The Shortest Distance Path (SDP) problem is a critical routing issue in communication networks, particularly in satellite networks. Typically, SDP is solved by graph-based iterative algorithms, while an explicit or analytic approach is challenging. However, considering the orbit dynamics and topology regularity, this paper proposes, for the first time, an explicit analytic phase-based algorithm STEPCLIMB to directly solve the SDP in low-Earth orbit (LEO) satellite networks. Based on the relationship between satellite phase and inter-satellite link distance, the SDP is modeled with the satellite phase, and SDP problem is converted into a total phase offset problem through theoretical derivations. Then STEPCLIMB is derived in two cases, respectively. Monte-Carlo simulations verify STEPCLIMB’s accuracy, which has zero error in the mono-valley case and has less than 0.1% error in the bi-valley case. The algorithm performs better in larger-scale constellations and can save over 99.4% computational cost compared to Dijkstra algorithm. Further, the SDP pattern and features in Starlink constellation are analyzed. The model proves that most inter-plane hops in the SDP occur successively, and the simulations further indicate that these hops prefer satellites in the higher latitude regions.

This paper proposes a new network approach towards the implementation of T-S fuzzy models referred to as disjunctive fuzzy neural networks (DJFNN). The proposed DJFNN involves a novel network architecture and a greedy learning algorithm. Being different from existing grid-based and clustering-based network architectures, the proposed architecture adds an OR neural layer positioned between the fuzzification layer and the rule layer. In this way, the implied constraint between the number of rules and the number of fuzzy labels is excluded so that a curse of dimensionality can be overcome and more interpretable models can be formed. Further, inspired by the core algorithm for building a decision tree, a top-down, non-backtracking, and greedy algorithm is proposed to learn the unknown parameters of the networks. The input space splits into smaller and smaller subspace along the pre-defined fuzzy grids in a supervised manner meanwhile the associated conditions of T-S fuzzy model are identified. The greedy algorithm is applicable to high-dimensional problems since there is no exponential growth in time or space as the dimensionality increases. The new network architecture and greedy learning algorithm make the proposed DJFNN a regression model of high interpretability and good prediction capability, particularly suitable for solving high-dimensional problems. DJFNN was experimented with using a synthetic dataset and 28 real-world datasets and compared with classical and state-of-the-art methods through non-parametric statistical tests. The results confirmed the effectiveness of DJFNN in terms of accuracy, interpretability, and computational cost.

Polynomial Regression Surface (PRS) is a commonly used surrogate model for its simplicity, good interpretability, and computational efficiency. The performance of PRS is largely dependent on its basis functions. With limited samples, how to correctly select basis functions remains a challenging problem. To improve prediction accuracy, a PRS modeling approach based on multitask optimization and ensemble modeling (PRS-MOEM) is proposed for rational basis function selection with robustness. First, the training set is partitioned into multiple subsets by the cross validation method, and for each subset a sub-model is independently constructed by optimization. To effectively solve these multiple optimization tasks, an improved evolutionary algorithm with transfer migration is developed, which can enhance the optimization efficiency and robustness by useful information exchange between these similar optimization tasks. Second, a novel ensemble method is proposed to integrate the multiple sub-models into the final model. The significance of each basis function is scored according to the error estimation of the sub-models and the occurrence frequency of the basis functions in all the sub-models. Then the basis functions are ranked and selected based on the bias-corrected Akaike’s information criterion. PRS-MOEM can effectively mitigate the negative influence from the sub-models with large prediction error, and alleviate the uncertain impact resulting from the randomness of training subsets. Thus the basis function selection accuracy and robustness can be enhanced. Seven numerical examples and an engineering problem are utilized to test and verify the effectiveness of PRS-MOEM.