On-chip optical network mapping method based on growth type annular SOM neural network

Abstract

The invention discloses an on-chip optical network mapping method based on a growth type annular SOM neural network. The method comprises the following steps: acquiring an IP core connectivity relation matrix in a core graph and a node communication loss matrix in a network topology; training the growth type annular SOM neural network by using a row vector in the node communication loss matrix to obtain a trained growth type annular SOM neural network; respectively acquiring a winning neuron index number of each row vector in the node communication loss matrix and a winning neuron index number of each row vector in the IP core connectivity relation matrix by using the trained growth type annular SOM neural network; and according to the winning neuron index number of the IP core connectivity relation matrix and the winning neuron index number of the node communication loss matrix, obtaining a mapping matrix between the IP core and the network topology node in the core graph. According to the method, the complexity of the mapping method is reduced and the solving time is shortened in a self-adaptive manner under the condition of ensuring the stability of the result.

Description

technical field [0001] The invention belongs to the technical field of on-chip networks, and in particular relates to an on-chip optical network mapping method based on a growing annular SOM (Self-Organizing Map, Self-Organizing Map) neural network. Background technique [0002] The on-chip optical network has many incomparable advantages of traditional electrical interconnection, such as high bandwidth, fast transmission rate and anti-interference, and is a development direction of VLSI in the future. In the on-chip optical network, mapping is a key aspect that needs attention; under the premise of satisfying certain mapping requirements, the mapping maps the IP cores in the core map to the nodes in the topology. The location of IP core mapping in the network topology will greatly affect the performance of the network. Therefore, how to reasonably map the IP core in the core map to the network topology under the premise of meeting certain performance requirements is a probl...

AI Technical Summary

Problems solved by technology

However, the existing algorithms have high computational complexity, long solution time, and are prone to fall into local optimal solutions.

Images