I have significant expertise in the design of all type of phase shifters, from very low frequencies all the way up to mm-wave frequencies. 

Fig. 1 Practical hybrid varactor-PIN diode phase shifter (a) and its circuit diagram (b)

Chia sẻ0

Tweet

Chia sẻ

My work on amplifiers goes all the way back to 2002. My work on amplifiers included discrete transistor characterisation (Volterra and Taylor series), design of Low Noise Amplifiers (LNAs) and the design efficiency enhancement amplifiers, such as Doherty amplifiers.

Chia sẻ0

Tweet

Chia sẻ

Of particular importance to modern filter design is not only excellent electrical performance, but also size and volume. This is principally true for filters operating at the lower end of the frequency spectrum (e.g., 700 MHz), where their physical volume and weight pose significant challenges to network equipment manufactures. In this regard, the need for excellent electrical performance inevitably increases the filter size. Consequently, RF filters tend to occupy a significant volume of a number of communication devices. In particular, cavity filters are still the mainstay in mobile cellular communication base stations, by virtue of their power-handling capabilities, cost effectiveness, good electrical performance (medium to high quality factor) and technological maturity. However, the attractive features of cavity filters are counterbalanced by an increased physical size and, equally importantly, weight. The bulky size can be alleviated at the expense of reduced electrical performance. For example, capacitive loading and a stepped resonant post are often deployed to reduce resonator profile, albeit at the expense of performance. Helical resonators can also be used to address the issue of bulky size. 

The frequency range of filters that my work focuses on is from 1 GHz to 150 GHz. Of particular importance to the filter design of my work is miniaturization with little or no impact on performance. As evidenced in my broad patent and publication portfolio, I have contributed to the knowledge and applications of a wide range of filters – from cavity to ceramics.  

Chia sẻ0

Tweet

Chia sẻ

The designed filter is shown in Figure 2. The total internal dimensions of the filter are 110 x 41 x 52 mm³. The filter has 3 cross couplers necessary to introduce 4 transmission zeroes – 2 on the lower skirt and the remaining two on the upper skirt of the response, Figure 3.

Highlights:

1. The correspondence between the simulated and measured filter responses is excellent.
2. Both simulated and fabricated filters meet all specifications of band 42, LTE standard.
3. As an example, the return loss is no less than 18.2 dB and insertion losses are kept well below 1.25 dB at the band edges.
4. Please contact us for more details.

Chia sẻ0

Tweet

Chia sẻ

Heinrich Hertz, 1887: First spark generator at Karlsruhe High-School for Technology

A century forward – base station technology in the digital age

Motivation – Modern day

1. Driven by size reduction – in base stations operating at lower GHz, size of diplexers over 70 % of the entire transceiver. 
2. Driven by cost – ceramic based filters have performance edge over cavity solutions, but it comes at cost. 
3. Driven by performance – assuming a base station with a power output of 50 dBm (100 W), losing 1 dB infers losing 20W. 
4. Frequency reconfigurability and tunability are becoming of greater importance as new bands are opening up, such as FR 2 (5G) and the likely need in 6G. 
5. Reconfigurability still in infancy, with great deal of effort dedicated to bulk-tuneable materials (Liquid Crystals, Ferro-electrics, Electro-chromic materials and Transition Metal Oxides)          

Distributed Resonator – Basic Idea with two resonators

Distributed Resonator – with any number of individual resonators

Resonance distribution can be extended to a large number of individual elements for reduced frequency of operation.

Distributed Resonator – Operation

Operating frequency can be reduced by folding individual resonant elements

The frequency of operation of the folded structure is always lower than the inline structure

Distributed Resonator – 3-Pole filter Results

In order to test the theory, a 3-pole filter operating at 1800 MHz with a bandwidth of 40 MHz was designed and fabricated

Each resonant chamber consists of 25 high frequency resonant elements – the tuning screw behaves as one of the resonant elements – unloaded Q of  1900.

Split-Distributed Resonator

-A more uniform distribution of electric field intensity among individual resonant elements should result in better electrical performance and better power handling

-Individual resonant elements are made in a distributed element fashion themselves

-Individual pairs couple among themselves (inter) and inside (intra) each other

-Coupling along the downward, left right diagonal of element pairs is achieved

-Resonant matrix can be arbitrarily extended

Split-Distributed Filter (5-pole)

-Each cylindrical element is split into 4 separate elements to obtain better distribution of electric field.

-Individual chamber size is 30 mm x 30 mm x 5 mm and consists of 3x3x4 (36) elements

-Maximum power handling of over 70 W (CW)

– Lower volume (over 40 % for same performance)

-Comparable power handling performance

Varieties of distributed resonators – Co-centric distributed resonator and miniature coax-resonator

PCB Distributed Resonator – 3-pole Prototype I

-3-pole filter fabricated using TMM 13 I substrate, with dielectric permittivity of 12.85

-Individual resonator dimensions are 30 x 30 x 3.175 mm3 , with unloaded Q of about 500

-Coupling walls achieved by subtractive manufacturing

Total filter dimensions 92 x 30 x 3.175 mm3

PCB Distributed Resonator – measured vs simulated results

-Good agreement of tuned measured filter with simulations

-Predicted insertion loss of 1.67 dB vs measured insertion loss of 1.8 dB

-Predicted return loss is 16 dB while measurements show 10 dB in the corresponding frequency range

-Downside on the need for post-production tuning

11 x 11 Multilayered PCB Distributed Filter

-Eliminate the need for post-fabrication tuning – achievable by increasing the number of individual resonant elements. This eliminates the zero-mean statistical imperfections, such as drill depth. It does not eliminate deterministic errors (non-zero mean), such as different dielectric constants in measurements as compared to simulations

-Example below fabricated in PCB multi-layer technology consisting of 5 layer (Tachyon)

-2-pole filter fabricated using 5 Tachyon substrates pressed together, with dielectric permittivities ranging from 3.04 to 3.25

-Individual resonator dimensions are 40 x 40 x 1.27 mm3 , with unloaded Q of about 400

-Coupling walls achieved by subtractive manufacturing

-Total filter dimensions 82 x 40 x 1.27 mm3

-Very high power handling – no performance degradation even at 100 W (CW)

-No post production tuning

Metal drawn mm-wave distributed filter at 28 GHz (9×9 resonant element matrix)

Basic principle

1. Use a large number of resonators
2. Any variations in the manufacturing process is overcome by the collective properties of the design.
3. Avoids postproduction tuning.

Characteristics

1. Qu = 500, 6-pole filter (6 stacked resonant cavities in 2×3 configuration)
2. Dimensions (each resonator): 5 x 5.5 x 0.4 mm3
3. No need for post fabrication tuning using cost effective metal drawing
4. Electrical height of only 13o
5. Insertion loss of less than 1 dB only

Summary

1.Distributed resonators explained – trading height for surface area

2.Resonant frequency no longer a strong function of height

3.Very-low profile for given operational frequency – down to 100 while maintaining performance

4.Possibility of dimensional averaging – no post-production tuning required

Bulk Tuneable materials

To date:

1. Liquid crystals (LC): nematic types examined, limited tuning range 20%-40%. Moderate bias voltages ~ 10 V (dc or low frequency ac). No memory effect.
2. Ferroelectrics (FE): moderate to high bias voltages needed (depending on the thickness of films). High values of dielectric permittivities.

New materials:

1. Dielectro-Phoresis (DP): movement of uncharged particles under the influence of gradient of electric field; slow. More research needed.
2. Phase change materials: Creation of Magneli phases in the interior of oxides of certain transition metals. Moderate bias voltage (up to 10 V).
3. Electro-Chromic (EC) materials: dielectric tunability discovered in 2016. Moderate bias voltages (~10 V dc), memory effect, easily tuneable absolute values of permittivity.

Dielectro-Phoresis 

Facts:

1. The movement of uncharged dielectric particles under the influence of non-uniform (gradient) of electric field.
2. Commonly used in biological sciences to transport different particles.
3. Different particles respond to different frequencies.

Research questions:

1. Can dielectric characteristics be manipulated – can we induce dielectric tunability?
2. Can we make RF devices based on DP?
3. Can we improve on the speed?

Phase Change Materials

1. Phenomenon not fully understood, believed to be due to impurities in oxides of transition metals – formation of Magneli phases.
2. Transition (Mott) from dielectric to metallic state induced by elevated temperature or application of DC bias.
3. Most widely examined oxide VO2.
4. Very promising research area.

Electrochromism

1. Electrochromism (traditionally) refers to changes in optical properties caused by movement of charges induced by external electric field
2. Exhibited by several organic and inorganic compounds
3. Electrochromic materials can revert to its original state by reversing electric field that produced the change

RF test measurement cell

1. Test well-known materials that exhibit optical EC effect for their RF potential (dielectric tunability).
2. Inorganic materials are preferred due to their long life (NiO and WO3).
3. Use height of LiNbO3 as parameter – two cases: 500 nm (cell 1) and 700 nm (cell 2)

Initial dielectric tunability results for 2 cells

1. Dielectric tunability of Cell 1 is between 5 % and 20 %, while for cell 2 it is between 8 % and 18 %
2. Loss tangent tunability is much greater – it is over 360 % for both cells

Tuneable optical properties of EC

1. Modulation of dielectric tunability at mm-wave frequencies extends to optical frequencies
2. Can be modulated in a simple manner – by changing heights of LiNbO3 layer (500 nm and 700 nm)

New EC cell structure – can we do better with standard NiO/LiNbO3/WO3  materials – LiNbO3/NiO/WO3/LiNbO3?

1. Major limitation of standard EC cells lies with the fact that upon actuation, chromic layers become conductive, thereby reducing the channel height.
2. A new EC cell structure shown below does not have such limitation
3. Initial results indicate a change of no less than 80 % is achieved – tremendous result

EC based Aperture-Coupled Reflectarray Element

Topology of the EC- based aperture- coupled reflectarray unit cell (not to scale), [1] (a) front view and (b) side view. The delay line is etched in the isolation layer (SiO2 of thickness 2 µm), however, it is shown as a separate entity.

EC based Aperture-Coupled Reflectarray (5,000+ antenna elements)

Test setup to be used for reflectarray measurements (M – mirrors, P – corner reflectors). The signal path is shown by yellow arrows.

Summary

1. Next generation technologies require high quality RF/mm-wave devices, filters among them.
2. Flexible spectrum usage is expected to play a very important role.
3. The focus is not only on new RF techniques, but the intricate interplay with new bulk tuneable materials.
4. Several promising candidates are identified, but more work needs to be done.
5. Fundamental challenges will need to be addressed by working closely with material scientists, chemists and engineers.

Contact us for more information on our services!

Chia sẻ0

Tweet

Chia sẻ

 Variable attenuators are widely used in radar systems, point-to-point radio, smart antennas and RF signal cancellation. The choice of the attenuator for a particular application is influenced by many factors, such as the dynamic range, the insertion/return losses, power consumption, linearity and power handling capability. Depending on the circuit configuration, attenuators can be clustered into two broad groups: transmission-type attenuators and reflection-type attenuators.  Transmission-type attenuators operate on the principle of attenuating the RF signal in the direct path, i.e. from input to output, while reflection-type attenuators usually employ a hybrid coupler whose coupled ports are terminated in variable resistance devices. Regardless of the attenuator type, for lower power handling capabilities and lower costs, the variation of attenuation is usually achieved using either variable resistance devices, such as PIN diodes or some kind of FETs. The main drawback of variable resistance devices is that they are usually current controlled, inferring that devices based on them are dc power consumers. As such, there exist a need for variable attenuators consuming no or very little power. This can only be achieved if the active device used to control the characteristic of the attenuator consumes no or very little power. One such device is a varactor diode, which is a voltage-controlled device. 

My work on attenuators is broad as evidenced by my publication and patent portfolio, from very low frequencies all the way up to 170 GHz. 

Contact us for more information on our services!

Chia sẻ0

Tweet

Chia sẻ