Оптимизация зоны покрытия LTE-сетей методом динамического программирования

Abstract

A method for planning a radio network by a cellular operator has been proposed. The urgency of the task is due to the desire of service providers to minimize the costs of the distribution and maintenance of the required number of base stations while ensuring the required signal / noise level within the coverage area. The problem is solved by a rational choice of the location of the repeater base points in several stages. First of all, an analysis of a site is carried out and a preliminary selection is made on it of a set of points where base stations can potentially be located. As a rule, this problem is solved using specialized geographic information systems. At the second stage, the calculation of the radio range is carried out, provided the location of the repeaters at the selected points. At the third stage, on the basis of preliminary calculations, the selection of the positions of the base stations is carried out directly. It is proposed to use a stochastic approach for a rational choice of positions. The expediency of this approach is due to the large number of random uncontrolled factors affecting the signal-to-noise ratio at the receiving point. In this case, the problem of optimal choice of position is considered as an extremal problem, in which the parameters of the conditions are random variables. The implementation of the method proposed by the authors in the paper is illustrated by the example of radio communication planning on a country-type site, which is intersected by two highways. As input parameters in the example, the length of the sections of roads crossing the coverage areas of the repeaters and the concentration of subscribers on the routes is used. The objective function and the system of restrictions are compiled in such a way as to ensure that subscribers are located within the coverage area and to minimize the areas of intersection of the coverage areas of the neighboring repeaters. The use of this technique will allow us to obtain a statistically optimal arrangement of base stations on a given piece of terrain, which is confirmed by simulation modeling.