Research On Robust Evolved Antenna

As a key part of radio equipment, antenna is widely used in the fields of communication, radar and etc. Conventional antenna design methods rely on the designers’knowledge of electromagnetic theory and experiences, and it’s difficult to find a solution in a short time while facing with challenging requirements. As a newer antenna design method, evolutionary antenna design method is now widely acknowledged. Evolutionary antenna design is a system project, is the product of interdiscipline, so that its development is related with several other fields. Evolutionary antenna design mainly relies on the fields of evolutionary algorithms, electromagnetic simulation software, antenna design, computer hardware and distributed computation. The field of evolutionary antenna design has grown in recent years as evolutionary algorithms have improved, computer speed has increased and electromagnetics simulators have improved. It is now possible to do large scale simulation through the evolutionary way. Though the evolutionary antenna is related with many other fields, it develops rapidly with its branches’ improvements. Evolutionary antenna design method is an important and widely used method now.In the field of evolutionary antenna, there are researches on evolutionary algorithms, antenna problem modeling and antenna structure design. The evolved antennas show significant advantages, but it is an aided design method for most designers. There is no systematic knowledge and technology, and a lot of problems are waiting to be solved in the process of design and production for the applications in a real project. The problem modeling, antenna structure design and algorithm improvement are the main research subjects for evolutionary antenna. The wire antenna is the most researched antenna type. The research results are mainly tested by simulation, only a few of them are built and measured, and there are several evolved antennas applied in large scale projects. It is hardly to find a paper about the robustness of the evolved antenna, and there are some papers considered it with a simple scheme. Evolutionary antenna design, as an import approach, is restricted with high cost due to the accuracy of manufacture, and this paper proposes a solution for the robustness of evolved antenna. At the begging of the evolutionary design, we should think about how to improve the robustness of the antenna. When the accuracy of manufacture is known, the robustness can be ensured for it using the robust process. The robust evolved antenna is not only good for reducing the accuracy of manufacture, but also making the antenna structure more stable for distortions. The scheme of robust evolved antenna reduces the cost of antenna design, production and even use by improving the robustness of the antenna, and this makes evolved antenna getting more competitive advantages.Evolutionary antenna design method is a way to design antenna automatically using computers, so that the evolutionary antenna design software is a base for all the works. This paper designed and implemented evolutionary antenna design software with high expansible framework which can handle with different algorithm, problem and simulation interface modules. The modules include simulation software interface modules, algorithm modules, problem modules, robustness module and distributed task manager module. The simulation software which provided by the third parties does the antenna electromagnetic simulation work for the evolutionary antenna design software. The simulation interfaces gives the ability to access the simulation software. Evolutionary algorithms are the optimization engines. The problem modules describe the antenna problems in programming code. The robustness module provides the antenna design with robustness attributes. Message Passing Interface is the low-level layer for mpi4py, and we design the distributed task manager module on mpi4py. We can start the computation task of robust antenna design while the project is deployed properly.In the design of evolved antenna, computer searches the solution spaces automatically to find the most optimized result. When a solution is found, without considering the limitation of simulation software, its performance is corresponding to the antenna. We will meet the robustness problem when we are going to manufacture the evolved antenna for real applications. Here, the orthogonal design is applied to evolutionary antenna design. The orthogonal design is a way to sample uniformly around the solution space, and the uniformity makes the samples are able to present the robustness of the solution. During the process of evolution, this paper uses the orthogonal design to simulate the error in real world’s production. The robustness of a solution will be key condition for selection, and this make it is able to design robust evolved antennas.2.4GHz band is an open band, and the most common used wireless local area network is working on2.4GHz. This paper designed a robust evolved antenna and made a real object of it for the requirement of2.4GHz omnidirectional antenna. As a robust evolved antenna, its simulation performance meets the requirements perfectly. From2.4GHz to2.483GHz, its return loss is less than-15dB, and the min gain is better than-0.3dBi. In order to test the robustness of the evolved antenna, this paper tried5000simulation experiments. These experiments simulate the antennas in production environment by adding random errors to the robust variables of the antennas. The results of the experiments showed that all the tested antennas have the return loss less than-15dB and min gain better than-0.3dBi. The simulation test prove the design is robust. The measured return losses at2.4GHz and2.483GHz are similar to the simulated return losses, the max error is less than5%. The measured data proved the usability of the robust evolved antenna. Using the-10dB standard, the simulation data of the antenna has a band width of600MHz (2.1GHz-2.7GHz), and the measured data of the antenna has a band width of800MHz (2.1GHz-2.9GHz). The measured band width is better than the simulated band width. This paper designed and implemented evolutionary antenna design software, applied the orthogonal design to robustness design of evolved antenna, and finally built the material object of the robust antenna. The simulated performance meets the requirements very well, the robustness test improves the robustness of the evolved antenna, and the measured performance shows that it is acceptable for real applications.