.png){kind=icon}
.png){kind=icon}
Return
Optimizing deep convolutional neural network with an IP-bounded binary optimization and optimal extreme learning machine for pulse repetition interval modulation recognition
.png){kind=icon}
PRE
AI
DOI：10.1007/s10470-026-02568-4.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The recognition of pulse repetition interval (PRI) modulation is one of the primary tasks of the modern electronic intelligence system (ELINT) and electronic support measure system (ESM) for accurately identifying threat radars. The PRI modulation recognition is a complex and challenging issue due to missing and spurious pulses and large outliers, which cause a very noisy sequence of PRI modulation changes in authentic settings. This paper presents an innovative three-phase technique to recognizing the five types of standard PRI modulation. First, an optimal deep convolutional neural network (ODCNN) structure is formed by the Internet Protocol-based bounded binary optimization (IP-BBO) using the data set as a feature extractor. Then, in the second step, the last fully connected layers of ODCNN are replaced by an extreme learning machine (ELM) to improve the time complexity of the proposed model and for real-time recognition. After that, in the third stage, BBO was introduced to adjust the biases and weights of ELM to reduce the complexity of the suggested method space. The proposed method performs better than other methods, with an accuracy of 99.22% and a training time of 109.46 s. The results indicate that the proposed method can be reliable and efficient in applications related to identifying PRI modulation by providing robust performance and high accuracy in all evaluation criteria. Also, the proper training time of this method has made it an ideal option for practical applications and sensitive environments that require fast and accurate processing.
Keywords:
Deep convolutional neural network
Pulse repetition interval
Bounded binary optimization
Extreme learning machines
Deep convolutional neural network
Pulse repetition interval
Bounded binary optimization
Extreme learning machines
Journal
IF:
1.4
Papers:
178
Citations:
2.1K