Parametric design method of Tx and Rx filter in bulk acoustic wave duplexer
-
摘要: Tx与Rx滤波器,作为体声波(BAW)双工器的关键部件,其中各薄膜体声波谐振器(FBAR)单元的谐振区面积会影响滤波器的插入损耗和带外抑制这两个重要指标。由于这两个指标是相互制约的,在设计时往往需要折衷考虑,这使得各FBAR单元的谐振区面积设计成为BAW双工器设计的难点之一。为了解决这一问题,提出了一种Tx与Rx滤波器的参数化设计方法。在ADS软件中以FBAR的Mason模型为基础构建了梯形拓扑结构的滤波器电路模型。设置其中各串联FBAR的谐振区面积值以及串并联FBAR的谐振区面积比值为两类优化参数。以给定的滤波器插入损耗与带外抑制为优化目标,使用ADS软件中基于梯度的优化算法得到各参数的优化值。根据优化结果可以简单地计算得到滤波器中各FBAR单元的谐振区面积。以一个工作在LTE band 7的BAW双工器为设计案例展示了该方法的应用流程。设计结果的仿真验证表明:Tx滤波器的插损小于2 dB、在Rx滤波器通带内的抑制大于40 dB;Rx滤波器的插损小于1.9 dB、在Tx滤波器通带内的抑制大于40 dB;满足BAW双工器中Tx与Rx滤波器的性能指标。由此验证了该方法的可行性和易用性。Abstract: In design of the bulk acoustic wave (BAW) duplexer, the resonance area of each thin film bulk acoustic wave resonator (FBAR) in Tx or Rx filters affects its performance of insertion loss and out of band rejection. However, these two properties of the filter in design often need to compromise, which brings difficulty for calculating the resonance area of each FBAR in filter. In order to solve this problem, a new parametric design method of Tx or Rx filter is proposed. The circuit model of filter was established by the foundation of FBAR Mason model in ADS software. The resonance area value of series FBAR and the ratio of resonance area value of parallel FBAR to series FBAR were made into two types of optimization parameters reasonably. According to the given insertion loss and out of band rejection of filter as the optimization objective, the optimized values were obtained by the algorithm based on gradient in ADS software. To design BAW duplexer used in LTE band 7 for example, simulation results show that the insertion loss of the Tx and Rx filter using this method is less than 2 dB and the rejection in passband of each other filter is greater than 40 dB, which meets the performance indicators of Tx and Rx filters and verifies the feasibility and ease of use of the method.
-
Key words:
- bulk acoustic wave /
- duplexer /
- filter /
- parametric design /
- Mason model /
- resonance area
点击查看大图
计量
- 文章访问数: 1283
- HTML全文浏览量: 214
- PDF下载量: 464
- 被引次数: 0