.png){kind=icon}
.png){kind=icon}
Return
Short-Term Prediction and Application Research of BDS-3 Satellite Clock Offset Based on the IWOA-Mamba network
.png){kind=icon}
PRE
AI
T
J
J
J
DOI：10.1016/j.asr.2026.05.019.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
To address the challenge of low prediction accuracy for nonlinear satellite clock offsets in traditional neural networks, this work presents a novel prediction network based on IWOA-Mamba. By introducing Kent chaotic mapping, adaptive inertia weight, t disturbance, and a greedy selection strategy into the whale optimization algorithm (WOA), a balance has been achieved between global search and local optimization, minimizing the likelihood of getting stuck in local optima and improving the algorithm’s optimization ability. Then, the hyperparameters obtained by the improved whale optimization algorithm (IWOA) were introduced into the Mamba network to construct the IWOA-Mamba network, and it was applied in the short-term clock offset prediction of the Beidou-3 navigation satellite system (BDS-3). The experimental findings indicate that, compared with the quadratic polynomial (QP), grey model (GM), Transformer, long short-term memory neural model (LSTM), Mamba network, and WOA-Mamba network, the average prediction accuracy of the IWOA-Mamba network for single-day prediction has increased by 35.23%, 38.94%, 26.90%, 19.34%, 14.69%, and 12.29%, respectively. The average prediction accuracy of multi-day predictions increased by 32.47%, 34.07%, 24.35%, 21.69%, 17.57%, and 11.12%, respectively. In precise point positioning (PPP) applications, the IWOA-Mamba network achieves average positioning standard deviations of 3.64 cm, 4.25 cm, and 7.54 cm in the E, N, and U directions, respectively. The ability of the IWOA-Mamba network to enhance the prediction accuracy of short-term clock biases has been verified.
Keywords:
IWOA-Mamba network
BDS-3 satellite clock offset
short-term prediction
whale optimization algorithm
chaotic mapping
Journal
IF:
2.8
Papers:
1.3K
Citations:
2.0W
Organization
No organization information available
Cited Papers