DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING, MAULANA AZAD NATIONAL INSTITUTE OF TECHNOLOGY BHOPAL, MADHYA PRADESH 462003
PROFESSOR CHANDRA MURTHY
DEPARTMENT OF ELECTRICAL AND COMMUNICATION ENGINEERING, INDIAN INSTITUTE OF SCIENCE, BENGALURU, KARNATAKA 560012
In ALOHA random protocol replicas of a packet are transmitted at randomly selected slots. For decoding, the receiver needs to have complete knowledge of the replica positions of a packet prior to decoding. Each replica contains a pointer containing the information of replica positions. In the present work, we present a deterministic setup (DS) for transmitting packets. The proposed DS model can be realized via a deterministic binary matrix, where users and slots are realized along the columns and rows respectively. Each user is assigned a unique column. The number of ones of a column dictates the number of replicas and the positions of ones dictate the slots where the replicas are transmitted. There are several advantages of using the DS model. Firstly, as the receiver has exact knowledge of the replica positions of every user’s packet, replicas don’t need to contain pointers. Secondly, one can bring the vast literature on deterministic binary matrices into play to choose matrices suitable for decoding the packets in the DS model. t-disjunct matrices are especially helpful in the decoding process due to its unique property. In particular, we show that if in a given timeframe the number of active users is less or equal to t+1, SIC (successive interference cancellation) decodes all the active users. In numerical analysis, we show that the throughput performance of DS model is comparable with respective ALOHA protocols.
Keywords:Deterministic Setup, interference cancellation, multiaccess communication, binary matrices