Research On Large Scale Sparse Array Layout And Beam Synthesis New Technique

Compared with the uniformly spaced array under the same aperture,sparse array can reduce the number of antenna elements while obtaining the equivalent even better pattern characteristics by using the potential degree of freedom of element position,which can reduce the cost,space and weight of the radiation system.Moreover,it can suppress the grating lobe by the position optimization,which is helpful for the design of wideband or ultra-wideband(UWB)array antennas.These advantages make sparse arrays attract widely attention in modern radar,sonar,wireless communication and radio astronomy.Sparse array with non-uniform amplitude weighting is widely used for achieving a complicated focused or shaped beam pattern.However,the feeding network of antenna system is required to equip with unequal power dividers,which increases the hardware complexity and system cost.Thus,this paper will consider the synthesis of sparse arrays with uniform amplitude.Firstly,the element rotation is exploited as a particular degree of freedom,and the element position and excitation phase are also combined to realize the synthesis of sparse arrays with focused/shaped beam pattern.Secondly,the design method of large-scale sparse planar array layout from the perspectives of stochastic optimization and iterative Fourier transform technique is studied in this paper.The main research of this paper is summarized as follows:1.Synthesis of rotated sparse dipole arrays with shaped pattern.Firstly,the vectorial shaped pattern expression for a rotated dipole array is derived in a two-dimensional case,and then an asymmetric mapping method is introduced to constrain the minimum element spacing of sparse planar array.Finally,a comprehensive learning particle swarm optimization(CLPSO)is adopted to jointly optimize the element position,rotation angle and excitation phase,and the synthesis of dipole arrays with different pattern shape requirements is achieved.This joint optimization strategy exploits the particular degree of freedom of element rotation under the condition of uniform amplitude to improve the array pattern performance.Compared with the traditional amplitude weighting methods,it avoids the usage of unequal power dividers in the feed network.2.Synthesis of sparse arrays with focused or shaped beam pattern based on element rotation considering mutual coupling.For rotated sparse arrays considering antenna element structure and mutual coupling,a refined strategy based on the combination of CLPSO algorithm and full-wave simulation is proposed.Since the array element position and rotation angle need to be optimized,the approximated active element pattern is used to the pattern synthesis,but there exists some deviations between the synthesized pattern and the real array pattern.To reduce these deviations to a negligible degree,synthesis and simulation are required to conduct several times,and the allowable range for element positions and rotation angles is required to gradually reduce as the number of the synthesis increases.Meanwhile,to meet the minimum element spacing constraint,an improved asymmetric mapping method is proposed to transform the array element positions into some variable matrices.Then CLPSO algorithm is used to optimize these matrices together with rotation angles and/or phases.According to the optimized position and rotation angle,full-wave simulation is carried out to obtain the active element pattern of each element.As the number of refinements increases,the deviation becomes smaller and smaller.This method is suitable for the synthesis of sparse linear and planar arrays with arbitrary antenna element structure considering mutual coupling.3.Synthesis of large scale sparse planar arrays based on stochastic optimization method.A differential evolution algorithm with new encoding mechanism and Cauchy mutation is proposed to optimize large scale sparse array layout with minimum element spacing constraint.In the new coding mechanism,each individual represents an array element position,rather than the entire array layout in the traditional stochastic optimization algorithm.In each individual updating,only the radiation contribution variation caused by one element movement is taken into account,which greatly reduces the calculation time.Moreover,when the population is updated,the resulting array layout can easily meet the minimum element spacing constraint.As the array size increases,each individual is always searching in two dimensions.In addition,the differential evolution algorithm is used as search engine,a Cauchy mutation based on chaotic mapping is proposed to enhance local search ability while maintaining population diversity.This method is more efficient than the traditional stochastic optimization algorithm,especially in the case of synthesizing large sparse planar array layout.4.Large scale UWB sparse planar array synthesis method based on iterative Fourier transform(IFT)framework.A cooperative coevolutionary iterative Fourier transform technique is proposed to efficiently optimize the UWB sparse planar array layout.Firstly,the element position grid with half a wavelength at the highest frequency is adopted to avoid the grating lobes.Then,a probabilistic-assisted element selection mechanism integrated with a minimum element spacing control is proposed to prevent premature convergence to some extent while enabling the obtained array to meet the minimum element spacing constraint.Meanwhile,inspired by the idea of population evolution,a cooperative coevolutionary strategy is presented to further prevent the premature convergence.In this strategy,an individual represents an excitation distribution,and the information interaction between different individuals is established to guide the selection of the element position distribution.In addition,simulated annealing mechanism is embedded into the cooperative coevolutionary process to maintain the diversity of the population.A set of examples for synthesizing large UWB planar arrays with different sizes and bandwidths demonstrate the effectiveness and superiority of the proposed method.