MATLAB and Simulink for Mixed-Signal Systems

Analyze, design, and verify analog and mixed-signal systems

Use MATLAB® and Simulink® for behavioral modeling, rapid design exploration, predesign analysis, and verification of mixed-signal systems.

For getting started with designing mixed-signal integrated circuits (ICs), you can use Mixed-Signal Blockset™ models of PLLs and ADCs. Building blocks are characterized with data sheet specifications and include analog impairments. Built-in analysis tools and measurement testbenches help you with reducing the verification effort.

For the design and analysis of high-speed links, such as PCI Express®, USB, DDR, and Ethernet, you can use SerDes Toolbox™ to build and assess your channel equalization scheme and automatically generate IBIS-AMI models for channel simulation.

With MATLAB and Simulink, you can:

  • Create behavioral models of PLLs, DACs, ADCs, SerDes, SMPS, and other mixed-signal systems
  • Evaluate analog-digital design tradeoffs following a top-down methodology
  • Link system-level models to EDA tools via co-simulation or by creating SystemVerilog modules and IBIS-AMI models
  • Verify designs including analog/digital hardware and control logic before producing test chips

"Circuit-level simulations took three days. Using MATLAB and Simulink, we reduced simulation time to just one minute."

Jun Uehara, Epson Toyocom

Using MATLAB for Mixed-Signal System Design

Mixed-Signal Analysis

At the highest level of abstraction, you can use MATLAB to analyze basic system architectures; for example, which is better: a second or third order sigma-delta modulator? What type of PLL is best? What do the Bode plots say about system stability?

Use analysis tools in MATLAB and Simulink to explore the design space and find the best starting point for your design. For example, Mixed-Signal Blockset uses MATLAB functionality to perform closed- and open-loop static analysis of PLLs and rapidly design loop filters.

MATLAB provides greater analysis and visualization functionality than spreadsheets or traditional programming languages like C/C++. However, you don’t have to abandon your existing investments; MATLAB works with Microsoft® Excel® and with C/C++.


Mixed-Signal Top-Down Design

Use and elaborate behavioral models and measurement testbenches to enable faster design and verification. In Simulink, you can simulate analog circuits together with control logic and digital hardware at different levels of abstraction.

Describe analog electronics either using continuous-time signals at the “transfer function” abstraction level or using Simscape Electrical™ to model voltages and currents and components such as RLC elements, op-amps, and switches.

Describe digital electronics at the algorithmic level using floating-point accuracy or perform bit-accurate simulations using fixed-point data types of arbitrary length, including quantization and saturation effects. Lastly, generate synthesizable HDL code for targeting ASICs and FPGAs.

Describe control logic and state machines at the algorithmic level using MATLAB functions or Stateflow®. You can use fixed-point data types and decide whether to target microcontrollers using embedded C/C++ code generation or generate synthesizable HDL code for targeting ASICs and FPGAs.


Mixed-Signal Verification

System-level models must be linked to the next stages in the design flow. You have different ways to use your MATLAB and Simulink models as test harnesses for SPICE models, HDL code, or hardware.

Cosimulation is a run-time link between different tools; at every simulation time step, data is exchanged between tools, enabling them to run together to simulate a model. In the analog domain, Cadence® Virtuoso® AMS Designer provides cosimulation links to Simulink. In the digital domain, HDL Verifier™ provides links to third-party HDL simulators and FPGAs boards for in the loop testing.

For regression testing and reuse in functional verification environments, you can export MATLAB algorithms and Simulink models as SystemVerilog modules taking advantage of the DPI-C interface.

You can analyze IC simulation results with MATLAB to visualize data more effectively and to further refine behavioral models using optimization, machine learning, or deep learning techniques.

The final level of mixed-signal verification is device testing. At this stage, MATLAB and Simulink integrate with a variety of test equipment, enabling you to build test systems that create test vectors via models, control test equipment, and analyze the results.


Phase-Locked Loops (PLL)

Transistor-level models are accurate, but extremely slow when it comes to phase-locked loop (PLL) design. The feedback loop often requires long simulations to capture the lock time and small simulation time steps to accurately predict the phase-noise effects. Simulink and Mixed-Signal Blockset use a variable step solver that results in very fast PLL simulation without the need of oversampling.

With its control design heritage, Simulink has a simulation engine that is extremely efficient at simulating systems with feedback loops. The combination of behavioral modeling and a faster approach to simulation enables engineers to cut simulation times for PLL designs from days to hours or minutes.


Data Converters (ADC/DAC)

The ability to rapidly simulate continuous- and discrete-time signals is key to the design and verification of analog-to-digital converters (ADCs). Because Simulink allows modeling in the same environment analog and digital hardware, you can design an ADC in a fraction of the time required by SPICE tools.

Doing rapid ADC design with Simulink enables faster parameter sweeps, allowing engineers to perform detailed verification in less time. Using Mixed-Signal Blockset testbenches, you can rapidly assess integral and differential nonlinearity and noise performance.


SerDes and High-Speed Links

The analysis and simulation of SerDes serial and DDR parallel equalization systems operating at high data rates can slow simulations to a crawl, which threatens project delivery times and limits the scope for design exploration.

The SerDes Designer app allows you to analyze arbitrary high-speed channel equalization schemes in a matter of minutes, including different architectures for pre-emphasis and equalization, using either NRZ or PAM4 signals. From the app, you can automatically generate Simulink models to further refine the adaptive equalization algorithms, or you can start from your own model and add your proprietary algorithms. For system integration and channel verification, you can automatically generate dual IBIS-AMI models using SerDes Toolbox.


Digital Predistortion (DPD) of RF Power Amplifiers

Digital predistortion is simple in theory, but difficult in practice. MATLAB provides a unifying environment for controlling test equipment, analyzing complex data, and building algorithms for DSPs or FPGAs while having an in-depth understanding of the effects introduced by RF power amplifiers (PA).

In MATLAB, you can easily build a PA model based on modified Volterra series, including memory and non-linearity, and simulate it using RF Blockset™ Circuit Envelope. Simulating the RF PA in closed loop with your own DPD algorithm allows you to estimate timing, quantization, and additional RF effects before going to the lab.