1.Nanjing Research Institute of Electronic Technology, Nanjing 210039, China 2.National Key Laboratory of Antennas and Microwave Technology, Nanjing 210013, China 3.Institute of Applied Physics, University of Electronic Science and Technology of China, Chengdu 610054, China
Abstract:The resolution of traditional far-field imaging system is generally restricted by half of the wavelength of incident light due to the diffraction limit. The more specific reason is that evanescent waves carrying sub-wavelength information cannot propagate in the far field and make no contribution to the imaging. However, higher imaging resolution is required in practical applications. To realize the far-field super-resolution imaging, the imaging system should be able to collect both propagating waves and evanescent waves. Many designs have been proposed to solve this issue. In 2007, a far-field superlens was proposed by Liu et al. (Liu Z W, Durant S, Lee H, Pikus Y, Fang N, Xiong Y, Sun C, Zhang X 2007 Nano Lett. 7 403) to realize far-field super-resolution in optical range, which consisted of a silver film and a nanoscale grating coupler. The silver film was used to amplify the evanescent waves, which were then converted into propagating waves by the sub-wavelength gratings. However, the special material properties limit the freedom of design. In microwave band, the incident components can be converted into Bloch modes by the resonant metalens, which consists of subwavelength resonators, and then be radiated to far field. Nevertheless, Green function between antenna and target is necessary, which is difficult to obtain due to the complex and even time-dependent imaging environment in practical applications, especially for super-resolution imaging system. It has been demonstrated in recent research that frequency information can be associated with spatial information of imaging target by localization resonant modes. Therefore, super-resolution imaging can be realized based on frequency information, without using Green function. Thus, a novel microstructure array is proposed to realize the far-field super-resolution scanning imaging based on a fractal resonator. The fractal resonator can work at several frequencies because of the self-similarity, which provides higher selectivity according to practical conditions. Several working statuses can be obtained for the resonator by adding photoconductive semiconductor switches, which are controlled by laser. On account of localization mode resonance, the array can realize the conversion between evanescent waves and propagating waves. Then with the help of antennas in the far-field to receive the frequency information, the location of imaging source can be confirmed according to the spectrum. Then by using the magnitude of resonant peak, sub-wavelength image can be reconstructed without using Green function. To verify the super-resolution scanning imaging characteristics of the array, an imaging simulation of “laugh face”-shaped target is performed. The image is reconstructed very well and the resolution determined by the period of the array is 20 mm, corresponding to λ/10. In view of the particularity of proposed fractal resonator, a novel scanning method is proposed. By combining the first and the third resonance, the imaging efficiency can be well improved compared with by the traditional point-by-point scanning method. Keywords:far-field/ super-resolution scanning imaging/ fractal resonator/ imaging efficiency
值得注意的是, 文献[25]将单个谐振器单元作为一个成像像素点, 本文则不同, 提出的分形单元具备更多的工作状态, 具体可以细分为多个像素点. 接下来对一阶和三阶谐振进行组合使用, 根据谐振器的谐振情况将该分形单元划分为5个区域, 如图1(b)所示, 分别对应1个一阶谐振和4个三阶谐振. 这样三阶谐振就将整个分形谐振器细分为4个区域, 具有了4个像素点. 对于该4个像素点区域, 打开其中1个区域的开关, 关闭其他开关, 通过开关通断得到的频谱即可判断该区域是否存在激励源, 成像像素得到细化, 提升了成像分辨率. 为了避免相邻像素点区域对远场频谱的影响, 须要分析单元之间的耦合问题, 模型结构如图3所示. 此时, 一个激励源位于单元1的近场区域, 初始状态时两个单元都处于非工作状态, 在远场频谱中不会观察到谐振现象. 当处于状态1时, 单元2的5个开关闭合, 断开单元1的5个开关, 此时单元1处于工作状态, 在远场频谱中会观察到明显的谐振现象, 如图4黑色曲线所示, 谐振频率位于1.47 GHz; 当处于状态2时, 单元1的5个开关闭合, 断开单元2的5个开关, 此时单元2处于工作状态, 可以看到此时的远场频谱远小于状态1时, 可以判断单元1和2之间的耦合可以忽略不计, 如图4红色曲线所示. 图 3 一阶谐振的两种工作状态设置 Figure3. Two working status setup of the first resonance.
图 4 一阶谐振两种工作状态的远场频谱 Figure4. Far-field spectra of the first resonance at two working status.
然后对三阶谐振进行类似的分析, 模型结构如图5(a)所示, 得到的结果与一阶谐振类似, 相邻像素点区域之间的耦合可以忽略不计. 图 5 三阶谐振的两种工作状态设置及远场频谱 (a)工作状态设置; (b)远场频谱 Figure5. Two working status and far-field spectra of the third resonance: (a) Working status setup; (b) far-field spectra.
图 1 (a)分形谐振器单元的物理尺寸; (b)开关设置模型
图 2 分形谐振器单元远场频谱仿真模型及两种工作状态下的远场频谱 (a)远场频谱仿真模型; (b)两种工作状态下的远场频谱
图 3 一阶谐振的两种工作状态设置
图 4 一阶谐振两种工作状态的远场频谱
图 5 三阶谐振的两种工作状态设置及远场频谱 (a)工作状态设置; (b)远场频谱
图 6 分形微结构阵列编号及成像模型设置 (a)编号模型; (b)成像模型
图 7 分形微结构阵列远场超分辨率扫描成像结果 (a) 8 × 8像素点成像结果; (b) 16 × 16像素点成像结果