Printed Antennas for 5G Communication

Abstract

Due to rapid demands placed on the system by growing traffic, it is desirable to have higher data throughput which would enable all the users to have seamless connectivity. However, the current system is approaching its bottleneck as explained by Shannon due to limited spectrum being utilized. [1] Thus newer and higher frequency bands need to be explored as viable alternatives for communication which could accommodate all the user traffic. FCC has already allocated 11GHZ of spectrum for the future 5g communications. The lowest frequency band is centred around 28Ghz. Thus, there is immediate need to develop prototypes of communication systems and RF components. One key component which needs more inspection and development is Antenna system for 5G Communication. Because of the stringent requirements of 5g communication in terms of low latency, low Signal to Noise Ratio, higher bandwidth aimed at having a higher carrying capacity, having a MIMO system to enable point to point communication without interference from spatially adjacent communication systems. For satisfying such requirements, it's important to develop a high gain and wide bandwidth antenna system. The currently developed antennas do not have wide enough bandwidth and have not fully explored the possibility of having a rapidly scalable design which can be seamlessly developed into an array as the requirement for gain is scaled up. Thus, it was essential to develop such an antenna system which has high gain, low profile, high bandwidth, and can be made into an array seamlessly and whose beam can be steered by using some phase compensation mechanism. Although current designs do exist which are later discussed but they do lack insightful discussions of certain topologies which could be indispensable for certain applications such as in development of total surface wave suppression antennas which do not interfere with the circuitry around it when encased at a base station with other RF components. Based on the above factors, it was decided to focus on a research project to develop an antenna for 28 GHz which would satiate all the requirements of the design without having to compromise upon either the gain or the bandwidth of the interference the antenna presents to circuitry around it. It was also desirable if the antenna has an array design such that it can be easily scale to enhance the gain. The study commenced by analysing the key designs presented by antenna designers around the world for Ka band( 26.5 – 40 GHz) and identifying the issues with each design. Most of the designs presented have at-least one drawback, such as the design in [4] has impedance bandwidth of only 10 % and more importantly it does not cover the first band of 5g communication which is centred around 28Ghz. Similar drawbacks in other designs are explored in detail later. Post Literature review, the most apt radiating element was decided upon : Magneto Electric Dipole Antenna fed with a transverse coupling aperture on E wall of Substrate Integrated Waveguide due to it's wideband resonance structure, uniform gain across the functioning spectrum and high front to back radiation ratio which leads to better directionality and lower interference with adjacent antenna systems. At 28 GHz, the profile of Magneto-Electric Dipole can be low enough to enable easy integration into PCB structures and further fabrication by laser etching. Existing literature related to such profiles were examined and one such design in [3] at 60 GHz was chosen as the starting point for further development. Development of antenna at 28 GHz raises one important question: what would the feeding structure be ? Major challenges which need to be solved to resolved this issue were tackled by firstly, literature review in Antenna Theory by Balanis and then examined via simulations. Microstrip line, Suspended Stripline and Substrate Integrate Waveguide were examined as viable feeding structures by running simulations on commercial software available in laboratory : Computer Simulations Technology ( CST) Microwave Studio. Upon examination of the results of the simulation software it was decided to choose Substrate Integrated Waveguide(SIW) as the feeding structure for the antenna so as to avoid spurious radiation at such high millimetre wave frequencies. The antenna acts as a transducer between free space and a waveguide which carries the communication signal, and as such it's crucial to have a proper impedance matching to avoid having high standing waves in the feeding lines and to radiate as much power as possible into the free space. The impedance matching was commenced with a radiating structure already decided and rough dimensions calculated as per the theoretical dimensions of an Electric Dipole and Magnetic Dipole from theory. The structural parameters were then varied within 20 % of their initial values to see their effect on the impedance and see their correlation with theory with respect to changes in input resistance and reactance of the antenna. Once the correlations were established, the antenna was tuned for having least reactance and most resistance. Antenna element has a final return loss of less than 10 dB impedance bandwidth of from (24.42 – 34.92) GHz which is 37.5 % assuming centre frequency of 28 GHz. The antenna is linearly(vertically polarized with a realized gain of 7.626 dB with coplanar to crossplanar ratio of 11.625 dB. Development of Substrate Integrated Waveguide (SIW) was also crucial to the development of antenna, such that it can allow propagation of waves above the cut-off frequency (21 GHz) seamlessly and also block higher order modes from propagating below 32 GHz. The design equations in [2] were followed to develop the fully functional SIW. For connecting with a commercially available waveguide so that power may be coupled into the Substrate Integrated Waveguide (SIW), a rectangular waveguide to Substrate Integrated Waveguide (SIW) transition structure was developed, inspired from [5] by redesigning some of the parameters in the original design so that the new transition is apt for 28 GHz. The waveguide to Substrate Integrated Waveguide (SIW) transition has a broad impedance bandwidth with return loss of -10 dB from (24.943 – 30.475 ) GHz which is 19.75 % considering centre frequency of 28 GHz. Considering the fact that for 5G communication, a gain of <8dB might not be enough, a new lens based on phase compensation by using Magnetic near Zero Material was studied as stated in [6]. The lens was readjusted and redesigned for fabrication on the substrate available in the lab and for working well with the Magneto Electric Dipole with an aperture in SIW wall. The antenna with the lens has a realized gain of 10.74 dB, a gain of 3.114 db from the original design. For the purpose of controlling the direction of mainlobe and also to enhance the gain, it's imperative to develop a novel antenna array which would enable the antenna to have a higher gain. But before developing the antenna array, a feeding network for delivering power to the array elements was developed. T- junction, 3 way power splitter and L-junction were all developed in order to develop the distribution network. Each of these elements was optimized to have a return loss of less than 20 dB over frequency band of (24 – 32) GHz which is 28.57 %. The antenna elements were closely packed and connected to the distribution network to make a power distribution network of 2x2. The Array has a gain of 11.99 dB. The same array with simulated with the lens element on top and it achieved an even higher gain of 13.15 dBi. Due to the physical constrains, the elements could not be spaced any closer, thus optimum gain of 14 db could not be achieved. The array has an impedance bandwidth of -10 db over a 20% bandwidth , from (25 – 30.6 GHz). It is worth mentioning that the array was simulation after the addition of the feeding transition, thus even its integral higher bandwidth might have been masked by the transition's intrinsic bandwidth. Seeking inspiration from Leaky wave antenna, the design was modified to accommodate closer spacing of antenna elements by having multiple Magneto-Electrical Dipoles onto a single Substrate Integrated Waveguide. By having such a system, one can have a bigger array, with closer radiating elements and higher gain. The development of this array is still in development phase and further work needs to be done in order to make it fabrication ready. Such a design has a potential to be a very wideband structure with very high gain. The beam's mainlobe can be swept by introducing phase shift in individual SIWs being fed by means of a variable phase shifter. Such phase shifter can easily be introduced by using microstrip to SIW transition and by using a micro strip line based varactor controlled phase shifter as indicated in [7]. Ultimately, for a system to be fully support 5g, it must have the controllable beam direction, however, owing to the paucity of time, the full-fledged development of project could not take place. But, for a real application based on the design developed in the paper here, it would be optimum if the additional design ideas which will be further explored down below are incorporated. However, if an application does not need the option of beam scanning, any one of the designs mentioned above could be chosen and scaled up depending on the intended gain and bandwidth using the CSR software, and the macros provided which have been developed in-house for standard development of stand-alone parts such as resonating cavity, siw, power divider to create optimal solution. Furthermore, for fine tuning the design, optimal settings for tuning the Magneto-Electric Dipole for any frequency using frequency sweep and CST's optimizers have been elucidated for reader's referral.