Настройки

Укажите год
-

Небесная энциклопедия

Космические корабли и станции, автоматические КА и методы их проектирования, бортовые комплексы управления, системы и средства жизнеобеспечения, особенности технологии производства ракетно-космических систем

Подробнее
-

Мониторинг СМИ

Мониторинг СМИ и социальных сетей. Сканирование интернета, новостных сайтов, специализированных контентных площадок на базе мессенджеров. Гибкие настройки фильтров и первоначальных источников.

Подробнее

Форма поиска

Поддерживает ввод нескольких поисковых фраз (по одной на строку). При поиске обеспечивает поддержку морфологии русского и английского языка
Ведите корректный номера.
Ведите корректный номера.
Ведите корректный номера.
Ведите корректный номера.
Укажите год
Укажите год

Применить Всего найдено 2. Отображено 2.
23-05-2023 дата публикации

Concurrent communication channel allocation method based on phased array antenna azimuth multi-beam

Номер: CN116156640A
Принадлежит:

The invention relates to the technical field of wireless communication networks of communication dedicated phased arrays or multifunctional phased arrays, and provides a concurrent communication channel allocation method based on phased array antenna azimuth multiple beams. When the phased array communication equipment receives, a certain number of digital beam forming units and baseband communication receiving processing units are added, and different azimuth receiving beam pointing parameters are given to the digital beam forming units, so that space multi-channel parallel receiving is realized; on the basis of a conventional TDMA (Time Division Multiple Access) mode, a network MAC (Media Access Control) protocol is combined with a multi-beam parallel receiving capability to uniformly distribute receiving time slots of all nodes, and all neighbor nodes send data to the nodes in the receiving time slots of the nodes; in consideration of transmission conflicts under a special network topology ...

Подробнее
01-08-2023 дата публикации

Multi-edge server low-delay semi-central federated learning method

Номер: CN116527520A
Принадлежит:

The invention relates to a federated machine learning model training method based on multiple edge servers. A semi-central federated learning system is built through multiple Internet of Things devices and edge servers; in the initial stage of each training period t, the edge server formulates an association strategy A (t) of the Internet of Things equipment and the edge server according to the channel information and the calculation and transmission capability of the Internet of Things equipment, and the aim is to achieve given training loss with low training time delay under limited bandwidth resources; in the training process, the Internet of Things equipment carries out local training and sends an updated local model to an edge server associated with the Internet of Things equipment for edge aggregation, and a new edge model is generated; the edge server transmits the updated edge model to an edge server close to the edge server; and after the models sent by all other edge servers are ...

Подробнее