Optically Transparent Antenna for Smart Glasses

The optically transparent antenna technology has been gaining attention in consumer electronic industry owing to its potential to release the antenna volume constraints in compact mobile devices. Earlier studies have been heavily focused on the transparent antenna integration to the display for smartphone applications, for example, to improve its spatial coverage of 5G mm-wave at the forehead side. To our knowledge, the application of the transparent antenna technology to smart glasses and augmented reality glasses, especially considering the constraints from the coexistence with optical features in the lens stack-up, has remained unexplored. Thus, for the first time, we suggest a feasible existence solution between antennas and the RF lossy layers in the lens stack-up, to reduce the negative impacts on antenna performances. A slot loop antenna formed by metal-mesh based transparent antenna film and metallic glass frame has been investigated. Experimental results of such an antenna on human head phantom shows a total efficiency better than −4.5 dB in 2.4 GHz band. Even though it is demonstrated through a 2.4 GHz antenna example, the same principle can be applied to any other sub-6 GHz antenna applications, i.e., LTE, WiFi, GNSS, etc.


I. INTRODUCTION
W E FUTURE immersive wearable computing platforms will connect people to the new experiences, from immersive education and training to the new possibilities in healthcare and the workplace, and beyond. The Augmented Reality (AR) and Virtual Reality (VR) devices are key to offer such immersive experiences. Consumer electronic industry has been heavily investing with commercial products already made available to the market from Meta (formerly Facebook), Google, Microsoft and many others. Despite the tremendous efforts from the industry, technology breakthroughs are demanded for developing the enabling technologies that are required to support the 3D virtual experiences and yet packing them into light-weighted, long-term lifetime, stylish and socially acceptable wearable form-factor devices. Specific to antennas on smart glasses, there are restrictions in enclosure on the conventional placement of antennas such as by Printed Circuit Board (PCB), Laser Direct Structuring (LDS), Flexible Printed Circuit (FPC), chip, metal plates, metal wires, etc. Therefore, the transparent antenna has been drawing attention from industry recently as one potential solution to address the antenna volume constraints and to achieve a high design freedom. Many transparent antenna research and application in consumer electronic domain have been focused on the transparent antenna and display integration for smart phone [1], [2], [3], [4], [5], [6], smart watch [7], [8], etc. Large space or high freedom of placement are desired in the design for antenna performance optimization, for example, antenna efficiency, antenna gain, bandwidth, antenna isolation, tuning capability, polarization, and so on. In particular, the implementation of transparent antenna and display integration provides an option to achieve 5G mm-wave coverage at the forehead side for the all-screen display phones [1]. Other transparent antenna applications in the industry include but not limited to car [9], [10], solar panel [11], [12], [13], [14] applications.
Antenna design has been one of the most significant engineering challenges in hardware design for smart glasses-the antenna design is constrained by the stylish industrial design requirements, while the antenna performances suffer from severe body effects, including detuning, attenuation, and shadowing effects. From a system level, AR applications demand high throughput and low latency while the conducted power for the antenna port is constrained by the battery life requirement as well as the compliance with human electromagnetic exposure requirement, i.e., Specific Absorption Rate (SAR). On the other hand, the lenses in the glasses are the single biggest component / module in the glasses form factor, and they offer the largest real-estate in the system and the best flexibility to tune the antenna performance towards the system requirements. A generic AR glass [15], [16] is illustrated in Figure 1. Some key optical functions, i.e., micro-LED, active or passive dimming, optical waveguide, eye tracking etc., are expected to be built into the lens which help connect people from the real world to the virtual world for the immersive experiences. Hence, a major technical challenge in implementing a transparent antenna is the coordination of the transparent antenna layer with other functions that already exist in the lens stack-up. In another word, the transparent antenna must ensure the minimal optical performance degradation within acceptable range, while still demonstrating the antenna performance benefits as compared to the conventional antennas in the glasses form factors, i.e., the LDS-or the PCB-antenna in the glass temple or frame areas. While optical functional layers in the lens stack-up is still evolving with huge technical challenges in their optical performances, the Light Control Layer (LCL) is identified as the material that has the largest negative impact on antenna performances among the existing functions, and the coexistence design of lossy LCL and  [17], [18], [19], [20]. transparent antenna becomes key for transparent antenna to be integrated into modern smart/AR glasses.
In this paper, the LCL and transparent conductive material made by metal mesh are firstly described. Through the studies of several basic antenna types, the transparent slot antenna concept emerges to be the best option for transparent antenna to coexist with the lossy LCL. The design has been simulated and validated with prototypes.

II. TRANSPARENT MATERIALS
This section provides a review of Light Control Layer (LCL) with the analysis to explain why it has the largest negative impact on antenna. This section also reviews the typical approaches to design optical invisible but electrically conductive films.

A. LCL
The general LCL shown in Figure 2 is also known as "active dimming layer." The main types are PDLC (Polymer Dispersed Liquid Crystal) [17], [18] which is good at scatter controlled and GHLC (Guest-Host Liquid Crystal) [19], [20] which has advantages of optical transmittance controlled. In both cases, the Liquid Crystal (LC) is sandwiched between two layers of transparent conductive film (typically Indium Tin Oxide: ITO), and the orientation of the LC can be changed by applying a voltage bias, thus allowing the optical properties of the layer to be changed electrically. The choice of GHLC vs PDLC depends on the application, i.e., AR vs. VR, and also the optical features that they have to co-exist with.
At radio frequency, the ITOs on LC have significant impacts on transparent antenna performances, because of the high sheet resistivity nature of the typical ITOs. For example, the typical resistance of ITO is about 10 to 500 /Sq, to maintain the acceptable optical properties. The RF impact from LC and mixed LC is minor even when the bias of LC is ON or OFF since the difference between the vertical and horizontal Dk and Df of LC is small. If a transparent antenna on lens is not designed properly, the proximity to the lossy ITOs leads to a severe performance degradation.

B. TRANSPARENT CONDUCTING FILMS AND METAL MESH
Transparent conducting films (TCFs) are very thin, and the material itself is optically transparent and electrically conductive at same time. Several transparent antennas were developed using these TCFs in the past [21], [22], [23], [24], [25], [26], [27]. ITO is the most widely used TCF in electronic devices. Silver Nanowire (SNW) has randomly conducted network made by silver wires which is scaled shorter than the wavelength of visible light. Graphene and Carbon Nanotubes (CNT) are also being developed in recent years as one of the promising TCFs.
On the other hand, Metal Mesh (MM), as one of the TCFs, has been mainly used in touch sensors on display and electromagnetic shield in the past. Recently, the use of MM to design transparent antennas has gained increasing attention from the industry owing to its low resistivity [1], [2], [7], [8], [27], [28], [29], [30], [31]. MM can be made from manufacturing methods, such as subtracting method by photo-lithography, additive method by plating up, printing, transfer method, etc., these process similar to the general fine pattern for PCB or Integrated Circuit (IC) manufacturing.
Among all TCFs mentioned above, there is always a fundamental trade-off between electrical resistance and optical properties as illustrated in Figure 3. Optical properties includes the well-understood visible light transmittance as well as scattering (e.g., Haze), color parameters (A * , B * , yellowness index, etc.), and the Moire effect caused by interference with other periodic structures specific to MM. Recently, both MM and ITO based transparent films are considered for antennas. Since the MM approach provides lower resistance and more flexibility to adjust the sheet resistance vs optical transparency, the consumer industry and transparent antenna research community are both trending to use MM as the major transparent antenna materials.
The sheet resistance of MM is related to its shape, the width/thickness, the pitch of wires and metal conductivity. The metal mesh structure can be any geometric periodic structure, i.e., square, rectangular, hexagonal, diamond, triangular, hybrids of different shapes, rotational angles of arrangement, periodicity staggered, even randomness. In any geometric pattern, the size and periodicity of the structure is on a scale sufficiently larger than the visible light wavelength (380 nm to 780 nm). Each has its own advantages and disadvantages, from both optical and electrical perspectives. This paper describes the major parameters for the case of a 45-degree rotation of a square. Figure 4 shows the metal mesh structure and resistance network schematics.
Re is the element resistance of mesh branch, N and M is the number of mesh intersections in the width direction, and in the current path direction, respectively. The aperture ratio Ap (%), the area that one intersection can cover AC (m 2 ), and the DC sheet resistance Rs ( /Sq) can be calculated from the following equations: where MW is the width, ML is the length, MP is the periodic pitch (ML and MP are the same for a squared shape), MT is the thickness, and σ is conductivity of metal mesh, respectively. The effective σ (s/m) is calculated from the measured dimensional values and DC sheet resistance Rs of the manufactured MMs. A photograph of the MMs used in this paper is given in Figure 5. MP is 200 um, MW is

III. TRANSPARENT SLOT ANTENNA ON LCL A. TRANSPARENT ANTENNA TYPES STUDY
A plot of calculated Ap and Rs of general configurations is given in Figure 6. The Ap based on only MP was changed to see the Rs values. The thickness of metal mesh MT has dependencies on the mesh width MW. In this calculation, we make an assumption that MT = MW, which is typical from the manufacturing process. The effective conductivity is set at 50% of Cu (i.e., 5.8 × 10 7 × 0.5). Ap is also correlated to optical transmittance (Tr) and scattering, indicating that there is a trade-off between Rs and optical properties of metal mesh. From optical visibility perspective, it is recommended that the mesh width is no more than 2um. From the optical transparency perspective which is correlated to AP, the wider mesh width MW, the better AP, which is indicated from Figure 6. When it comes to antenna design, it is reminded that there are three dominant materials to form antenna elements: the transparent conductive material, e.g., metal mesh antenna layer, the lossy layer which locates in close proximity to the antenna layer, e.g., ITOs of LCL, and the metallic or plastic enclosure from the glasses. Figure 7 shows not only layer stack-up of transparent antenna and LCL on lens but also the possible antenna types that can be fit into the form factor. The actual layer stack-up should have a complicated combination of base materials and Optical Clear Adhesive (OCA) for each function. However, they're not included in this study since they're especially thin and can be negligible at radio frequency.
There are a total of five types of transparent antenna on lossy material, with Types (A) -(C) consisting of MM antenna and lossy LCL only, and Types (D)-(E) also including metal enclosures. In all types, they are multi-layer structure similar to PCBs but without vias. Type (A) is a dipole antenna. Type (B) is a Microstrip Antenna (MSA). Type (C) is a slot antenna within the transparent region. Type (D) is a monopole antenna locates at the edge of the transparent region and using metal enclosure as ground plane. Type (E) is a slot antenna formed between the transparent region and the metal enclosure. The film size of the transparent sheet is given as 40 mm by 50 mm, and its metal enclosure is a rectangular ring with a 2 mm by 2 mm cross section. In these studies, each parameter and the results of the analysis are summarized in Table 1. Except antenna size (W and L), the same materials were used in all antenna types as a fair comparison, for example, the sheet resistance of MM is 2 ( /Sq.), and the sheet resistances of ITOs in LCL, Rs2 and Rs3, are both 200 ( /Sq.). It is noted that the value of Rs1 on MM is determined from the target value of optical properties (e.g., Tr), and the values of H, Dk, Df , H2, Dk2, and Df 2 are from the manufacturing process and materials. It is well understood that when the lossy ITO materials are in close proximity of the antenna, the antenna efficiency drop.
In the design of transparent antenna, the design variables include the antenna pattern, the mesh width, mesh pitch, the gap between MM layer and metal enclosure G1, and the gap between ITO layer and metal enclosure G2. The simulation results of all five types of antenna designed shown in Table 1 is described in this paragraph. All antennas are well-matched at 2.4 GHz. The radiation efficiency of Type (A) to (D) is significantly lower because the most of the RF power converts into Ohmic loss due to absorption of the resistive sheet of Rs1, Rs2, and Rs3. For the same reasons, it yields a large impedance matching therefore the discussion of return loss here is negligible. It is found that among all five antenna types under investigation, the type (E) edge slot antenna performs the best. In edge slot antenna design, the transparent antenna film matches the pattern of the ITO in LCL and is directly placed against the ITO layer. With such a configuration, the metal mesh layer and ITO layers are strongly coupled to each other. Majority of the currents follow the least impedance path and is distributed at the metal enclosure and the conductive metal mesh layer. The lossy effect of the ITOs can be reduced. The additional analysis with the removal Rs2 and Rs3 lossy LCL layers from Type (E) slightly improves antenna radiation efficiency from 35.88% to 38.10% indicating that the negative  impact from LCL is minor for the proposed antenna. G1 and G2 are the key parameters to increase radiation efficiency and kept as 1 mm throughout the study. When G1 = G2 at Type (E) is modified up to 0.5, 1.0, 1.5, and 2.0 mm, its radiated power increases to 18.21%, 35.88%, 48.18%, and 56.44%.

B. ANTENNA CONCEPT AND DESIGN
A Type (E) antenna, i.e., edge slot antenna between transparent conductor and metallic enclosure was designed for 2.4 GHz applications. The shape of glass frame refereed from [15], [16]. For the antenna input impedance and resonance of the edge slot antenna, the tuning parameters are the locations of antenna feed and shorting PINs. Changing these parameters is equivalent to changing the length of the slot antenna or impedance match. A simulated antenna radiation pattern and current distribution are shown in Figure 9. Since the proposed edge slot antenna is slot formed between the transparent sheet (i.e., including transparent antenna layer Rs1, LCL transparent conductor Rs2, and Rs3) and metal enclosure, the radiation pattern radiates both to the outside and to the human side. The proposed edge slot antenna has multiple resonance including 2.16 GHz as second harmonic of a full wave length ('b' -'A' -'B' -'D' -'c') for best radiation efficiency close to 2.4GHz application, 1.6 GHz as with parasitic slot ('c' -'C' -'b') plus fundamental slot ('b' -'A' -'B' -'D' -'c') or, slot loop will potentially cover GPS L1 band.
The visual and representative dimensions of the model used in the analysis are shown in Figure 8. Figure 10 shows the results of the analysis of the radiation performances and reflection characteristics of the designed antenna in free space. Total efficiency of −4.40 dB, as well as excellent matching are obtained at 2.4 GHz.

IV. PROTOTYPE AND MEASUREMENT RESULTS
This chapter describes the results of the prototype evaluation for LCL and antenna performance. Figure 11 shows the prototype with one side has transparent antenna on lens and another side has lens only, both have no LCLs. The gap between MM layer and metal enclosure G1 and the gap between ITO layer and metal enclosure G2 are 1mm which is the same as the design in the previous chapter. As shown in the figure, human eyes can hardly differentiate the discrepancy between the two sides of lens.

A. LCL
LCL with PDLC was implemented in the prototype shown in Figure 12. The reason was that it was possible to correctly evaluate the relatively high loss transparent resistance that affects the antenna performance, and it is easy to fabricate. After processing and mounting the electrodes for bias,  LCL was mounted on the lens with transparent antenna electrode. As shown in the figure, the transparency is apparently changed by turning bias ON or OFF. In additional measurement, there was almost no impact from LCL bias status on antenna performances.

B. ANTENNA
Antenna characteristics were evaluated. Figure 13 demonstrates the over-the-air measurement setup and radiated antenna pattern at 2.4 GHz. The radiation pattern indicates that the majority of the total gain shows on the world side due to the shadowing effect of the human head phantom. The measurement results of prototype with LCL are also shown Figure 14. In each of the radiation and reflection characteristics, both with and without the human head phantom were measured. A comparison of the measured results without the human head phantom and the simulated results in the previous chapter shows an excellent agreement. After taking account of 0.28 dB cable loss at 2.4 GHz, the estimated measured antenna efficiency with head phantom is −4.36 dB.

V. CONCLUSION
The transparent antenna is a good candidate to solve the antenna volume constraints for smart glasses and augmented glasses form factors. This paper summarizes the practical constraints of transparent antenna designs and performances when they co-exist with the RF lossy optical features/functions in the lens stack-up. For the first time, the edge slot antenna concept has been proposed to address the coexistence issue with the optical features, where, more specifically, a matched metal mesh layer is directly attached to the lossy light control layer. With such configuration, the RF loss from the lossy optical features can be significant eliminated. Experimental results on human head phantom verified our simulation with a total efficiency of −4.36 dB at 2.4 GHz.