Deep learning has achieved remarkable results in various fields, such as image recognition and classification. However, in the recognition of radio modulation methods, deep learning for different modulation methods of radio signal recognition results are not satisfactory. In this paper, we propose to use densely connected convolutional networks combined with bidirectional recurrent neural networks to identify the radios of 11 different modulation methods. The final results show that our method is more accurate than the traditional convolution neural network in modulation recognition.

No abstract

No abstract

In order to meet the requirement of heavy computation and hard real-time in SAR system, a software cache technique was designed for the geometric correction of SAR images. We studied the address mapping, pixel picking algorithm and cache replacement rule in line with the process feature of geometric correction. A result image pixel ergodic schemes was figured out for improving the cache hit probability. In addition, an algorithm was developed to figure out the most efficient cache line strategy adaptively. The software cache module was tested in two platforms, resulting in remarkable efficiency improvement of SAR geometric correction.