2; 2003. p. 1038–42 Breed Gary. Wireless ad hoc networks: basic concepts. High Freq Electron 2007:44–6. Baskar Raju, Ahmed Gulfishan. Different approaches on cooperation in wireless ad hoc networks. Int J Comput Appl 2011;28(3):36–41. Johnson David B, Maltz David A, Broch Josh. DSR: the dynamic source routing protocol for multi hop wireless ad hoc networks. In: Ad hoc networking; December, 2000. p. 139–72. Luo Haiyun, Zerfos, Kong Jiejun, Lu Songwu, Zhang Lixia. Self-securing ad hoc wireless networks. In: The Seventh IEEE symposium on computers and communications; 2002. p. 1–17. Lee Seoung-Bum, Ahn Gahng-Seop, Zhang Xiaowei, Campbell Andrew T. INSIGNIA: An IP-based quality of service framework for mobile ad hoc networks. J Parallel Distr Comput 2000;60:374–406. Varshney Upkar, Vetter Ron. Emerging mobile and wireless networks. Mag Commun ACM 2000;43(6):73–81. Günes, Sorges, Bouazizi I. ARA-the ant-colony based routing algorithm for MANETs. In: International conference on parallel processing workshops (ICPPW-02). IEEE Computer Society Press; 2002. p. 79–85. Ali Zulfiqar, Shahzad. Critical analysis of swarm intelligence based routing protocols in ad-hoc and sensor wireless networks. In: International conference on computer networks and information technology; 2011. p. 287–292. Cheng Hui, Yang Shengxiang, Cao Jiannong. Dynamic genetic algorithms for the dynamic load balanced clustering problem in mobile ad hoc networks. Expert Syst Appl 2013;40(4):1381–92. Mahmood Toqeer, Nawaz Tabassam, Ashraf Rehan, Adnan Shah Syed M. Gossip based routing protocol design for ad hoc networks. IJCSI Int J Comput Sci Issues 2012;9(1):177–81. MENG Limin, SONG Wenbo. Routing protocol based on grover’s searching algorithm for mobile ad-hoc networks. China communications; March 2013. p. 145–56. Pi Shangchao, Sun Baolin. Fuzzy controllers based multipath routing algorithm in MANET routing. In: International conference on applied physics and industrial engineering. Part B, vol. 24; 2012. p. 1178–85. Ejaz Waleed, Manzoor Kamran, Kim Hyung Joo, Jang Byung Tae, Jin Gwang-Ja, Kim Hyung Seok. Two-state routing protocol for maritime multi-hop Wireless networks. Comput Electr Eng 2013;39(6):1854–66. Amri Huda Al, Abolhasan Mehran, Wysocki Tadeusz. Scalability of MANET routing protocols for heterogeneous and homogenous networks. Comput Electr Eng 2010;36(4):752–65. Guo Lei, Zhang Lincong, Peng Yuhuai, Jingjing Wu, Zhang Xiaoying, Hou Weigang, et al. Multi-path routing in spatial wireless ad hoc networks. Comput Electr Eng 2012;38(3):473–91. Wang Z, Chen Y, Li C. CORMAN: Anovel cooperative opportunistic routing scheme in mobile ad hoc networks. IEEE J Select Areas Commun 2012;30(2):289–96. Nancharaiah B, Chandra Mohan B. Routing in mobile ad hoc networks (MANETS) using fuzzy aided ant colony optimization (ACO) technique. J Theor Appl Inform Technol 2013;53(2):227–35. Thanushkodi, Deeba K. Hybrid intelligent algorithm [improved particle swarm optimization (PSO) with ant colony optimization (ACO)] for multiprocessor job scheduling. Sci Res Essays 2012;7(20):1935–53. Nancharaiah B, Chandra Mohan B. MANET link performance using ant colony optimization and particle swarm optimization algorithms. In: IEEE international conference on communications and signal processing (ICCSP); April, 2013. p. 767–70.
B. Nancharaiah received a Bachelor’s degree in Electronics and Communications Engineering from the SRKR Engineering College in Bhimavaram in 1999 and a Master’s degree in Electronics and Communications Engineering from the Pondicherry Engineering College at Pondicherry Central University in 2003. He is pursuing a PhD in JNTU, Hyderabad. He is currently working as a faculty member at the NRI Institute of Technology, Guntur. His research interests are wireless communications and networks.
1264
B. Nancharaiah, B. Chandra Mohan / Computers and Electrical Engineering 40 (2014) 1255–1264
B. Chandra Mohan received a Bachelor’s degree in E.C.E. from the Bapatla Engineering College in Bapatla in 1990 and a Master’s degree in Microwave Engineering from the Cochin University of Science and Technology in 1992. He obtained his PhD from JNT University in Hyderabad in 2009. Presently, he works as a professor and the head of the ECE Dept. in BEC in Bapatla. His research interests include image watermarking, image compression, and communications.