Our 5G future: in the fast lane with numerical simulation
04 October 2018
5G and IoT are among the hottest topics being discussed in the RF and microwave industry. Every day, activities and technological advancements depend more than ever on reliable, fast data communication. This piece explains how designers are now faced with one of their biggest challenges, as they strive to take real-time data usage and availability to the next level...
For the digital issue of this piece, please visit this link – or click here to register for EPDT's magazine.
To achieve such a milestone, companies require access to the best design tools, and significant advances in signal processing, device-centred communications and evolving technical standards.
Predictions state that 5G will also need to utilise higher frequency spectrums, in the millimetre wave (mmwave) range, when deploying active electronically scanned array (AESA). This enables multi-beam multiplexing and massive multi-input-output (MIMO) technologies (see Figure 1 – links to ActiveMag). Researchers working on the frontlines of forging this ultra-fast and high bandwidth successor to 4G LTE are relying on modelling and simulation tools to optimise product development and test cycles.
Simulation supports engineers throughout the design cycle by allowing them to virtually evaluate multiple design ideas, and implement physical prototypes based on the most promising concepts. Another advantage is the possibility to investigate different boundary conditions: simulation allows engineers to efficiently measure and test several scenarios without damaging a prototype (in cases such as extreme temperature variation, structural deformation or chemical reactions).
The goal of simulation is to mimic the real world as closely as possible, so that the prototype is based on numerical results that achieve the expected performance in fewer design and test iterations.
Design simulation via apps
In preparation for the 5G rollout, designers are working through numerous obstacles, including frequency choices, propagation, reliability, battery life and interference, to name a few. Each of these challenges is represented by a unique blend of physics, requiring a simulation specialist in that specific area, who is equipped with the right tools to set up the underlying mathematical model properly. The symbiosis between designers and simulation specialists must be optimised to deliver the right product at the right time.
Simulation experts are typically the only ones who can safely provide the input data needed to get a useful output from a model. They therefore have to be involved in the iteration process every time there is a new request or change to be made in the device being simulated. Additionally, results or outputs are often presented in an environment only familiar to the specialist, so distributing information to their colleagues often requires presenting an explanation and interpretation of the results.
But what if simulation specialists could easily build simulation apps? In other words, wrap an intuitive interactive user interface around a complex mathematical model. What if users without any previous experience using simulation software could run apps specifically designed for them?
Simulation apps make it possible for simulation specialists to efficiently and effectively support the designers working on the next breakthrough in the ultra-competitive landscape of wireless communication. Supplied with the right tools, designers focused on 5G implementation can freely collaborate and complement their skills with those of their colleagues and collaborators, who specialise in physics and numerical analysis.
What simulation apps can do for wireless communication design
Let’s take the example of active electronically scanned arrays, or phase antenna arrays. They have become popular for military use in radar and satellite applications and are now being applied to commercial purposes, due to the growing needs of higher data rates in communication devices. The size of a simple component can easily exceed tens of wavelengths, making its numerical analysis very memory-intensive.
As a result of the above, models take a very long time, even when approximated values would be sufficient to evaluate a proof-of-concept design. Rapid prototyping would help reveal performance tendencies and determine design parameters quickly.
Figure 2 shows a simulation of a 4x2 phased microstrip patch antenna array, which can steer the beam toward the desired direction. This example is significantly more memory intensive and will take a longer time to compute than a single microstrip patch antenna (see Figure 3). The simulation results shown in Figure 3 are based on a full finite element method (FEM) model of a single slot-coupled microstrip patch antenna – built on low-temperature co-fired ceramic (LTCC) layers, initially operating at 30 GHz.
Can we use the analysis of a single antenna to describe the behaviour of the entire array? The power and flexibility of COMSOL Multiphysics software allows simulation specialists to perform an accurate simulation of a single microstrip patch antenna, and then take into account user inputs such as array size, arithmetic phase progression and angular resolution to describe, for example, the 3D far-field of the entire array.
COMSOL makes it easy for specialists to couple physics interfaces already available with equations or algorithms needed to model a specific application. In this case, the two-dimensional antenna array factor has been implemented to include the translational phase shifts and array element weighting coefficients needed to determine the radiation pattern of the entire array.
Can we present such a model to designers in a user friendly way? Simulation specialists are now provided with an intuitive workflow to create custom user interfaces based on their multiphysics simulation model. An app built for simulating the antenna array is shown in Figure 4.
This app allows the designer to change the physical size of the single microstrip patch antenna, as well as the thickness and material properties of each layer, in addition to other relevant parameters determined by the simulation specialist. In this particular example, the simulation specialist has included an interactive user experience by indicating whether the chosen design parameters are appropriate or not – by comparing the computed S-parameter (S11) value to the pass/fail target criterion.
The app also includes a results report and documentation that concisely explains how the app is working. This last feature can be used in a variety of practical ways – from building reports for stakeholders and management, to use as a training tool for new hires in the company. Apps can also be easily deployed to colleagues and collaborators through a local installation of the COMSOL Server product, allowing authorised users to access apps through COMSOL Client or a major web browser.
We have a lot of work ahead of us before 5G is unveiled to the public. When designers are equipped with the right set of tools, they can freely collaborate with colleagues throughout their organisation and beyond. Working cross-departmentally will be key to competing and succeeding in the 5G race.
Contact Details and Archive...