Harnessing AI at the Edge with iWave’s Agilex™ 5 System on Module

Artificial Intelligence (AI) is rapidly redefining embedded computing, powering next-generation solutions across robotics, industrial automation, and medical devices. As data processing increasingly moves closer to the edge, platforms that deliver efficient AI performance with low latency and high customization are becoming indispensable.

iWave’s iW-RainboW-G58M System on Module (Figure 1), powered by the Altera Agilex™ 5 SoC FPGA, represents a significant leap forward in edge AI design. Built on Altera’s first FPGA architecture with AI capabilities infused directly into the fabric, this compact module delivers a unique combination of flexibility, compute power, and efficiency for AI-driven embedded systems.

Figure 1: The iWave iW-RainboW-G58M System on Module. (Image source: iWave)

Compact Design, Powerful Compute

The iW-RainboW-G58M SoM comes in a 60 mm × 70 mm form factor and supports the Altera Agilex™ 5 SoC FPGA E-Series family with the B32A package. It is available in two groups:

• Group A: A5E065A / 052A / 043A / 028A / 013A SoC FPGA
• Group B: A5E065B / 052B / 043B / 028B / 013B / 008B SoC FPGA

The Agilex™ 5 series integrates an asymmetric heterogeneous processing system, combining dual Arm Cortex-A76 and dual Cortex-A55 cores for an optimized balance of performance and power efficiency. Enhanced DSP blocks with integrated AI Tensor Blocks accelerate deep learning workloads, while GTS transceivers up to 28.1 Gbps, PCIe® 4.0 x 8, and display interfaces such as HDMI and DisplayPort provide high-speed connectivity for complex multimedia and AI applications.

AI Software Ecosystem and Development Tools

The Agilex™ 5 platform integrates seamlessly with leading AI frameworks (Figure 2), simplifying the development of custom AI inference solutions. It supports TensorFlow, PyTorch, and the OpenVINO™ toolkit, which optimizes deep learning models for FPGA inference using Altera’s FPGA AI Suite and Quartus® Prime design software.

Figure 2: The Agilex™ 5 platform integrates seamlessly with leading AI frameworks. (Image source: iWave)

With the Altera FPGA AI Suite, developers from hardware engineers to data scientists can efficiently build and deploy optimized AI pipelines using familiar workflows. This toolkit bridges FPGA design with high-level AI frameworks, accelerating time-to-market while maintaining deterministic performance and hardware-level efficiency.

Cloud vs Edge AI: A Balanced Approach

In modern architectures, AI processing is split between the cloud and the edge (Figure 3). Cloud AI provides massive scalability but often faces latency and privacy limitations. Edge AI, in contrast, performs real-time processing locally enabling instant decision-making, enhanced security, and continuous operation even without network connectivity.

Figure 3: Cloud AI provides massive scalability but often faces latency and privacy limitations. Edge AI performs real-time processing locally, enabling instant decision-making, enhanced security, and continuous operation even without network connectivity. (Image source: iWave)

In a hybrid setup, edge nodes handle time-critical analytics, while data centers aggregate and process larger datasets for trend analysis or predictive modelling. The iW-RainboW-G58M is built to excel in this hybrid paradigm providing both computational agility at the edge and seamless integration with cloud infrastructures.

Development and Longevity Support

iWave provides early access samples of the Agilex™ 5 SoM and its comprehensive evaluation kit, which includes detailed documentation, BSPs, software drivers, and pre-validated reference designs. This helps developers rapidly prototype, evaluate performance, and accelerate product deployment.

The company’s Product Longevity Program guarantees availability for 10+ years, ensuring sustained support for long lifecycle industrial and embedded products. In addition, iWave offers ODM design services—including carrier card design, thermal management, and production support allowing customers to scale from prototype to production efficiently.

iWave is a global embedded systems company specializing in FPGA and SoC-based System on Modules (SoMs) and ODM design services. With over two decades of experience in industrial, medical, automotive, and avionics markets, iWave delivers end-to-end embedded computing platforms that enable customers to bring reliable, high-performance products to market faster and at lower cost.

For more information, visit www.iwave-global.com or contact mktg@iwave-global.com.