TY - JOUR
T1 - The performance evaluation of a new neural network-based traffic management scheme for a satellite communication network
AU - Ansari, Nirwan
AU - Liu, Dequan
N1 - Funding Information:
*This work was partially supported by the NASA Lewis Research Center, Cleveland, OH, under contract number NAG3-1244, and NJIT Sponsored Research Program under contract numbers 421860 and 421250. This paper is the expanded version of a conference paper [3] presented at GLOBECOMPl. * Corresponding author. Email: [email protected]
PY - 1995/8
Y1 - 1995/8
N2 - This paper presents the first attempt to incorporate the map configuration task with the routing task in the traffic management scheme of a satellite communication network. The map configuration task, which is automated by a self-organization scheme modified from Kohonen's model, consists of three stages. The first stage is the pattern recognition task, which selects an exemplar map that best meets the current network requirement. The second stage consists of analyzing the discrepancy between the chosen exemplar map and the state of the network, and adaptively modifying the chosen exemplar map to conform closely to the network requirement (input data pattern) by self-organization. Based on certain performance criteria, whether a new map is generated to replace the original chosen map is decided in the third stage. The routing task is implemented by a state-dependent routing algorithm which arranges the incoming call to some proper path to make the network more efficient, and lowers the call block rate. The new scheme inheriting merits of both the map configuration and routing tasks provides the best compromised performance in terms of block rates at various load conditions as compared to status quo approaches - schemes with either only the map configuration task or only the routing task.
AB - This paper presents the first attempt to incorporate the map configuration task with the routing task in the traffic management scheme of a satellite communication network. The map configuration task, which is automated by a self-organization scheme modified from Kohonen's model, consists of three stages. The first stage is the pattern recognition task, which selects an exemplar map that best meets the current network requirement. The second stage consists of analyzing the discrepancy between the chosen exemplar map and the state of the network, and adaptively modifying the chosen exemplar map to conform closely to the network requirement (input data pattern) by self-organization. Based on certain performance criteria, whether a new map is generated to replace the original chosen map is decided in the third stage. The routing task is implemented by a state-dependent routing algorithm which arranges the incoming call to some proper path to make the network more efficient, and lowers the call block rate. The new scheme inheriting merits of both the map configuration and routing tasks provides the best compromised performance in terms of block rates at various load conditions as compared to status quo approaches - schemes with either only the map configuration task or only the routing task.
KW - Kohonen
KW - Map configuration
KW - Satellite communication network
KW - Self-organization
KW - State-dependent routing
KW - Traffic management
UR - http://www.scopus.com/inward/record.url?scp=0029347579&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0029347579&partnerID=8YFLogxK
U2 - 10.1016/0925-2312(95)00040-D
DO - 10.1016/0925-2312(95)00040-D
M3 - Article
AN - SCOPUS:0029347579
SN - 0925-2312
VL - 8
SP - 263
EP - 282
JO - Neurocomputing
JF - Neurocomputing
IS - 3
ER -