Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is Embedded System Security?

Embedded system security protects purpose-built computers inside physical products, where the attacker can hold the hardware, memory is measured in kilobytes, and the code runs for a decade without an administrator.

Illustration for the cybersecurity term Embedded System Security
Home / Blog / Cybersecurity Terms / What Is Embedded System Security?
17 November 2026

What Is Embedded System Security?

What Is Embedded System Security?

An embedded system is a computer built into a product to do one job: the controller in a lift, the board inside a smart meter, the module that drives an infusion pump. Embedded system security is the practice of keeping those computers trustworthy when the attacker can buy the product, open it, and spend a weekend with it. That single fact separates the field from server security, where physical access is assumed to be controlled.

The constraints are real. A microcontroller may have 256 KB of flash and 64 KB of RAM, no memory management unit, and a real-time operating system with no process isolation. There is no room for an endpoint agent, no user to notice odd behaviour, and often no way to collect a log.

Where the Weaknesses Live

Hardware

Debug interfaces are the usual way in. Production boards frequently leave JTAG and UART pads active, giving a root shell or full memory access to anyone with a £20 adapter. External flash is a second route: a SOIC-8 clip and a cheap programmer read an unencrypted SPI chip in seconds, and chip-off analysis recovers the contents of soldered memory when a clip will not do.

Firmware

What comes out of that flash is usually revealing. Hardcoded administrative accounts, private keys embedded in the image, and debug builds left in production are routine findings. A 2015 study of several thousand firmware images found hundreds of unique private keys reused across many thousands of internet-facing devices, which means one extracted key unlocks an entire product line.

Memory safety

Embedded code is overwhelmingly C, and the libraries beneath it are shared by millions of devices. Ripple20, disclosed by JSOF in June 2020, covered 19 vulnerabilities in the Treck TCP/IP stack used in medical, industrial, and aviation equipment, two of them allowing remote code execution. Urgent/11 did the same for the VxWorks IPnet stack in 2019, AMNESIA:33 for four open-source stacks in December 2020, and BadAlloc for memory allocation routines across 25 products in April 2021. In each case the flaw sat in a component the device maker did not write and often could not name.

Defences That Fit the Constraints

Building It into Development

Retrofitting is painful because many of these controls are decisions made at silicon selection. Choose a part with secure boot and key storage before the schematic is finished, do threat modelling that covers physical access, and test an engineering sample before tooling is committed. IEC 62443-4-1 describes a secure development lifecycle for industrial products, and ETSI EN 303 645 sets the consumer baseline that UK law now draws on.

Where UK Rules Bite

Consumer connectable products sold in the UK have been subject to the Product Security and Telecommunications Infrastructure Act 2022 since 29 April 2024, which ends universal default passwords and requires a published vulnerability contact and update period. Medical devices follow the MHRA’s expectations and the relevant UK Conformity Assessed route; industrial equipment is typically covered through the operator’s obligations under the NIS Regulations 2018.

At Privia Security, our Hardware Penetration Test Services work on engineering samples and production units alike, from bus capture to firmware review.

FAQ – Frequently Asked Questions

Is an RTOS More Secure Than Embedded Linux?

Neither is inherently safer. An RTOS has a smaller attack surface but usually no memory isolation between tasks, so one overflow compromises everything. Embedded Linux brings process separation, users, and mature hardening options, along with a far larger codebase to patch. The deciding factor is whether the vendor ships updates and whether you can apply them.

How Do I Find Out What Components My Firmware Contains?

Extract the image and analyse it. Tools such as binwalk, EMBA, and FACT identify filesystems, kernels, and embedded libraries, and matching version strings against vulnerability databases surfaces known issues. For products you buy, ask the supplier for a software bill of materials in CycloneDX or SPDX format and make it a contract condition.

Can I Secure a Device That Has No Secure Boot?

Only up to a point. Without a verified boot chain, an attacker with physical access can replace the firmware and defeat everything above it. You can still reduce exposure by encrypting external flash, locking debug interfaces, keeping secrets in a secure element, and restricting what the device can reach on the network, but the physical attacker wins eventually.

What Does an Embedded Security Assessment Cover?

Board inspection and interface identification, bus capture over SPI and I2C, debug access over JTAG or UART, firmware extraction and static review, key and credential discovery, update mechanism testing, and a look at every network or radio service the device exposes. Findings are then mapped back to the design so fixes land in the next hardware revision.

Related Terms

You May Be Interested In These

What Is Sysmon?

Cybersecurity Terms

17 November 2026

What Is Sysmon?

Read More
What Is a Keylogger?

Cybersecurity Terms

17 November 2026

What Is a Keylogger?

Read More
What Is an Access Review?

Cybersecurity Terms

17 November 2026

What Is an Access Review?

Read More