Increased support for RF MIMO systems design and development with MathWorks updates

Increased support for RF MIMO systems design and development with MathWorks updates

Technology News |
By Jean-Pierre Joosting

I recently spoke to Ken Karnofsky, Senior Strategist for Signal Processing at MathWorks and we discussed these significant challenges which designers will face in next generation wireless systems. New and demanding roles and skills will be required of designers coupled with the need to accelerate product cycles. The high costs of verification will need to be minimized and dealt with as well.

To ease a designer/R&D team’s burden, MathWorks has made some crucial new updates to the RF Toolbox , SimRF, and Antenna Toolbox that will strengthen design support for digitally-assisted RF MIMO System.

As a part of Release 2016a , these new updates will help engineers ramp-up on RF simulation, assist in performing a first order RF budget analysis that is extendable with advanced models, and help integrate the results of RF design in system-level simulation. The new capabilities will now allow wireless R&D engineers working on 5G and other advanced wireless communications systems to use MATLAB and Simulink directly for RF and antenna modelling, avoiding the need to learn and maintain separate, specialized tools.

Karnofsky said that designing a next generation wireless system will require a minimum of seven different skills in order to achieve a robust system that will meet all the expected demands by users who will expect 5G to provide service anytime and anywhere with high speed capability. See Figure 1 for those skills.

Figure 1: The seven skills designers/R&D teams will need to accomplish the needs of 5G (Image courtesy of MathWorks).


System designers and R&D teams will include Digital, and RF expertise and will need to model in order to create algorithms, do end-to-end simulations and have a “golden” reference for verification so that they can fully design a functional and robust system. See Figure 2.

Figure 2: Modelling: Digital and RF, and System Perspectives (Image courtesy of MathWorks).
Figure 3: Do your next generation Wireless design going from MATLAB design with rapid and flexible algorithm exploration including HDL and C code generation for FPGAs, processors, and ASICs, design, and analysis to SIMULINK modelling and simulation of digital, RF, and antenna elements. (Image courtesy of MathWorks).

New enhancements have been added to the design phase, bringing more tools into the environment leading to end-to-end RF simulations done quickly.

Digital and RF co-design is made possible so that designers/R&D teams are able to look at such things as Digital Pre-Distortion (DPD) in a base station Power Amplifier (PA) using a designer’s unique DPD algorithm or a basic algorithm in MATLAB. Then monitoring the spectrum all the way through the signal chain with a simulated spectrum analyzer can be achieved.

This system works with key manufacturer’s components and demo boards/reference designs such as the Analog Devices AD9361 AGILE transceiver. See Figure 4.

Figure 4: Create a transceiver design using an exact, verified model of Analog Devices AD9361 Agile Transceiver along with RF components, digital filters, and control logic. Designers can configure and debug in simulation, predict impact of RF behaviour on system performance, and stimulate with LTE and other reference signals. (Image courtesy of MathWorks).

Doing a top-down RF analysis and simulation

Figure 5: Designers/R&D teams can import, visualize and export RF data (S-parameters), automate analysis of RF measurements, build and analyze RF networks. (Image courtesy of MathWorks).
Figure 6: Do a Top-down design of RF architecture and specs and integrate and simulate RF and control algorithms as well as test and debug implementation before going into the lab. (Image courtesy of MathWorks).


New: RF Budget Analyzer app

Figure 7: The RF Budget Analyzer simplifies design workflow, implements power/noise/IP3 RF link budget, generates models and test benches for Circuit Envelope simulation and proves consistency between analytical and simulation results. (Image courtesy of MathWorks).

New: Automatically Generate Simulation Model from RF Budget Analysis

The RF Budget Analyzer will calculate the cascaded gain, noise figure, and 3rd order intercept (IP3) of a chain of RF stages. Each stage is represented by a parameterized component model, specified with its own stage gain, stage noise figure, and either an output referred stage IP3 (OIP3) or an input referred IP3 (IIP3).

First, specify the system by defining a gain, noise figure, OIP3, and name for each stage and create an RF Chain object to represent the cascaded system. Designers can then view the calculated cascaded values of gain, noise figure, and IP3.

The RF Budget Analyzer can generate SimRF models from the analysis results, as well as test benches to validate analytical results against circuit envelope simulations. The models can be used in end-to-end simulations in Simulink.

Figure 8: Simulation Model (Image courtesy of MathWorks).
Figure 9: Test Bench (Image courtesy of MathWorks).

Antenna Design and Analysis

About a year ago, an Antenna Toolbox was added to MATLAB. Now with this new addition, a wireless system designer can model antennas within the system without having to do full CAD design or become an antenna expert.

Figure 10: Designers/R&D teams receive a great deal of help with a parameterized antenna library and antenna arrays. Seamless integration is easy with the ability to model the antenna together with signal processing algorithms and fast iteration of different antenna scenarios for your system design. (Image courtesy of MathWorks).

Since antennas are frequently mounted onto substrates, new additions to this system allows the designer to choose from catalogue or to define with custom dielectric materials. See Figures 11 and 12.

Figure 11: A dielectric catalogue will help designers in their task. Dielectric properties affect resonant frequency, bandwidth, efficiency, pattern and more. (Image courtesy of MathWorks).
Figure 12: Dielectric Antenna substrate models are available and custom dielectric materials may be chosen from a software catalogue as seen in Figure 11. (Image courtesy of MathWorks).

Next, designers/R&D teams can incorporate their antenna and RF models into the Downstream Workflow. See Figures 13 and 14.

Figure 13: Designers/R&D teams can model RF/Microwave behaviour such as impedance, resonance, return loss, bandwidth and more. (Image courtesy of MathWorks).
Figure 14: Antenna behaviour can be modelled and seen such as Radiation patterns, beamwidth, E-Plane and H-Plane, and Polarization. (Image courtesy of MathWorks).

5G technology R&D

With 5G being so close to implementation, tough requirements need to be addressed, such as Enhanced mobile broadband of greater than 20 Gbps, Ultra-low latency of less than 10 msec, massive machine-type connectivity, ubiquitous coverage and service everywhere and anytime, and LTE/WLAN co-existence.

Some candidate technologies to make the above requirements a part of the 5G system reality, new technologies must be implemented such as Massive MIMO and beamforming, Active phased array antenna systems, emerging waveforms like UFMS, FBMC and GFDM must be addressed, higher bandwidth and mmWave bands employed and so much more.

Figure 15: MIMO beamforming can be implemented into an development/evaluation system in an end-to-end analysis of system performance. (Image courtesy of MathWorks).

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News


Linked Articles