Based on simulation samples of open source data collection and communication system signal pretreatment methods, we design a Convolution Neural Network (CNN) that contains three full connection layers and three convolution layers to recognize 11 different kinds of wireless signal modulation. In additive white Gaussian noise(AWGN) model, CNN networks can realize automatic modulation classification of the wireless signal by learning the deep characters of sample data, and the average recognition accuracy is up to 81% under different ratio of signal to noise (SNR). We analyze the influence between modulation recognition accuracy and network hyperparameters such as network structures, convolutional layers, dropout value, batch size, and network training methods.

With the development and universal application of satellite technology, an important way to expand the function of satellites is setting up inter-satellite networks to make them work together. Traditional satellite networking methods generally adopt a fixed time slot allocation method, which is not suitable for small satellite groups with low latency and high throughput requirements. In order to solve this problem, it has been proposed to apply the traditional Wifi protocol in satellite networking. As there are differences between satellite networks and terrestrial networks, it’s necessary to improve the traditional 802.11 protocol. The Media Access Control (MAC) protocol in 802.11 is improved in this paper, which mainly includes the adaptive algorithm of maximum contention window size and the growth algorithm of Contention Window (CW) size. The maximum contention window is adjusted according to the conflict state of the current network, which makes the network accommodate more satellite nodes. The CW growth algorithm improves the traditional Binary Exponential Back-off (BEB) algorithm, where the CW is designed according to the priority of the data frame or the network load. In this way, high-priority satellite accusation information will have higher reliability or tolerate greater network load.