Embedded RoT Design: Choosing Between MCU and FPGA for Hardware Security

In a connected-device ecosystem where embedded systems increasingly operate at the edge, in industrial automation, IoT gateways, and critical infrastructure, cybersecurity is no longer optional — it’s foundational. Among the most important architectural decisions for product security is how to implement a hardware root of trust (RoT). Should this be embedded in a lightweight secure MCU, or should it be realised in an FPGA fabric? The choice influences cost, power, flexibility and long-term resilience. In this article, we compare MCU-based and FPGA-based RoT implementations, analyse key decision criteria, examine use-cases and provide guidance for embedded security architects.

Why the Hardware Root of Trust Matters

Embedded devices often form the first line of interaction with the physical world. They control actuators, monitor sensors, manage connectivity and often endure long lifecycles. That makes them attractive targets for attackers seeking to infiltrate supply chains, extract IP, compromise device integrity or pivot into higher value assets.

A hardware root of trust — a minimal, immutable hardware component that anchors device identity, attests firmware integrity, controls secure boot and manages cryptographic operations — becomes the foundation of trust for the entire device. Without a robust RoT, software-only protections are undermined by physical or side-channel attacks, firmware rollback vulnerabilities and long-term key-exposure threats.

When designing embedded systems, the decision of MCU-based RoT vs FPGA-based RoT can thus determine whether the device is secure by design or security by after-thought.

MCU-based Security Approach

Benefits

Lower unit cost and lower power consumption: Secure MCUs from vendors such as NXP or Infineon provide integrated cryptographic engines, secure boot, key storage and built-in RoT at much lower cost and power than larger programmable fabrics.
Faster time to market: Established toolchains, libraries, vendor support and known hardware ecosystems make it quicker to embed RoT into an MCU-based design.
Simplicity of development: With MCUs, firmware development follows traditional MCU patterns, reducing the need for HDL or custom logic design.

Limitations

Fixed hardware logic: The MCU architecture defines the cryptographic engines, memory layout and I/O. For advanced or evolving threat models (e.g., custom crypto, side-channel countermeasures) flexibility is limited.
Lower parallel-processing ability: MCUs often cannot absorb complex side-channel counter-logic, high-throughput encryption or real-time monitoring of physical monitors (voltage, current, magnetic) as reconfigurable hardware might.
Less readiness for long device lifecycles and emerging threats: Embedded devices often operate 5-10+ years; the hardware logic of an MCU-based RoT may become obsolete or vulnerable to quantum-era threats if not designed for agility and future proofing.

FPGA-based Security Approach

Benefits

Customisable hardware root of trust: FPGAs enable designers to implement custom logic for cryptography, authentication, key-management, secure update mechanisms and side-channel protections (e.g., power balancing, masking, PUFs).
Higher resistance to side-channel attacks and custom physical threats: Because logic is implemented directly in hardware fabric, designers can adopt advanced countermeasures that MCUs may not support.
Flexible firmware monitoring, rollback protection and post-quantum readiness: FPGA fabric allows the integration of runtime monitors, rollback detectors, secure update state machines and even post-quantum cryptography cores that can be upgraded as standards evolve.

Limitations

Higher cost and power: FPGAs traditionally consume more power and cost more than MCUs. That can impact deeply embedded, battery-constrained designs.
Higher complexity and expertise requirement: FPGA security logic demands hardware-design skills (HDL, timing closure, side-channel design) beyond standard MCU firmware development.
Longer development time: Customising FPGA logic for a RoT adds design, verification and qualification time compared to using secure MCU building blocks.

Decision Factors: Which Approach Fits What Situation?

When selecting MCU or FPGA for RoT, embedded architects should assess the following dimensions:

Device class and lifecycle: If the device is low-cost, battery-powered, with a short lifecycle (1–3 years) and limited connectivity, an MCU-based RoT may suffice. If the device is an edge gateway, industrial controller or a device with 5-10+ year lifecycle, FPGA-based RoT offers longer-term flexibility.
Threat model: For simple secure boot and basic cryptography, MCU solutions are appropriate. For high-value assets, exposed physical access, side-channel threat risk or post-quantum readiness, FPGA is a stronger choice.
Upgradeability and field update: If over-the-air (OTA) updates, reconfigurable functionality or future-proofing (e.g., PQC) matter, FPGA logic provides more headroom.
Cost and power budget: Battery-driven endpoints or cost-sensitive mass-markets will favour MCU. Edge systems with power budget and performance headroom can adopt FPGA.
Ecosystem and skills: Using secure MCUs leverages existing firmware toolchains and vendor support. FPGA-based RoT requires hardware logic design, side-channel expertise and careful verification flows.

Hybrid and Combined Approaches

It is increasingly common to see hybrid architectures, where a secure MCU provides baseline RoT and an FPGA complements with advanced security logic or hardware acceleration. For example, an MCU handles system configuration and life-cycle management, while an FPGA fabric implements custom cryptographic blocks, dynamic monitoring, PUF-based identity or real-time side-channel immunity. Such co-designs allow balancing cost/power and flexibility. This aligns with observations that modern security solutions can “have your cake and eat it too” by combining benefits of both platforms. Embedded Computing Design+1

Use Cases: Practical Illustration

Low-cost IoT sensor endpoint: Battery-powered device with simple secure boot and encrypted telemetry — MCU-based RoT is appropriate.
Industrial edge gateway: Located in production environment, connected to networks, susceptible to physical attack, long lifecycle — FPGA-based RoT offers custom logic, side-channel protections and upgradeable crypto.
Telecom network appliance: High value, long service horizon, field-reconfigurable protocols — hybrid MCU+FPGA RoT allows future-proofing.
Automotive domain controller: Needs functional safety and cybersecurity for 10–15 years — FPGA might be preferred if heavy custom security logic is needed; MCU may suffice for simpler subsystems.

Implementation Considerations for Designers

Secure boot chain and authentication: The RoT must authenticate firmware via immutable hardware roots; for FPGA this may include encrypted bitstreams and runtime verification.
Key management and PUFs: FPGAs support hardware PUFs (Physical Unclonable Functions), enabling unique chip identity; MCU solutions may rely on vendor secure elements. Wikipedia
Side-channel counter-measures: In FPGA implementations, designers can implement power noise, asynchronous logic, dynamic reconfiguration; MCU solutions rely on vendor-integrated countermeasures.
Post-quantum readiness: FPGA fabric can accommodate updated PQC cores; MCU cores may be fixed in silicon and become obsolete faster.
Certification & lifecycle management: Both approaches must integrate OTA update paths, rollback protection, hardware-rooted identity, and remote attestation frameworks (e.g., PSA Certified). Wikipedia
Verification and manufacturing cost: FPGA logic requires extensive verification (timing, side-channel leakage), increasing R&D cost and time; MCU platforms benefit from pre-certified security building blocks.

When the Trade-Offs Tip the Balance

Choose MCU-based RoT when cost/power are dominant constraints, the threat model is modest, and time-to-market is critical.
Choose FPGA-based RoT when longevity, flexibility and high security are non-negotiable — especially in industrial, automotive, telecom or infrastructure domains.
Consider a hybrid approach when you need both: secure baseline via MCU, custom logic and future security headroom via FPGA.

In many real designs, what matters is not “which is better overall” but “which is better for this device under this threat model”.

AI Overview: MCU-based vs FPGA-based Hardware Root of Trust

The decision between MCU- and FPGA-based hardware roots of trust (RoT) is central to embedded cybersecurity: MCUs offer low cost, low power and quick deployment, while FPGAs provide customisable logic, side-channel resistance and long-term flexibility.

Key Applications: secure IoT endpoints, industrial gateways, automotive controllers, telecom infrastructure, edge-security controllers.
Benefits: improved device trust, hardware-anchored identity, firmware integrity, advanced cryptography, future-proofing.
Challenges: balancing cost vs performance, developing hardware security logic, managing power and lifecycle, verifying complex designs, ensuring upgradeability.
Outlook: over the next 3–5 years, devices will increasingly embed flexible RoT logic — FPGA or hybrid — enabling post-quantum readiness and dynamic threat adaptation.
Related Terms: hardware root of trust, secure MCU, secure FPGA, side-channel protection, PUF, post-quantum cryptography, PSA Certified, firmware rollback protection.