Dynamic Bandwidth Management and Information Protection in Service Channels of Non-Stationary Objects

Abstract

The paper is devoted to development and integration of the dynamic method of bandwidth management in service (control) communication channels for non-stationary objects. Polar codes (PC) were chosen as a scheme to combat interference in communication channels. Firstly, polar codes allow us to achieve the asymptotically possible bandwidth of a binary channel without memory. Secondly, they have the opportunity to freely choose the required code distance within the framework of the Hamming metric. The Bhattacharyya distance (BD) is used to provide the second advantage. This is a universal mechanism for determining the weight of a matrix row in a common generating matrix for the subsequent allocation of the most “strong/significant” matrix rows and the construction of any code vectors. It is important to note that the paper does not use the classical scheme for determining the likelihood ratio logarithm (LLR) but a modified one. For this purpose, there are used properties of the erasing communication channel (a wide erasure interval), integer values of the likelihood ratio logarithm to minimize computational costs (leveling time delays) and the assumption that the scope of variation changes quasi-linearly in the case of mathematical expectation. The paper describes the switching modes and the mechanism of bandwidth management that is determined by the tasks facing the designer and the operator of the communication system. Simulation was performed on FPGA Xilinx Artix-7 XC7A100T.