scispace - formally typeset
Search or ask a question
Author

Han Duan

Bio: Han Duan is an academic researcher. The author has contributed to research in topics: Null (mathematics). The author has an hindex of 1, co-authored 1 publications receiving 2 citations.

Papers
More filters
Journal ArticleDOI
TL;DR: A novel beamforming algorithm (named as SVR-CMT algorithm) is presented for controlling the sidelobes and the nullling level and can improve the output signal-to-interference-and-noise ratio (SINR) performance even if the direction-of-arrival (DOA) errors exist.
Abstract: Minimum variance distortionless response (MVDR) beamformer is an adaptive beamforming technique that provides a method for separating the desired signal from interfering signals. Unfortunately, the MVDR beamformer may have unacceptably low nulling level and high sidelobes, which may lead to significant performance degradation in the case of unexpected interfering signals such as the rapidly moving jammer environments. Via support vector machine regression (SVR), a novel beamforming algorithm (named as SVR-CMT algorithm) is presented for controlling the sidelobes and the nullling level. In the proposed method, firstly, the covariance matrix is tapered based on Mailloux covariance matrix taper (CMT) procedure to broaden the width of nulls for interference signals. Secondly, the equality constraints are modified into inequality constraints to control the sidelobe level. By the ε-insensitive loss function for the sidelobe controller, the modified beamforming optimization problem is formulated as a standard SVR problem so that the weight vector can be obtained effectively. Compared with the previous works, the proposed SVR-CMT method provides better beamforming performance. For instance, (1) it can effectively control the sidelobe and nullling level, (2) it can improve the output signal-to-interference-and-noise ratio (SINR) performance even if the direction-of-arrival (DOA) errors exist. Simulation results demonstrate the efficiency of the presented approach.

3 citations


Cited by
More filters
Journal ArticleDOI
TL;DR: A novel selection technique is introduced that will choose the best particle among the population within the search domain to achieve a high-performance exploration and a dynamic parameter strategy is proposed for further facilitating the algorithm and tradeoff between exploration and exploitation searches.
Abstract: Quantum inspired particle swarm optimization (QPSO) stimulated by perceptions from particle swarm optimization and quantum mechanics is a stochastic optimization method. Although, it has shown good performance in finding the optimal solution to many electromagnetic problems. However, sometimes it falls to local optima when dealing with hard optimization problems. Thus, to preserve a good balance between local and global searches to avoid premature convergence in quantum particle swarm optimization, this paper proposed three enhancements to the original QPSO method, the proposed method is called modified quantum particle swarm optimization (MQPSO) algorithm. Firstly, a novel selection technique is introduced that will choose the best particle among the population within the search domain to achieve a high-performance exploration. Secondly, a new mutation method is used to preserve the easiness of available QPSOs. Also, a dynamic parameter strategy is proposed for further facilitating the algorithm and tradeoff between exploration and exploitation searches. The experimental results obtained by solving standard benchmark functions and an electromagnetic design problem which is the superconducting magnetic energy storage (SMES) system available in both three parameters and eight parameters problems are reported to showcase the usefulness of the proposed approach.

27 citations

Journal ArticleDOI
TL;DR: In this paper , a two-step learning-by-examples (LBE) strategy based on support vector regression (SVR) is proposed to learn the complex relation among the heterogeneous parameters of the power plant.
Abstract: |In this paper, an innovative machine learning ( ML ) approach for the prediction of the output power generated by photovoltaic ( PV ) plants is presented. Toward this end, a two-step learning-by-examples ( LBE ) strategy based on support vector regression ( SVR ) is proposed to learn the complex relation among the heterogeneous parameters affecting the energy production of the power plant. More speci(cid:12)cally, the (cid:12)rst step is aimed at down-scaling the weather forecasts from the standard air temperature and the solar irradiance to the local module temperature and the plane-of-array ( POA ) irradiance. Then, the second step predicts the output power pro(cid:12)le given the down-scaled forecasts estimated at the previous step. The advantages and limitations of the proposed two-step approach have been experimentally analyzed exploiting a set of measurements acquired in a real PV plant. The obtained results are presented and discussed to point out the capabilities of the proposed LBE method to provide robust and reliable power predictions starting from simple weather forecasts.

5 citations

Journal ArticleDOI
Feng Yang1, Guochen Pei1, Lingna Hu, Lianghui Ding1, Yang Li 
TL;DR: The goal is maximizing the minimum signal-to-interference-plus-noise ratio (SINR) while depressing the maximum sidelobe level (SLL) of all users to solve the problem effectively.

4 citations