A PSO Based on SA for the Sparse Linear Array Synthesis with Multiple Constraints

Pattern synthesis of sparse linear arrays based on the atomic norm minimization and alternating direction method of multipliers approach

A new method of sparse array synthesis based on quantum search algorithm is proposed. The method ensures the constraints include the number of elements, the array aperture and the minimum element spacing can still be satisfied by setting rational parameters of achimedean spiral reasonably when array elements are distributed along archimedean spiral. It turns the optimization problem of sparse plane array into the optimization of asymmetrical linear arrays, which will reduce the complexity of solving problems greatly. The quantum search algorithm (QSA) is a new global optimization methods, which is inspired by quantum mechanics theories, and has excellent global optimization performance. Nowadays, the application of QSA in sparse plane arrays synthesis is few. The method combines archimedean spiral and the QSA, make the sparse plane array produce lower peak sidelobe level (PSLL). The simulated results show that the approach is high efficiency and robust, and can apply to problem of sparse plane antenna array synthesis with the constraint of minimum element spacing and aperture.

Synthesis of sparse arrays is a promising area of research for a wide range of applications including radar and millimeter-wave wireless communication. The design goal of array thinning problems is to reduce the number of elements of an array without significantly affecting its performance. This work presents a technique for synthesizing a sparse phased-array antenna from a 16×16 uniform rectangular array (URA). The proposed approach reduces the number of elements by 50% without any significant increase in the peak sidelobe level (PSLL) for all possible scan angles in the azimuthal and elevation plans within a finite range of scan angles. The synthesis includes an artificial neural network (ANN) model for estimation of the excitation weights of the URA for a given scan-angle. The weights of the sparse array are computed by the Hadamard product of the weight matrix of the URA with a binary matrix that is obtained using particle swarm optimization (PSO) to minimize the PSLL.

In this paper, the synthesis of sparse time-modulated linear arrays with minimum number of elements and controlled harmonic radiations is investigated. The proposed iterative approach based on a particle swarm optimization is aimed at flnding the array conflguration with the minimum number of elements and the optimal pulse sequence that afiords a beam pattern with the same features of a reference one also limiting the amount of sideband radiations under a speciflc threshold. A set of representative numerical examples is discussed to assess the efiectiveness and the reliability of the proposed approach.

The present article proposes a novel hybrid approach for addressing the synthesis problem of rectangular planar arrays under multiple constraints, through joint optimization of the weight function and the improved grey wolf optimization (IGWO) algorithm. Firstly, the grey wolf optimization (GWO) algorithm is improved by using tent chaotic mapping, nonlinear convergence factor, dominant wolf dynamic belief strategy, and opposition‐based learning strategy to increase the population diversity and the ability to jump out of the local optimum. Secondly, the array elements are weighted using the weight function to generate the position distribution matrix, which reduces the thinned matrix optimization time and improves the optimization efficiency. Finally, the position distribution matrix is used to generate the thinned array, and the IGWO algorithm is applied to perform the sparse optimization with multiple constraints. The effectiveness of the method is verified through numerical simulation and full‐wave simulation experiments, demonstrating its capability to enhance array antenna performance and reduce peak sidelobe level (PSLL). These experimental results hold significant engineering implications and provide valuable references for addressing the array distribution problem under multiple constraints.

This article proposes the hybrid Nelder–Mead (NM)–Particle Swarm Optimization (PSO) algorithm based on the NM simplex search method and PSO for the optimization of multimodal functions. The hybrid NM–PSO algorithm is very easy to implement, in practice, since it does not require gradient computation. This hybrid procedure performed the exploration with PSO and the exploitation with the NM simplex search method. In a suite of 17 multi-optima test functions taken from the literature, the computational results via various experimental studies showed that the hybrid NM–PSO approach is superior to the two original search techniques (i.e. NM and PSO) in terms of solution quality and convergence rate. In addition, the presented algorithm is also compared with eight other published methods, such as hybrid genetic algorithm (GA), continuous GA, simulated annealing (SA), and tabu search (TS) by means of a smaller set of test functions. On the whole, the new algorithm is demonstrated to be extremely effective and efficient at locating best-practice optimal solutions for multimodal functions.

This paper presents a novel method for synthesizing sparse dipole arrays with shaped patterns by joint optimization of element positions, rotations and phases. Such joint optimization exploits maximum available degrees-of-freedom under the assumption of uniform amplitude excitations to improve the shaped power pattern synthesis performance, and it leads to a reduced number of elements and simpler feeding network without using unequal power dividers for the array. The copolarization (COP) and cross-polarization patterns for a rotated dipole array are derived in a 2-D case, and then the vectorial shaped pattern synthesis problem is formulated as finding the optimal solution by maximally approaching the desired COP pattern shape under multiple constraints on the sidelobe level (SLL) and cross-polarization level (XPL). In addition, an asymmetric mapping method is incorporated into the proposed joint optimization, so that the minimum element spacing constraint can be also used for the synthesis of planar sparse dipole arrays. The comprehensive learning particle swarm optimization (CLPSO) algorithm is adopted to optimize the element positions, rotation angles and phases. A set of synthesis experiments including linear and planar rotated dipole arrays with a different pattern shape requirements are conducted to validate the effectiveness and robustness of the proposed approach. Synthesis results show that the proposed approach can save about 17.24% – 55.86% elements as well as many unequal power dividers for test cases.

This paper addresses a novel rotationally symmetric technique with multiple constraints for sparse conformal array synthesis. The purpose is to synthesis a sparse optimal common element positions on the conformal surface varying multiple patterns of wide angle scanning with the behavior of low sidelobe levels (SLL). The conformal surface aperture is partitioned into several rotationally symmetric sections. The element positions and element numbers of only one section need to be optimized, which contribute to the reduction of optimizing variables and computation resources. We formulate the synthesis problem as a constrained optimization problem, which takes the peak sidelobe level (PSLL) as the fitness function, and sets the total number of array elements, the minimum spacing between two adjacent elements to form multiple constraints. The Brain Storm Optimization (BSO) is further exploited into the synthesis problem with multiple constraints. A set of representative numerical examples are presented to assess the advantages and effectiveness of the proposed method.

Neural Network system for SoC is one topic in the world. Recently, SoC began to combine AI technique by Python. When developers consider Intelligent system on SoC, the co-design software and hardware are significant. In this case, small devices and embedded system are target system. For IoT, those techniques will grow up in the world. Those cases, avoiding premature convergence while keeping performance is a challenge in training the neural network, especially in a case of large NNs or a large number of training data. In this case, one problem is a huge memory and processing power. For tiny memory and processing system as SoC, we propose an improved particle swarm optimization (PSO) algorithm called the PSOseed2 algorithm for training NN. The PSOseed2 algorithm solves the premature convergence of the standard PSO (SPSO) algorithm by slightly modifying the velocity update function without adding many computational tasks to the SPSO algorithm. We evaluated this algorithm on field programmable gate array (FPGA)-based NN and software-based NN and trained these NNs with different PSO algorithms that are SPSO, PSOseed, PSOseed2, and dissipative PSO. Experimental results with different datasets confirmed that the NNs trained by the proposed PSOseed2 algorithm had better recognition rates and lower global learning errors than the NN trained by other PSO algorithms. In this talk, I would like to talk those relational research in the world and our results and discussion.

For the element position synthesis of 2-dimension sparse rectangular arrays with the design constraints of the number of elements, the aperture and the minimum element spacing, an improved genetic algorithm (IGA) based on chromosome element resetting is presented in this paper. Compared with the synthesis method of thinned arrays, this new approach could exploit more degree of freedom of elements to control the characters of the sparse arrays. When the aperture, the number of elements and the minimum element spacing are fixed identical, the new approach can make the sparse plane array produce lower peak sidelobe level (PSLL). The simulated results show that the fitness function value of thinned plane arrays with 108 elements mentioned by Haupt can reduce 5.626 dB when the minimum element spacing of half wavelength was introduced into the optimization. And as well, the great efficiency and the robustness of IGA are shown in this paper

In view of the problem of premature convergence of traditional particle swarm optimization (PSO) algorithm in path planning, which is easy to converge to local optimal solution and poor path quality, some theory about the corresponding PSO algorithm of simulated annealing optimization is studied in this paper. While planning the moving path of the robot, it analyzes the effect of initial temperature and cooling coefficient on path length and iteration times from the major contributing factors of simulated annealing algorithm. Thus deduce the law of its change and seek the optimal parameter matching. Simulated annealing algorithm can not only move the updated particle position on the basis of the particle swarm optimization formula, but also select the updated position with a certain probability. The method is used to avoid the particle converging into the local optimal solution in the whole iterative process. The capabilities of the global optimization is strengthened. Compared with the traditional PSO algorithm, the simulated annealing PSO in complex environment has better optimization ability, shorter path and fewer iterations in the simulation results.KeywordsParticle swarmPath planningSimulated annealingLinear inertia weight

In this paper, a new scheme of Joint Channel Estimation and Partial Parallel Interference Cancellation using Simulating Annealing (SA) or Particle Swarm Optimization (PSO) algorithms is proposed. This scheme is proposed for multiuser Multiple-Input Multiple-Output /Space Division Multiple Access (MIMO-SDMA) Orthogonal Frequency Division Multiplexing (OFDM) systems. The proposed scheme tries to reduce the cancellation error in Conventional Parallel Interference Cancellation (CPIC) detection schemes used in MIMO SDMA-OFDM system. The SA and PSO algorithms are employed in searching the optimal weights of the PPIC for multiple access interference cancellation. The proposed scheme is shown to provide a performance improvement as compared to MMSE and CPIC detectors especially in an overloaded scenario where number of users is high as compared to the Base Station (BS) antennas. In this scenario, channel estimation becomes more challenging, owing to the increased number of independent transmitter-receiver links to be estimated and the constraint imposed by the rank of the MIMO channel matrix. The proposed scheme provides a reasonable solution to the multiuser MIMO channel estimation problem and multiuser detection in the above-mentioned overloaded scenario.

The pattern synthesis technique of antenna arrays has a wide range of applications in radar and communication systems, which is an important research direction in the field of smart antennas. In order to improve the optimization performance of pattern synthesis of linear antenna arrays, an optimization method based on the Modified RIME Optimization Algorithm (MRIME) is proposed. RIME is a new heuristic algorithm inspired by the condensation process of frost ice in nature, which has a unique update mechanism with strong convergence as well as randomization. In the present work, MRIME is used for optimal pattern synthesis of a linear antenna array (LAA). One part of the present study is to optimize the side lobe level by optimizing the antenna current amplitude while maintaining a uniform spacing; the other part is to optimize the antenna position while assuming a uniform excitation in the synthesis of a sparse LAA, where constraints are imposed on the spacing of the array elements and aperture length, and suppression of the side lobe level with the position of the zeros in the specified direction is also achieved. In this article, simulation experiments for uniform linear arrays as well as sparse linear arrays are presented in detail, and the results show that the MRIME optimization algorithm outperforms most of the existing evolutionary classes of optimization algorithms in optimizing the side lobe level. This illustrates the potential of utilizing the MRIME optimization algorithm for antenna arrays and various other electromagnetism-related challenges.

Designing sparse arrays with fewer antenna elements and lower peak sidelobe level (PSLL) is critical and challenging for antenna arrays in terminal devices in Internet of Things (IoT) systems. This article presents a deep learning method based on denoising diffusion probabilistic models (DDPM) for synthesizing sparse arrays. With predefined constraints such as aperture size, number of elements, and minimum element spacing, a training dataset that enables the proposed DDPM is generated to directly learn the distribution characteristics of these sparse arrays. To simplify the network architecture, the backbone network U-Net is replaced by a multilayer perceptron (MLP). In addition, an innovative conditional generation strategy is introduced to enhance the training accuracy, categorizing the training data into diverse cross-conditional labels based on PSLL and sparse ratio. During the training process, two separate regression networks are employed to predict the sparse ratio and the PSLL, and a Pareto-optimal weighting scheme (PWS) is designed to address the trade-off between them. Examples of two planar sparse arrays and a conformal sparse array demonstrate the effectiveness of the proposed DDPM-based method, highlighting its superiority in reducing PSLL and increasing sparse ratio, reducing inference time compared with the existing methods.

To solve the strong randomicity and slow convergence of the Particle Swarm Optimization(PSO) algorithms, two new particle’s position renewal formulas were analyzed on the basis of extrapolation in mathematics. A new modified PSO algorithms (called Leading PSO algorithms) was put forward. The Direct Torque Control(DTC) System was built in the environment of Matlab(Simulink). The weight and threshold values of BP Neural Network were trained using the modified PSO algorithms. Some disadvantages such as slow convergence speed and easily plunging into the local solution were avoided effectively. The simulation result shows that the system works well, and the rotor speed identifier has great static and dynamic performance.