Performance Improvement of GPS GDOP Approximation Using Recurrent Wavelet Neural Network

Tafazoli, Sadaf; Mosavi, Mohammad Reza

doi:10.4236/jgis.2011.34029

Cited by 9 publications

(7 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The computation of gradient descent is also expensive, as each iteration involves a time-consuming process of line search. Tafazoli and Mosavi developed a recurrent wavelet neural network (RWNN) to compute GDOP, where the activation function uses gradient steepest descent to improve the learning speed and network reliability [12]. However, due to the similar learning process as BPNN, the RWNN also suffers from the problem of local minimum.…”

Section: Introductionmentioning

confidence: 99%

Model Predictive Filtering Based Neural Networks for GPS GDOP Approximation

Yi¹,

Shesheng²,

Zhong³

et al. 2016

J Aerosp Eng Mech

View full text Add to dashboard Cite

This paper presents a new method to calculate the geometric dilution of precision (GDOP) of GPS by incorporating the concept of model predictive filtering in the training process of neural networks to learn the relationship between GDOP and the azimuth and elevation of satellite. This method overcomes the shortcomings of the traditional back propagation neural networks, such as the slow convergence speed and easily falling into local minimum. A model predictive filtering algorithm is developed by using network weights as system state variables to optimize the network weights based on the neural network's error correction. During the training process, the neural network model error is corrected by compensating the deviation between the actual and target output via the model predictive filtering. Experimental results and comparison analysis demonstrate that the proposed method can effectively approximate GDOP with improved accuracy and reduced training time.

show abstract

Section: Introductionmentioning

confidence: 99%

Model Predictive Filtering Based Neural Networks for GPS GDOP Approximation

Yi¹,

Shesheng²,

Zhong³

et al. 2016

J Aerosp Eng Mech

View full text Add to dashboard Cite

show abstract

“…Almost all satellite selection optimization methods based on ML algorithm are mainly divided into two categories, that is, approximation and classification. A group of methods based on approximate calculation are proposed in order to simplify the calculation of GDOP, such as NNs [27], SVM [28]. However, the GDOP approximation methods just treat satellite selection as a regression problem of GDOP calculation, and the optimal subset is also need to be selected from all satellite subsets by the original brute force means.…”

Section: Related Workmentioning

confidence: 99%

“…The experimental results including minimum NWGDOP and the convergence iteration number of 16 sets of tests are also shown in Table 1. Put the minimum NWGDOP of each group of tests into (26) to get S/N NW GDOP , put the iteration number into (27) to get S/N iteration , and then add the two SNRs to get the required minimum SNR. As a result, we obtain the optimal combination of parameters : K = 10, α = 1, β = 5, ρ = 0.7, Q = 100.…”

Section: S/n Nwmentioning

confidence: 99%

A Navigation Satellites Selection Method Based on ACO With Polarized Feedback

Hong

Xia

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Selecting the optimal satellite subset for positioning from all satellites in view can not only achieve positioning accuracy but also reduce the computational burden. In this paper, a navigation satellites selection method is proposed based on ant colony optimization with the improvement of polarized feedback (ACO-PF). Firstly, the satellite selection problem is described as a combinatorial optimization problem, and the noise weighted geometric dilution of precision (NWGDOP) is defined as the criterion for satellite selection. Then the ant colony optimization (ACO) is incorporated to solve the problem, and a polarized feedback mechanism is presented to improve the convergence speed of algorithm. Meanwhile, a perturbation operator is designed to improve the global searching ability of the algorithm. The numerical experimental results show that ACO-PF can select the superior satellites combination which provides highprecision positioning. And its convergence outperforms the related algorithms by up to 50%. Besides, the achieved NWGDOP of ACO-PF is usually 0.065 smaller than ACO. Therefore, the ACO-PF method can be considered as a promising candidate for satellite selecting in navigation applications.INDEX TERMS Ant colony optimization (ACO), noise weighted, geometric dilution of precision (GDOP), polarized feedback, perturbation operator

show abstract

“…Various kinds of neural works [8][9][10][11][12] are established to obtain GDOP directly from observation vector of satellites. Mosavi [13] uses genetic algorithm to obtain the approximated value of GDOP without trainings needed in neural networks.…”

Section: Problem Formulationmentioning

confidence: 99%

“…Another group of methods based on approximate calculation are also proposed in order to simplify the calculation of GDOP. Various kinds of neural works [8–12] are established to obtain GDOP directly from observation vector of satellites. Mosavi [13] uses genetic algorithm to obtain the approximated value of GDOP without trainings needed in neural networks.…”

Section: Problem Formulationmentioning

confidence: 99%

Fast satellite selection method for multi‐constellation Global Navigation Satellite System under obstacle environments

Peng

2014

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

Satellite selection plays an important role in a multi-constellation Global Navigation Satellite System (GNSS) receiver. A well selected satellite combination can provide better positioning accuracy and lighter computational load. In this study, typical obstacle environments in urban areas and their effects on satellite selection are discussed. A distribution-free satellite selecting method is proposed. The main idea of this proposed method is to expand the selected satellite subset sequentially. The performance of the proposed methods is validated and compared with four other methods by simulation in typical scenarios. Simulation results show that the proposed method is suitable for urban GNSS applications.

show abstract

Performance Improvement of GPS GDOP Approximation Using Recurrent Wavelet Neural Network

Cited by 9 publications

References 9 publications

Model Predictive Filtering Based Neural Networks for GPS GDOP Approximation

Model Predictive Filtering Based Neural Networks for GPS GDOP Approximation

A Navigation Satellites Selection Method Based on ACO With Polarized Feedback

Fast satellite selection method for multi‐constellation Global Navigation Satellite System under obstacle environments

Contact Info

Product

Resources

About