Reliable Random Access for Decentralized UAV Networks Based on Raptor Codes

Abstract

In this paper, we propose the Raptor coded random access (RCRA) scheme to enable reliable transmission in the decentralized unmanned aerial vehicle (UAV) network. The considered network is composed of several overlapped random access systems with interference nodes, and the proposed RCRA scheme reduces bit-error ratio (BER) of the random access systems by three steps. First, we choose the number of slots based on a derived lower bound, which is necessary for reliable random access. Second, error-correcting codes are incorporated as the precode before random access, and then the access probability is optimized to achieve the minimum BER. Third, by correlating two consecutive slots, an idle-slot-filling approach is designed to further improve the efficiency of the random access systems. Numerical results show the proposed RCRA scheme reduces significantly both block-error ratio (BLER) and BER at moderate-and high-signal-to-noise ratio (SNR) region. With Es/N0 equal to 0 dB, the RCRA scheme saves 20% slots, compared with the existing frameless ALOHA scheme, to achieve a target BER of 10-4.