Building Next-Generation Software-Defined Radios with RFSoC System-on-Modules

Software-defined radio (SDR) represents one of the most significant transformations in the field of wireless communication. Unlike conventional radios that rely on fixed analog circuitry for filtering, mixing, and modulation, SDRs shift much of the processing to the digital domain. By replacing hardware-centric functions with software-driven algorithms, SDRs gain an unmatched level of flexibility allowing designers to upgrade features, adapt to new protocols, and extend system lifecycles without redesigning the hardware.

This ability to reconfigure on the fly makes SDRs indispensable in a broad spectrum of applications, from defense systems and aerospace to 5G infrastructure, satellite communications, and electronic test equipment.

How SDR differs from traditional radio systems

In a traditional RF receiver, analog components handle most of the workload: mixers down-convert incoming signals, filters shape the spectrum, and modulators or demodulators recover information. This analog chain can be inflexible and susceptible to noise, requiring redesign for each new frequency band or standard.

By contrast, an SDR reduces the analog front end to the bare minimum—typically just the antenna and a basic RF front-end circuit (Figure 1). Once the incoming waveform is digitized by an analog-to-digital converter (ADC), the heavy lifting is performed in software. Modulation, demodulation, channel filtering, error correction, and decoding all happen digitally. Similarly, for transmission, a digital-to-analog converter (DAC) transforms processed data back into RF signals, again controlled by software routines.

Figure 1: The basic SDR processes. (Image source: iWave Global)

This shift unlocks tremendous agility: the same radio hardware can support Wi-Fi today, a 5G band tomorrow, and secure tactical communications the next—all with a software update.

RFSoC: An ideal platform for SDR

Building a high-performance SDR requires ultra-fast converters, a powerful processing fabric, and low-latency data paths. AMD’s Zynq™ UltraScale+™ RFSoC family addresses these needs by integrating:

• Multi-gigasample RF-ADCs and RF-DACs
• FPGA programmable logic for real-time DSP
• Embedded Arm® processors for software control
• High-speed memory and transceiver interfaces

By consolidating what previously required multiple discrete chips into a single device, RFSoC dramatically simplifies board design. This integration lowers power consumption, reduces latency, and improves signal integrity. For real-time RF applications where timing precision and performance are non-negotiable, RFSoC delivers a monolithic solution with ultra-low latency and tight synchronization.

The power of direct RF sampling

One of RFSoC’s defining advantages is its ability to support sampling rates in the multi-GSPS range. Its RF-ADCs can capture signals directly at RF frequencies, while its RF-DACs can generate extremely wideband outputs—both without relying on intermediate down-conversion stages.

This enables an “almost all-digital” radio architecture, where standards like Wi-Fi at 2.4 GHz, 5G New Radio around 3.5 GHz, and cellular bands from 800 MHz to 1.8 GHz can be directly digitized and processed. By contrast, many off-the-shelf SDR platforms are limited to sampling rates of a few tens or hundreds of MHz, making them dependent on analog mixers to shift signals down to an intermediate frequency.

By eliminating those analog stages, RFSoC-based SDRs achieve higher fidelity, lower latency, and a more compact design (Figure 2).

Figure 2: Comparison of a single-chip RFSoC SDR solution, with multi-chip alternatives. (Image source: Software-Defined Radio with Zynq® UltraScale+™ RFSoC)

Comparing SDR architectures: single-chip vs multi-chip

The benefits of RFSoC integration become clear when compared to conventional SDR architectures (Table 1).

Table 1: Comparison of RFSoC to conventional SDR solutions.

With ADCs, DACs, FPGA logic, and processors all inside one package, RFSoC avoids the pitfalls of inter-chip communication. For developers, this translates to shorter design cycles, reduced cost, and superior end performance.

Why choose a system-on-module for RFSoC SDRs?

While RFSoC itself is highly integrated, designing a custom board around it can still be daunting. Power sequencing, clock distribution, and multi-gigabit layout require advanced expertise. A system-on-module (SoM) provides a practical solution.

By delivering a compact, pre-validated module that houses the RFSoC, memory, power management, and high-speed interfaces, SoMs let engineers:

• Accelerate prototyping and minimize design risk
• Focus on application-specific innovation rather than baseboard integration
• Achieve compact, SWaP-optimized (size, weight, and power) designs suited for aerospace and defense
• Rely on long-term availability and production-grade quality

Carrier boards can be tailored to each use case while the SoM remains constant, allowing teams to reuse intellectual property (IP) and reduce total development cost.

Figure 3: iWave carrier board for RFSoC SDRs. (Image source: iWave)

iWave’s RFSoC system-on-module portfolio

iWave offers a comprehensive set of RFSoC SoMs and evaluation platforms, each tuned for high-performance SDR and RF applications:

• iG-G42M – ZU49/ZU39/ZU29DR RFSoC SoM
  • Features 16 ADCs (2.5 GSPS) and 16 DACs (10 GSPS).
• iG-G42P – RFSoC PCIe Card (ZU49/ZU39/ZU29DR)
  • PCIe Gen3 connectivity, NVMe storage, SMA I/O, and FMC+ expansion.
• iG-G60M – ZU48/47/43/28/27/25DR RFSoC SoM
  • Up to 8-channel ADC/DAC (5 GSPS / 9.85 GSPS).
• iG-G60V (Coming Soon) – RFSoC ADC/DAC 3U VPX Plug-in Module
  • Ruggedized form factor for aerospace and defense.

Figure 4: iWave RFSoC SoMs. (Image source: iWave)

These modules are backed by robust software stacks including Linux BSPs, JESD204B/C support, GStreamer pipelines, and reference applications, ensuring a seamless path from prototyping to production.

Real-World Impact of RFSoC SDRs

The combination of direct RF sampling, integrated digital processing, and module-level deployment results in SDR systems that are:

• Highly flexible – configurable for multiple wireless standards
• Compact and efficient – optimized for SWaP-sensitive platforms
• High fidelity – with minimal signal degradation
• Scalable – from lab prototypes to deployed defense and telecom infrastructure

Whether in unmanned aerial systems conducting real-time surveillance, 5G base stations supporting dynamic spectrum allocation, or portable test equipment analyzing wideband signals, RFSoC SDRs enable solutions that were once impractical with discrete designs.

Conclusion

Software-defined Radio is reshaping wireless communication by making radios more flexible, upgradeable, and future-ready. AMD’s Zynq UltraScale+ RFSoC brings this concept to life by collapsing converters, FPGA fabric, and processors into one silicon die. Pairing RFSoC with a System-on-Module unlocks faster time-to-market, reduced risk, and production-grade reliability.

With over 25 years of expertise in FPGAs and embedded systems, iWave delivers RFSoC SoMs and ODM services that balance performance, cost, and long-term support.

To explore how iWave’s RFSoC portfolio can accelerate your SDR projects, contact us at mktg@iwave-global.com.