14 March 2025 in IEEE Transactions on Communications
The 3rd Generation Partnership Project (3GPP) 5G Non-Terrestrial Networks (NTN) adopt Orthogonal Frequency Division Multiplexing (OFDM) to enable integrated space-ground networks via Low Earth Orbit (LEO) satellite global connectivity. However, the rapid movement of LEO satellites induces a significant non-uniform Doppler shift across subcarriers, resulting in the signal bandwidth changes that leads to sampling point offsets and severely impacting demodulation performance. Traditional Doppler compensation algorithms focus mainly on addressing the uniform Carrier Frequency Offset (CFO) caused by crystal oscillation. In LEO satellite broadband communication systems, the Sampling Frequency Offset (SFO), caused by the non-uniform Doppler frequency shift, can be tens of times greater than the CFO, leading to non-negligible phase rotation, inter-carrier interference (ICI), and inter-symbol interference (ISI). Consequently, a low-complexity algorithm is required to address the fast time-varying SFO — one of the core challenges in these systems. In this paper, we derive a closed-form expression for the signal distortion and propose a model-driven compensation method that leverages the predictability of satellite trajectories. The proposed method effectively removes phase rotation and mitigates ICI and ISI through fast Fourier transform (FFT) window adjustment and phase rotation compensation. The accuracy of the model is validated through extensive simulations and a real satellite communication trial. Results demonstrate that the proposed method supports higher-order modulations, leading to an average spectral efficiency improvement of approximately 50%. This pioneering research promises to ensure robust performance in dynamic LEO satellite environments.
20 April 2023 in IEEE Signal Processing Letters
In this letter, we propose a new nonlinear precoding algorithm, List encoding of Vector Perturbation precoding (LVP) for Multiple Input Multiple Output (MIMO) system. Different from the traditional VP based on sphere search strategy to minimize the Frobeniuas norm of the precoded signal with a exponential search complexity, LVP adopts multiple encoders design, a list of encoder perform only one dimensional integer optimization through several iterations in parallel. During each iteration each encoder searches through several perturbation results and retain one of them to the next iteration. When the iterations ends, the historical optimal perturbation result with the least Frobenius norm is output as the precoding result. The search complexity of LVP is $\mathcal {O}(M)$ which increases linearly with the antenna number $M$ . This complexity is much smaller than sphere search based strategy. Compared with other low dimensional search algorithm represented by D2VP, the search complexity drops from $\mathcal {O}(M^{2})$ to $\mathcal {O}(M)$ , with a significant BER gain thanks to the multiple encoder design. To the best of our knowledge, with similar BER performance the LVP algorithm has the lowest complexity.
1 April 2022 in Journal of Physics: Conference Series
In the on-line radiation monitoring of γ -ray source, the active online monitoring module can realize real-time recording of the radiation dose equivalent value and count rate in the detector. Because the amount of radiation at different measuring points is different, their values vary from place to place, so there is no exact value that is completely suitable for all kinds of measurement situations, so the nonlinear relationship between counting rate and dose rate is deduced. For the difference of the dose at the measuring point, the relationship between them will increase with the continuous accumulation of radiation dose, which will have an impact on the iterative algorithm. In this paper, the LM algorithm is used to solve the fitting function model, and the feasibility of the algorithm is verified by MATLAB software programming. The results show that the parameters of the objective function model can be obtained by the LM algorithm, and the relative error of the eighth polynomial can be minimized and a better fitting effect can be achieved.