Download Comparative Analysis Of shortest Path Optimization

326
International Journal of Advances in Electrical and Electronics Engineering
Available online at www.ijaeee.com & www.sestindia.org
ISSN: 2319-1112
Comparative Analysis Of shortest Path
Optimization Techniques in Packet Switching
Network based on Neural Network
Kamakshi 1, Mr Rakesh Kumar Sharma 2, Pradyut Kumar Mohanty
1
Mtech scholar, IET Alwar
[email protected]
2
Asst.Prof. IET, Alwar
[email protected]
3
Lecturer. Webuniv, New Delhi
[email protected]
3
ABSTRACT:- The more effective neural network algorithm for optimization of routing in communication networks is
proposed. As it was well known from literature, various optimization and very ill-defined problems may be solved using
appropriately designed neural networks, because of their high computational speed and the possibility of working with
uncertain data. Under some considerations, the routing in packet-switched communication networks may be considered as
optimization issues, more precisely, as a shortest-path problem. Routing algorithm neural network is designed to find the
optimal path, meaning, the shortest path (if possible), but considering the traffic conditions: the incoming traffic flow,
routers occupancy, and link or connection capacities, preventing the packet loss due to the input buffer overflow. The
performance of the proposed neural network is applicable in different traffic conditions and for different full-connected
networks with both symmetrical and non-symmetrical links.
Keywords: Routing, shortest path, optimization, Rosenblatt algorithm, packet Switching network.
I. INTRODUCTION
In modern real communication networks, specifically in packet switched networks, routing is an important and
prominent process that has a significant impact on the network’s performance. Ideal routing algorithm comprises
finding the “optimal” path(s) between source and destination router, enabling very high-speed data transmission and
avoiding a packet loss. Because modern communication traffic is characterized by huge amount of source-destination
pairs, high variability, nonlinearity and unpredictability, the routing policy is a very difficult task. Under some
assumptions the optimal routing can be considered as the shortest path (SP) computations that have to be carried out
in real time. This creates neural networks very good can-didates for solving the issue, because of their high
computational speed and the possibility of working with indefinite data.
II SHORTEST PATH OPTIMIZATION
2.1Basic Description: The shortest path problem can be defined for graphs whether undirected, directed, or mixed
graph.
The problem is also sometimes called the single-pair shortest path issues, to distinguish it from the following
variations:
•
The single-source shortest path issue, in which we have to find shortest paths from a source vertex v to all
other vertices in the graph.
•
The single-destination shortest path issue, in which we have to find shortest paths from all vertices in the
directed graph to a single goal vertex v. This can be reduced to the single-source shortest path problem by
reversing the arcs in the directed graph.
ISSN: 2319-1112 /V2N2:326-330 ©IJAEEE
327
Co mpa ra tive Ana ly sis O f shortest Pat h O pt imiza tio n Techniques in Pa cket Sw it ching
Netwo rk ba sed o n Neura l Network
•
The all-pairs shortest path issue, in which we have to find shortest paths between every pair of
vertices v, v' in the graph.
2.2 Application of Shortest Path: Shortest path algorithms are applied to automatically
Find directions between physical status, such as driving directions on web mapping websites like Map
quest or Google Maps. For this approach fast specialized algorithms are available.[2]If one represents a
nondeterministic abstract machine as a graph where vertices describe states and edges describe possible transitions,
shortest path algorithms can be applied to find an optimal sequence of choices to reach a certain goal state, or to
accomplish lower bounds on the time needed to reach a given state. For example, if vertices explore the states of a
puzzle each directed edge corresponds to a single move, shortest path algorithms can be applied to find a solution that
uses the minimum possible number of moves.
III. ALGORITHMS USED FOR SHORTEST PATH OPTIMIZATION
The most important algorithms for resolving this problem are:
Rosenblatt Algorithm:- The Perceptron presented by Rosenblatt in 1959. The essential and important innovation
was the introduction of numerical weights and a special interlinked pattern. In the original Rosenblatt model the
computing domains are threshold elements and the connectivity is determined stochastically. Learning occurs by
adapting the weights of the network with a numerical algorithm. Rosenblatt’s model or tool was refined and
perfected in the 1960s and its computational properties were carefully discussed by Minsky and Papert [15]. In
the following, Rosenblatt’s model will be named the classical Perceptron and the model discussed by Minsky and
Papert the Perceptron.
The Perceptron forms a network with a single node and set of input links along with a dummy input which is
always set to 1 and a single output lead. The input state which could be a set of numbers is applied to each of
the connections to the node.
. Thus the preceptron equation for the class labels cK.
Ck = w0 + w1i1 + w2 i2 ………………. wnin
b) Algorithm description of Hebb: ---- Perceptron is the generic name given by the Frank Rosenblatt because it is a
model of EYE. Perceptron was an attempt to understand human memory, learning & cognitive process. Training
algorithm of perceptron is supervised learning.
It accept no. of inputs xi (i=1, 2, 3…………..n) & compute a weighted sum of these inputs. The sum is then
compared with a threshold θ of an output Y (which is either 0 or 1)
n
y=1
if
∑
wi*xi ≥ θ
i=1
n
y=0
if
∑
i=1
ISSN: 2319-1112 /V2N2:326-330 ©IJAEEE
wi*xi ≤ θ
IJAEEE ,Volume 2 , Number 2
Kamakshi et al.
The perceptron is the single transmission network consists of sensor modes, association unit & final output of
response unit.
c) Hopfield Neural Network: A simple single layer recurrent neural network. The Hopfield neural
network is accompanied with a special algorithm that defines it to
Learn to recognize schemes. A Hopfield network is a network of N such artificial neurons, which are
fully connected. The connection weight from neuron j to neuron i is given by a number Wij .
The collection of all such numbers i s represented by the weight matrix W , whose components are
Wij.
Nodes(N)network
Learning rate(ℓ)
Source node(SN)
Destination node(DN)
Connecting paths(C)=
N=7
ℓ=.4
Source node(SN)=2
Destination node(DN)=1
Connecting paths(C)=9
N=7
ℓ=.4
Source node(SN)=3
Destination node(DN)=5
Connecting paths(C)=9
Rosenblatt
Algorithm
Hebb Algorithm
Hop
Algorithm
Processing Time3.0061
Processing Time3.3365
Processing Time3.5360
Shortest distance2.350
Shortest distance2.88785
Shortest Distance2.9524
Processing time3.0115
Shortest distance2.4387
Processing Time3.3238
Shortest distance2.6550
Processing Time3.5338
Shortest distance3.1513
IV RESULT ANALYSIS& DISCUSSION
Three different Algorithms are taken for result Analysis n discussion (shortest path distance and Processing
time).Rosenblatt Algorithm, Hebbian Algorithm and Hopfield Algorithm. We can see Rosenblatt Algorithm is
superior to Hebbian and Hopfield Algorithms in terms of shortest path distance and processing time. The table form
result analysis is depicted below:-
ISSN: 2319-1112 /V2N2:326-330 ©IJAEEE
329
Co mpa ra tive Ana ly sis O f shortest Pat h O pt imiza tio n Techniques in Pa cket Sw it ching
Netwo rk ba sed o n Neura l Network
V CONCLUSION
This study proposed an analysis and comparison of shortest path Optimization using Rosenblatt and Dijkstra
Algorithms.
•
In Shortest path optimization using Rosenblatt Algorithm, various Algorithms are used for finding
shortest path optimization.
•
Shortest path optimization using Rosenblatt over other different algorithms has various advantage
.In case of shortest path optimization using Rosenblatt Algorithm over Hebbian Algorithm and
Hopfield Algorithm allows the best and least Processing time and shortest path distance covered from source
node to destination node in neural network .
Refrences:
[1] J. J. Hopfield and D. W. Tank, “‘Neural’ computations of decision in optimization problems,” Biol. Cybern., pp.
141–152, 1985.
[2] M. Ali and F. Kamoun, “Neural networks for shortest path computation and routing in computer networks,” IEEE
Trans. on Neural Networks, vol. 4, no. 6, pp. 941–953, 1993.
[3] L. Bodin, B. L. Golden, A. Assad, and M. Ball, “Routing and scheduling of vehicles and crews: The state of the
art,” Comput. Oper.Res.,vol. 10, pp. 63–211, 1983.
[4] A. Ephremides and S. Verdu, “Control and optimization methods in communication network problems,” IEEE
Trans. Automat.Contr., vol. 34, pp. 930–942, 1989.
[5] J. K. Antonio, G. M. Huang, and W. K. Tsai, “A fast distributed shortest path algorithm for a class of
hierarchically clustered data networks,” IEEE Trans. Comput., vol. 41, pp. 710–724, 1992.
[6] S. Jun and K. G. Shin, “Shortest path planning in distributed workspace using dominance relation,” IEEE Trans.
Robot. Automat.,vol. 7, pp. 342–350, 1991.
ISSN: 2319-1112 /V2N2:326-330 ©IJAEEE
IJAEEE ,Volume 2 , Number 2
Kamakshi et al.
[7] P. Soueres and J.-P.Laumond, “Shortest paths synthesis for a car-like robot,” IEEE Trans. Automat.Contr., vol. 41,
pp. 672–688, 1996.
[8] W. Stalling, High-Speed Networks: TCP/IP and ATM Design Princi-ples. Englewood Cliffs, NJ: Prentice-Hall,
1998.
[9] M. K. Ali and F. Kamoun, "Neural networks for shortest path computation and routing in computer networks,"
IEEE Trans. Neural Networks, vol. 4, pp. 941-954, Nov. 1993.
[10] D. C. Park and S. E. Choi, "A neural network based multi-destination routing algorithm for communication
network," in Proc. Joint Conf. Neural Networks, 1998, pp. 1673-1678.
[11] C. W. Ahn, R. S. Ramakrishna, C. G. Rang, and I. C. Choi, "Shortest path routing algorithm using hopfield
neural network," Electron. Lett.,vol. 37, no. 19, pp. 1176-1178, Sept. 2001.
[12] M. Munemoto, Y. Takai, and Y. Sato, "A migration scheme for the genetic adaptive routing algorithm," in Proc.
IEEE Int. Conf. Systems, Man, and Cybernetics, 1998, pp. 2774-2779.
[13] Y. Leung, G. Li, and Z. B. Xu, "A genetic algorithm for the multiple destination routing problems," IEEE Trans.
Evol.Comput, vol. 2, pp. 150-161, Nov. 1998.
[14] R.Rojas, Neural Network :ASystamaticintroduction,Springer-Verlag,Berlin 1996.
[15] Rolf Pfeifer,DanaDamian,RudolfF¨uchslin, Neural Netwok.
[16] “Exploring Genetic Algorithm for Shortest Path Optimization in Data Networks” Global journal of computer
science and technology volume 10 issue 11 ver(1.0)October 2010 by Dr. Rakesh Kumar, Mahesh Kumar.
[17] Park, Y.-K., & Lee, G. (1995). Applications of neural networks in high speed communication networks. IEEE
Communications Magazine, 33(10), 68–74.
[18] Marbach, P., Mihatsch, O., &Tsitsiklis, J. N. (2000). Call admission control and routing in integrated service
networks using neuro-dynamic programming. IEEE Journal of Selected Areas of Communication,February.
[19] Pornavalai, C., Chakraborty, G., &Shiratori, N. (1995). A neural network approach to multicast routing in realtime communication networks.Proceedings of the 1995 International Conference on Network Protocols.
[20] By Mustafa, K., Ali, M., &Kamoun, F. (1993). Neural networks for shortest path computation and routing in
computer networks. IEEE Transactions on Neural Networks, 4(6), 941–954.
[21] A. W. Mohemmed, N. C. Sahoo, T. K. Geok, “Solvingshortest path problem using particle swarm
optimization,”Applied Soft Computing, vol. 8, no. 4, pp. 1643–1653,Sept. 2008.
ISSN: 2319-1112 /V2N2:326-330 ©IJAEEE