
Проведен обзор имеющихÑÑ Ð°Ð»Ð³Ð¾Ñ€Ð¸Ñ‚Ð¼Ð¾Ð² автоматичеÑкого раÑÐ¿Ð¾Ð·Ð½Ð°Ð²Ð°Ð½Ð¸Ñ Ð°Ð½Ð°Ð»Ð¾Ð³Ð¾Ð²Ñ‹Ñ… модулÑций, подробный обзор параметричеÑких алгоритмов. Получены завиÑимоÑти значений параметров Ñигналов, необходимых Ð´Ð»Ñ Ñ€Ð°Ð±Ð¾Ñ‚Ñ‹ иÑÑледуемых алгоритмов, в некотором диапазоне отношений Ñигнал/шум, а также пороговые Ð·Ð½Ð°Ñ‡ÐµÐ½Ð¸Ñ Ð´Ð»Ñ Ð°Ð»Ð³Ð¾Ñ€Ð¸Ñ‚Ð¼Ð° дерева решений. Проведено Ñравнение ÑффективноÑти метода на оÑнове дерева решений и методов машинного Ð¾Ð±ÑƒÑ‡ÐµÐ½Ð¸Ñ â€” иÑкуÑÑтвенные нейронные Ñети прÑмого раÑпроÑÑ‚Ñ€Ð°Ð½ÐµÐ½Ð¸Ñ Ð¸ метод k-ближайших ÑоÑедей. Получены вероÑтноÑти верного и ошибочного Ñ€ÐµÑˆÐµÐ½Ð¸Ñ Ð´Ð»Ñ ÑƒÐºÐ°Ð·Ð°Ð½Ð½Ñ‹Ñ… алгоритмов. Ðа Ñовременной программируемой Ñлементной базе предÑтавлена Ñ€ÐµÐ°Ð»Ð¸Ð·Ð°Ñ†Ð¸Ñ Ð½Ð°Ð¸Ð±Ð¾Ð»ÐµÐµ Ñффективного автоматичеÑкого алгоритма раÑÐ¿Ð¾Ð·Ð½Ð°Ð²Ð°Ð½Ð¸Ñ Ð¼Ð¾Ð´ÑƒÐ»Ñции по критериÑм минимума вероÑтноÑти ошибочного Ñ€ÐµÑˆÐµÐ½Ð¸Ñ Ð¸ вычиÑлительных реÑурÑов.
A review of the available algorithms for automatic recognition of analog modulations, a detailed review of parametric algorithms. The dependences of the values of the parameters of the signals necessary for the work of the investigated algorithms, in a certain range of signal-to-noise ratio, as well as the threshold values for the decision tree algorithm are obtained. A comparison was made of the effectiveness of the method based on a decision tree and machine learning methods — feed-forward artificial neural networks and the method of k-nearest neighbors. The probabilities of the correct and erroneous solution for the specified algorithms are obtained. On the modern programmable element base, the implementation of the most effective automatic modulation recognition algorithm is presented according to the criteria of minimum probability of an erroneous solution and computational resources.
клаÑÑиÑикаÑÐ¸Ñ Ð¼Ð¾Ð´ÑлÑÑий, деÑево ÑеÑений, machine learning, automatic modulation recognition algorithms, меÑод k-ближайÑÐ¸Ñ ÑоÑедей, decision tree, маÑинное обÑÑение, иÑкÑÑÑÑвеннÑе нейÑоннÑе ÑеÑи, алгоÑиÑÐ¼Ñ Ð°Ð²ÑомаÑиÑеÑкого ÑаÑÐ¿Ð¾Ð·Ð½Ð°Ð²Ð°Ð½Ð¸Ñ Ð¼Ð¾Ð´ÑлÑÑий, modulation classification, k-nearest neighbors method, artificial neural networks
клаÑÑиÑикаÑÐ¸Ñ Ð¼Ð¾Ð´ÑлÑÑий, деÑево ÑеÑений, machine learning, automatic modulation recognition algorithms, меÑод k-ближайÑÐ¸Ñ ÑоÑедей, decision tree, маÑинное обÑÑение, иÑкÑÑÑÑвеннÑе нейÑоннÑе ÑеÑи, алгоÑиÑÐ¼Ñ Ð°Ð²ÑомаÑиÑеÑкого ÑаÑÐ¿Ð¾Ð·Ð½Ð°Ð²Ð°Ð½Ð¸Ñ Ð¼Ð¾Ð´ÑлÑÑий, modulation classification, k-nearest neighbors method, artificial neural networks
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
