Studying on Multi-constraint Multicast QoS Routing Problem in SRIO Embedded Network
This paper focus on the multi-constraint multicast Quality of Service (QoS) routing problem in SRIO network. First, the system model of SRIO network is created by graph theory; second, a novel algorithm named bandwidth-delay-jitter-constrained multicast r
- PDF / 666,077 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 45 Downloads / 176 Views
Abstract This paper focus on the multi-constraint multicast Quality of Service (QoS) routing problem in SRIO network. First, the system model of SRIO network is created by graph theory; second, a novel algorithm named bandwidth-delayjitter-constrained multicast routing based on ant colony system (BDJCMRACS) algorithm is designed for solving the multi-constraint multicast QoS routing problem in SRIO networks; finally, the effectiveness of the BDJCMRACS algorithm is tested by simulation. Keywords SRIO
QoS Multicast Ant colony algorithm
1 Introduction SRIO is a kind of high-bandwidth, low-latency, and reliable connecting technology, which has been widely adopted in embedded processor for Radar, Sonar, and Tele-communication [1]. As SRIO network may contain 65536 endpoints and unlimited switches theoretically, sometimes it will be large in scale and complicated in topology. In order to utilize the network resources effectively and set the multicast routing tables efficiently, this paper focus on the multi-constraint multicast Quality of Service (QoS) routing problem in SRIO network. SRIO’s X. Zhang (&) M. Gao H. Luo Y. Ma Beijing Institute of Technology, Beijing, China e-mail: [email protected] M. Gao e-mail: [email protected] H. Luo e-mail: [email protected] Y. Ma e-mail: [email protected]
Z. Wen and T. Li (eds.), Foundations of Intelligent Systems, Advances in Intelligent Systems and Computing 277, DOI: 10.1007/978-3-642-54924-3_76, Springer-Verlag Berlin Heidelberg 2014
811
812
X. Zhang et al.
multicast is based on the relay method of spanning tree. Through the construction of the multicast tree which covers the source node and the destination nodes, the packet sent by the source node will pass through the multicast tree concurrently and reaches the destination nodes, respectively. The data packets are duplicated only at the branching [2]. There are many types of multicast tree, including the shortest path tree, the smallest spanning tree, the Steiner tree, the single-constraint Steiner tree, and the multi-constraint Steiner tree. The Steiner tree is the least-cost spanning tree which covers the source node and the destination nodes. Constructing the Steiner tree is a NP-Complete problem [3], whose optimal solution cannot be obtained within polynomial time. The Multi-constraint Steiner tree is the least-cost multicast tree which meets the constraints required. It may be obtained through combinatorial optimization algorithms, such as the neural network algorithm [4], genetic algorithm [5], and ant colony algorithm [6]. In the SRIO network, the multi-constraint multicast QoS routing is constructing the multi-constraint Steiner tree, which covers the source and destination nodes and meets the bandwidth, delay and delay jitter requirements. First, the system model of SRIO network is created by graph theory; second, a novel algorithm named bandwidth-delay-jitter-constrained multicast routing based on ant colony system (BDJCMRACS) algorithm is designed for solving the multiconstraint mult
Data Loading...
