Optimizing MxFE Front Ends with Complex Gain Control Settings

Organization

• Analog Devices

Challenges

• Accurately model complex mixed signal chains in wideband RF designs

• Optimize front-end performance in context, before using custom hardware

Solutions

• PathWave System Design integrated with MATLAB modeling

• ADI Sys-Parameter Library with models of hundreds of RF amplifiers and mixers

Results

• Flatter frequency response in a wide system bandwidth

• “Early look” capability for de-risking designs

• Estimated six months or more reduction in  design time

Analog Devices, Inc., based in Wilmington, Massachusetts, is a long-time industry leader in analog signal-processing semiconductor technology. As radio frequency (RF) systems continue shifting toward digital technology, Analog Devices (ADI) is incorporating more high-frequency mixed-signal elements in its chip designs.

A key piece of ADI's portfolio is the MxFE™ Platform, mixed-signal front ends designed for direct RF sampling, up/down conversion, and filtering. MxFE targets applications in the communications, aerospace and defense, and test and measurement industries. Uses include software-defined radios, phased array transmitters and receivers, and wideband instrumentation. 

MxFE offers higher front-end integration, helping reduce RF system SWaP (size, weight, and power). Still, applying MxFE chips plus other RF chips in the signal chain can be challenging with wider bandwidths and tighter specifications for crucial metrics like error vector magnitude (EVM) and adjacent channel power ratio (ACPR).

Members of a system integration team at ADI are using Keysight PathWave System Design to create and share a 2-24 GHz RF transmit and receive MxFE front-end reference design. Using modeling and simulation, ADI’s customers can start with this reference design, explore and tune parameters, then customize, optimize, and test their design virtually with confidence, cutting risk and time out of their design cycle.

Challenge: Capturing and Optimizing Complex  Mixed-Signal Chains

Often, the only help customers have when integrating a part is the manufacturer's published data sheet, maybe some application notes, and, ideally, access to a field applications engineer with the right experience. But this practice is inefficient and inadequate for a complex part. Different application contexts using the exact same part can produce wildly different and unpredictable results. In RF systems, real-world effects can show up and ruin hardware prototypes – unless simulation has enough fidelity to spot and fix those problems before committing to hardware.

Design for context is gaining momentum for RF systems, enabled by high-fidelity modeling and multidomain electromagnetic (EM) simulation tools that capture and optimize complex performance details under real-world virtual conditions. Imagine a designer with access to parameterized models created by a part manufacturer, a third party, or internally. With those models, they can make their customized design, then run simulations under conditions they expect in the field.

"The idea for this MxFE reference design came from our internal application expertise and talking to customers about specifications and functionality," says David Brown, Systems Integration Engineer in ADI's Aerospace and Defense organization. "We were using a basic cascade analysis tool, ADISimRF, doing a spot check at just a few frequencies to see if something worked.” ADISimRF is ADI’s free tool for gain, power, and frequency sweeps and level planning using measurement-based models of ADI parts. As shown in, it gives a first look as a stepping stone to higher-level simulation tools.