IMITATIVE STUDY OF THE UPPER LIMIT OF THE NODES-REPEATERS AMOUNT IN THE OLSR PROTOCOL

Abstract

Background. At present, the demand in MANET networks is determined by the inherent advantages of these networks over fixed networks, especially in emergency situations. The topology in such networks is extremely dynamic, which required the development of special routing protocols. Although the basics of the theory of the OLSR protocol, which is widely used in these networks, are well known, however, the development of mobile networks requires constant refinements of theoretical positions in accordance with new data on how to calculate the number of multipoint repeaters. Objective. The aim of the paper is to find analytical expressions for calculating the upper limit of the number of multipoint repeaters, the probability of the location of two-hop nodes at the maximum distance from the reference, and checking the correspondence of the results obtained using simulation. Methods. Analytical methods of calculation are used, as well as simulation modeling in the NS-2 software package. Results. Analytical expressions are obtained for calculating the upper limit of the number of multipoint repeaters and the probability of the location of two-hop nodes at the maximum distance from the reference node. A simulation was performed to confirm the theoretical results. Conclusions. The upper bound of the number of one-hop relay nodes is estimated. It is shown that with a low density of nodes in the area under consideration, the value of the upper limit is two times less than that previously determined in the scientific literature. Modeling in the NS-2 software package showed the correspondence of analytical expressions for the calculation of the upper boundary of multipoint repeaters. There are cases when analytical expressions give false results. For these cases, it was proved that regardless of the network density, each two-hop node that is at the maximum distance from the reference node must have its own multipoint repeater. The probability of the location of two-knot nodes at the maximum distance from the reference node is determined.