具有衬底集成波导和微带谐振器的混合非均匀Q损耗滤波器资源-CSDN文库

80 浏览量 2021-02-25 00:35:32 上传评论收藏 1.83MB PDF 举报

混合非均匀Q损耗滤波器是一种新型的射频/微波电路的被动组件，具有重要的研究意义和应用价值。这种滤波器的设计方法主要基于非均匀Q方法，其工作原理是通过构建一个高Q值的衬底集成波导(SIW)耦合路径和一个低Q值的微带耦合路径来实现。衬底集成波导(SIW)是一种新型的平面波导技术，它通过在介质基板上刻蚀金属化孔，实现对电磁波的引导。这种方法可以有效地保持波导的高Q值特性，同时实现平面化的结构设计，大大减小了设备体积，且易于与微带技术集成。因此，在混合非均匀Q损耗滤波器中，SIW被用作高Q值的耦合路径，以提升滤波器的性能。微带谐振器是一种常见的微波频率选择元件，通常由介质基板上的金属导带构成。它的特点是尺寸较小，成本低，易于集成，因此在微波集成电路中广泛应用。在混合非均匀Q损耗滤波器中，微带谐振器作为低Q值的耦合路径，有助于降低滤波器的成本和复杂度。非均匀Q方法是指通过调整滤波器中各个谐振器的Q值，以实现特定的频率选择特性和带宽。这种设计方法可以有效地提升滤波器的性能，但同时也增加了设计的复杂性。在混合非均匀Q损耗滤波器中，通过适当选择低Q值和高Q值的比率，可以获得较宽的0.2dB带宽。文章中提到的两种调整Q值比率的技术，一种是通过在每个微带谐振器上安装芯片电阻来降低其未加载Q值，从而降低滤波器的整体Q值比率。这种方法可以显著地增加滤波器的带宽，但是会增加滤波器的插入损耗。另一种方法是通过调整微带谐振器的辐射损耗来改变Q值，这主要是通过改变输入/输出馈电结构来实现。通过这种方法，可以获得更宽的带宽，但也会带来一定程度的插入损耗的增加。此外，文章还提到，通过优化设计，可以减小混合平面滤波器的占用面积，同时使通带的平坦度远远优于均匀Q滤波器。这通过模拟和实测结果得到了验证。减小占用面积可以使得滤波器更加紧凑，有助于集成度的提高和成本的降低。总体而言，混合非均匀Q损耗滤波器通过集成不同Q值的传输路径，结合不同调整Q值的技术，为射频/微波电路提供了性能优越、体积小、成本低的滤波器解决方案。这种滤波器的提出和研究，对于提高通信设备的性能和降低成本有着重要的意义。随着无线通信技术的快速发展，这种新型滤波器的研究和应用将会更加广泛。

资源推荐

资源详情

资源评论

IET Microwaves, Antennas & Propagation

Research Article

Hybrid non-uniform-Q lossy filters with

substrate integrated waveguide and

microstrip resonators

ISSN 1751-8725

Received on 23rd January 2017

Revised 21st August 2017

Accepted on 13th September 2017

doi: 10.1049/iet-map.2017.0060

www.ietdl.org

Liang-Feng Qiu

, Lin-Sheng Wu

, Wen-Yan Yin

1,2

, Jun-Fa Mao

Key Laboratory of Ministry of Education of Design and Electromagnetic Compatibility of High-Speed Electronic Systems, Shanghai Jiao Tong

University, Shanghai, People's Republic of China

Innovative Institute of Electromagnetic Information and Electronic Integration, College of Information Science and Electronic Engineering,

Zhejiang University, Hangzhou, People's Republic of China

E-mail: wallish@sjtu.edu.cn

Abstract: Three lossy filters with flat passband are proposed with planar hybrid configurations, designed with the non-uniform-Q

method. The hybrid filters are built up with one higher-Q substrate integrated waveguide (SIW) coupling path and the other

lower-Q microstrip coupling path. By properly selecting the lower-Q-to-higher-Q ratio, the 0.2 dB bandwidth of the first filter is

obtained to be 215 MHz (17% improvement), which is equivalent to a uniform unloaded Q-factor of 820 at the cost of additional

in-band insertion loss (IL) of 0.5 dB, when compared with the full-SIW filters. Further, two techniques are utilised to adjust the Q-

factor ratio. One method is implemented by mounting a chip resistor on each microstrip resonator to reduce its unloaded Q-

factor and lower-Q-to-higher-Q ratio of filter. Then, the 0.2 dB passband of 243 MHz is achieved for the second filter, equivalent

to a uniform unloaded Q-factor of 1420. The other method is realised by adjusting the Q-factor due to radiation loss of microstrip

resonators. By changing the input/output feeding structures, the 0.2 dB bandwidth of the third filter is enhanced by 32%, at the

cost of minimum IL increased by 0.9 dB. The occupied area of the proposed planar hybrid filters is reduced by 15–21% and the

passband flatness is much better than that of the uniform-Q filter counterparts, as demonstrated by the simulated and measured

results.

1 Introduction

Bandpass filters are important passive components of radio-

frequency/microwave circuits and systems. Modern filters are

oriented towards low cost, low weight and small size with high

performance. The design of high-performance miniaturised

bandpass filters is being a challenge, usually due to their low

unloaded Q-factors provided by compact resonators. Low-Q filters

suffer relatively high in-band insertion loss (IL), degraded

passband flatness and frequency selectivity of band edges,

especially for narrow bandpass filters [1]. Hence, in order to realise

high performance, the dissipation of the resonators cannot be

ignored. Recently, some techniques have been proposed for filter

design with the consideration of losses (called lossy filter),

including predistortion [1, 2], lossy coupling matrix [1, 3] and non-

uniform Q methods [4–10]. The predistorted filters achieve a good

passband flatness by reflecting more power in the passband. Thus,

the return loss (RL) of the predistorted filters is significantly

deteriorated. As for the other two kinds of lossy filters, they have

improved frequency selectivity and passband flatness by absorbing

power rather than reflection. Thus, the performance is improved at

the cost of introducing additional in-band IL. These methods can

be applied in the applications where the in-band amplitude flatness

is highly required while the additional IL can be tolerated, such as

in a satellite IMUX and alternative receiver architectures [4]. Also,

they can be used in the design of receive filters where the

additional IL can be compensated by pre-amplifiers that would

have little influence on the overall noise figure of the whole system

[1].

The concept of non-uniform-Q filter is first introduced in [4]

also with resistive couplings, which is difficult to apply for higher-

order filters. A new method with non-uniform-Q dissipation based

on the synthesis technique of predistorted reflection-mode filters is

then proposed by the same group [5], where resistive nodes are

required. A simplified method based on the utilisation of resonators

with different prescribed unloaded Q-factors only is presented in

[6], and it can easily be implemented since no additional resistive

coupling is needed. Using this method, dual-mode [7] and triple-

mode [8] microstrip resonators are exploited to design third-order

non-uniform-Q filters for miniaturisation. In these filters, the

unloaded Q-factor of microstrip resonator was tuned by loading

resistors. However, the parasitic parameters of lumped resistor,

especially the parasitic capacitance, may impact on the

performance of filter and the design process. The position of the

attached resistors should be carefully selected to reduce the

unwanted effects. However, when the operating frequency of lossy

filter is relatively high, the resistor-loading method cannot be used

anymore.

Substrate integrated waveguide (SIW), a new group of low-cost

integrated guided wave structures, shows many advantages, such as

high Q, low loss, high isolation, low leakage and high power

handling, when compared with other planar transmission lines [11,

12]. Recently, SIW filters have attracted much attention [13–20]. It

is noted that SIW resonator can provide relatively high unloaded

Q-factor with moderate size while microstrip resonator is usually

with relatively low unloaded Q-factor but small size. When

combining microstrip and SIW structures within one filter, more

flexible performance can be provided with compactness [20, 21]. It

is also feasible to realise a non-uniform-Q filter with hybrid

structures of SIW and microstrip line.

In this paper, the fourth-order hybrid non-uniform-Q lossy

filters with SIW and microstrip elements are proposed. It is

constructed by two signal paths, one with two higher-Q SIW

cavities and the other with two lower-Q microstrip resonators. The

impact of the lower-Q-to-higher-Q ratio on the bandwidth (BW)

and IL of the filter is analysed. The influence of additional phases

introduced by the T-junctions which connect the two signal paths

of the filter on the passband is also studied and it is proved that the

location of the T-junction can be designed with freedom. Further,

two techniques are utilised to adjust the Q-ratio of the filter. One

method is realised by mounting resistors on the microstrip

resonators, and the other method is implemented by tuning the

radiation loss of the microstrip resonators. So, the resistors and

their unwanted parasitic effect are avoided. Finally, three filter

prototypes are developed and fabricated together with a full-

IET Microw. Antennas Propag.

microstrip and a full-SIW filter counterpart for comparison. The

measured results demonstrate that the hybrid filters have much

flatter passband.

2 Analysis and design

2.1 Non-uniform-Q method

In a transversal coupled-resonator filter, the unloaded Q-factor of

each resonator has an almost independent influence on the response

around its resonant frequency. Therefore, the resonators whose

resonant frequencies are close to its band edges play more

important roles on the frequency selectivity and passband flatness

[6]. It is easy to transform the transversal topology of a fourth-

order filter with a symmetrical response to the scheme of parallel

coupled pairs, as shown in Fig. 1. When the pair of resonators

corresponding to the band edges are built up with higher-Q

structures and the other pair of resonators related to the middle

passband are realised by lower-Q structures, small in-band

variation and sharp edges can be obtained. The synthesis procedure

of the non-uniform Q method for an Nth order filter is presented in

[6, 22].

In order to achieve a filtering response with improved passband

flatness, the first-order approximation of the |S

| variations due to

the dissipation factors, i.e. δ

, introduced by finite unloaded Q-

factors at the reflection zeros of ideal lossless response should

satisfy the following condition:

Δ S

= P

= d

⋮

(1)

where δ

= δ

and δ

= δ

, d is the uniform deviation of

transmission coefficient and P is the sensitivity matrix. The

element of P(i, j) is defined as the derivative of |S

| with respect to

at the jth reflection zero. There are four equations and five

unknowns, i.e. δ

to δ

and d. The other four unknowns can be

solved when the maximum or minimum dissipation factor is pre-

determined.

From (1), it is derived that the ratio of Q

) to Q

) is

∑

k = 1

2, k

−1

∑

k = 1

1, k

−1

(2)

It shows that the ratio is independent from d if the first-order

approximation in (1) is applied. Thus, a flat passband of filter can

be obtained when the unloaded Q-factors of all resonators are

selected proportionally.

2.2 Influence of the ratio of lower-Q to higher-Q

A filter prototype of a fourth-order generalised Chebyshev

response centred at 5 GHz with a fractional BW (FBW) of 5% is

designed as an example. Its in-band RL is better than 20 dB and a

pair of transmission zeros are located at ±2j. Its coupling matrix

as the parallel connection of two folded topologies is given by

0 0.5248 0.8788 0 0 0

0.5248 −jδ

0 0 1.278 0

0.8788 0 −jδ

−0.6904 0 0

0 0 −0.6904 −jδ

0 0.8788

0 1.278 0 0 −jδ

0.5248

0 0 0 0.8788 0.5248 0

(3)

where the real part of the coupling matrix is synthesised according

to the method in [2] while the imaginary part of the matrix can be

obtained from (1) when the maximum or minimum unloaded Q-

factor and FBW are given. The ratio of lower-Q to higher-Q,

denoted as k = Q

= δ

/δ

, is synthesised to 0.27 to achieve an

improved passband flatness.

The impacts of ratio k on the relative incremental BW and the

incremental IL with respect to the uniform case, i.e. k = 1, are

depicted in Figs. 2a and b, respectively. As shown in Fig. 2a, the

passband flatness, which is in terms of the BW with 0.2 dB

variation (0.2 dB BW), becomes better when k decreases from 1.

However, the passband IL increases at the same time, as shown in

Fig. 2b. Hence, there is a tradeoff between the IL and the passband

flatness. The lower-Q-to-higher-Q ratio k can be selected according

to the design specifications of IL and passband flatness

simultaneously. Furthermore, it is seen that the maximum

improvement of 0.2 dB BW decreases with the increase of Q

. The

non-uniform-Q method provides little improvement (<5%) in the

0.2 dB BW when Q

> 2500 while the incremental IL is still

observable. For a given value of k, the higher the Q

is, the lower

the in-band IL is.

2.3 Consideration for T-junctions

As illustrated in Fig. 1, the higher-Q and lower-Q signal paths in

the parallel folded topology are connected by two T-junctions,

usually in the form of transmission lines. It is noted that the

practical T-junction, e.g. a microstrip T-junction, has additional

phases and losses in comparison with the ideal one, due to the

parasitic effects of discontinuity. The additional phases, denoted by

Fig. 1 Coupling scheme of parallel coupled pairs

Fig. 2 Responses as functions of k with respect to the uniform case (k = 1)

(a) Relative incremental BW, (b) Incremental IL

2 IET Microw. Antennas Propag.

剩余6页未读，继续阅读

评论收藏

内容反馈

weixin_38614112

粉丝: 3
资源: 929

具有衬底集成波导和微带谐振器的混合非均匀Q损耗滤波器

具有改善的通带平坦度的有损衬底集成波导（SIW）滤波器

具有高阶模式抑制功能的多层衬底集成波导（SIW）滤波器

基于衬底集成波导三分之一等边三角形谐振器的紧凑型带通滤波器

高品质因子干法蚀刻铌酸锂脊形波导微环谐振器

Lumerical编程,lumerical教程

微环谐振器

半导体激光器工作原理及基本结构PPT学习教案.pptx

一种内反射耦合的单纵模半导体激光器模型

半导体激光器工作原理及基本结构10150PPT学习教案.pptx

平面工艺制作的定向输出微腔半导体激光器

单片光电子集成技术-综合文档

显示/光电技术中的单片光电子集成技术

光纤通信原理 自考试卷

基于CORDIC的反正弦和反余弦计算的FPGA实现

BA无标度网络中的SIR模型

使用3DCNN和卷积LSTM进行手势识别学习时空特征

基于三次贝塞尔曲线的类汽车曲率连续路径平滑

基于机器学习的设备剩余寿命预测方法综述

基于维纳过程的退化模型，具有递归过滤算法，可用于估计剩余使用寿命

基于FPGA的奇异值和特征值分解的快速实现。

磁悬浮系统自适应模糊PID控制器的设计

基于BP神经网络的人口预测

两轮平衡车的建模与控制研究

无人机协同目标的多无人机协同搜索方法

基于改进遗传算法的六自由度机器人时间最优轨迹规划

一种基于深度学习的机械臂抓取方法

基于深度神经网络的交通流量预测

最新资源

光纤通信原理自考试卷