Proposal of Secure and Fast Vector Quantization for Edge Computing

Abstract

The use of cloud computing system, which is the basic technology supporting ICT, is expanding. However, when each sever is connected with much terminals, the server shows low capability compared to being connected with small number of terminals. The limit of capacity of servers for cloud system leads to significant processing time delay. To avoid it, edge computing system for IoT has been proposed. In the edge system, a plural of servers called edges are connected directly or close distance between the server and the terminal (or thing). In the previous paper, we proposed the effective BP learning method for the edge system. In this paper, we propose edge computing systems k-means and Neural Gas method and show the effectiveness of them in some simulations.