Knowledge Base

Embedded Systems & IoT Engineering Glossary

A comprehensive reference of technical terms in embedded systems, IoT infrastructure, PCB design, robotics, and firmware engineering — defined by practising engineers at our Thiruvananthapuram laboratory. This glossary is maintained as a living document and updated as technologies evolve.

Browse by Category

Glossary

Embedded Systems Terms

Microcontroller (MCU)

A compact integrated circuit designed to govern a specific operation in an embedded system. Combines a processor core, memory (RAM/Flash), and programmable input/output peripherals on a single chip. Common families include ARM Cortex-M (STM32, NXP, TI), ESP32 (Espressif), AVR (Microchip), and PIC (Microchip). Unlike a microprocessor (MPU), an MCU is self-contained and optimised for low-power, real-time control applications.

→ Related: Embedded Systems Development | STM32 vs ESP32 Guide

ARM Cortex-M

A family of 32-bit RISC processor cores licensed by ARM Holdings, purpose-built for microcontroller applications. The Cortex-M series ranges from the ultra-low-power Cortex-M0/M0+ to the DSP-capable Cortex-M4/M7. Key features include a nested vectored interrupt controller (NVIC), optional memory protection unit (MPU), and a deterministic instruction set. Cortex-M cores are the de facto standard in professional embedded development, used in over 90% of 32-bit MCUs shipped globally.

→ Related: Embedded Systems | Cortex-M Power Optimisation

STM32

A family of 32-bit ARM Cortex-M microcontrollers manufactured by STMicroelectronics. STM32 MCUs span from the ultra-low-power STM32L0/L4 series (ideal for battery-powered IoT sensors) to the high-performance STM32H7 series (suitable for real-time DSP and motor control). STM32 is widely adopted in industrial, medical, and IoT applications due to its rich peripheral set, comprehensive HAL/LL library ecosystem, and strong community support. At Hexcode Plus, STM32 is our primary MCU platform for professional embedded development.

→ Related: Firmware Development | STM32 vs ESP32

ESP32

A low-cost, low-power system-on-chip (SoC) from Espressif Systems featuring integrated Wi-Fi and dual-mode Bluetooth. Built around a Tensilica Xtensa LX6 or RISC-V processor, the ESP32 is popular for IoT prototyping and connected devices. While excellent for Wi-Fi-enabled applications, it consumes significantly more power than an STM32 in active mode and offers less deterministic real-time performance. For battery-powered deployments without Wi-Fi requirements, an STM32 + external radio is often the better engineering choice.

→ Related: IoT Solutions | MCU Comparison

Real-Time Operating System (RTOS)

An operating system designed for applications that must process data and respond to events within strictly defined time constraints. Unlike general-purpose OSes (Linux, Windows), an RTOS guarantees deterministic task scheduling with bounded interrupt latency. FreeRTOS is the most widely deployed RTOS in embedded systems, running on over 40 architectures. Other options include Zephyr (Linux Foundation), ThreadX (Eclipse), and embOS (SEGGER). An RTOS enables concurrent task management, inter-task communication (queues, semaphores, mutexes), and software timer services without sacrificing real-time behaviour.

→ Related: Embedded Systems | Firmware Development

FreeRTOS

An open-source, market-leading real-time operating system for microcontrollers and small microprocessors. FreeRTOS provides a preemptive or cooperative multitasking kernel, queues, semaphores, mutexes, software timers, and event groups. It is distributed under the MIT licence, ported to 40+ architectures, and integrated into AWS IoT services. FreeRTOS is the default RTOS in STM32CubeIDE and ESP-IDF, making it the natural choice for most embedded projects. Its minimalist kernel can run in under 10 KB of Flash and under 1 KB of RAM.

→ Related: Firmware Services | Embedded Development

Glossary

IoT & Connectivity Terms

MQTT (Message Queuing Telemetry Transport)

A lightweight, publish-subscribe messaging protocol designed for constrained devices and low-bandwidth, high-latency networks. MQTT uses a central broker to route messages between publishers and subscribers based on topic strings. It is the dominant protocol in IoT due to its minimal overhead (2-byte fixed header), support for QoS levels (0/1/2), persistent sessions, and last-will testament. MQTT 5.0 added session expiry, topic aliases, and request-response patterns. Commonly used brokers include Mosquitto, EMQX, and HiveMQ.

→ Related: IoT Solutions | IIoT in Manufacturing

LoRaWAN (Long Range Wide Area Network)

A low-power, wide-area (LPWA) networking protocol designed for battery-operated devices communicating over kilometres-range distances. LoRaWAN uses chirp spread spectrum modulation on the physical layer (LoRa) and a star-of-stars topology managed by gateways and a network server. In India, LoRaWAN operates in the delicensed 865–867 MHz ISM band, making it accessible without spectrum licensing fees. Key parameters: spreading factor (SF7–SF12) trades data rate for range; adaptive data rate (ADR) optimises per-node performance. Typical rural range: 8–15 km with line-of-sight.

→ Related: IoT Infrastructure | LoRaWAN vs NB-IoT

Edge Computing

A distributed computing paradigm that brings computation and data storage closer to the source of data generation — i.e., at the edge of the network — rather than relying solely on a centralised cloud. In IoT, edge computing means running analytics, filtering, and decision logic on gateways or on-device processors (STM32, ESP32, Raspberry Pi, Jetson) before sending aggregated results to the cloud. Benefits include reduced bandwidth costs, lower latency (sub-millisecond local decisions), and continued operation during connectivity loss.

→ Related: IoT Engineering | Embedded Systems

Modbus RTU/TCP

An open, serial communication protocol developed in 1979 for industrial automation and still dominant on factory floors today. Modbus RTU runs over RS-485 physical layer with a master-slave architecture. Modbus TCP encapsulates the same protocol in TCP/IP packets for Ethernet connectivity. It is the most common interface for retrofitting legacy industrial machinery with IoT monitoring because almost every PLC, VFD, and industrial sensor supports it. A typical IIoT retrofit reads Modbus registers from existing equipment and bridges them to MQTT for cloud ingestion.

→ Related: IIoT Retrofitting Guide | IoT Solutions

NB-IoT (Narrowband IoT)

A licensed-spectrum LPWAN technology standardised by 3GPP (Release 13) that operates within existing cellular network infrastructure. Unlike LoRaWAN (which uses unlicensed spectrum), NB-IoT requires a carrier subscription (Jio, Airtel, Vodafone Idea in India). NB-IoT offers higher data rates (~250 kbps) and deeper indoor penetration than LoRaWAN, at the cost of higher power consumption and per-device SIM fees. Best suited for urban deployments where cellular coverage exists and power constraints are less critical.

→ Related: LPWAN Comparison | IoT Solutions

AWS IoT Core / Azure IoT Hub

Managed cloud services from Amazon and Microsoft respectively that enable secure, bidirectional communication between IoT devices and the cloud. Both provide device provisioning (X.509 certificate-based), message routing (to databases, analytics, Lambda/Azure Functions), device shadow / digital twin, and over-the-air update orchestration. AWS IoT Core uses MQTT natively; Azure IoT Hub supports MQTT, AMQP, and HTTP. At Hexcode Plus, we integrate with both platforms depending on the client's existing cloud infrastructure.

→ Related: IoT Cloud Integration | Firmware & OTA

Glossary

PCB Design Terms

Controlled Impedance

The practice of designing PCB traces with a specific characteristic impedance (typically 50Ω for single-ended signals, 90–100Ω for differential pairs). Impedance is controlled by trace width, trace-to-reference-plane distance, and dielectric constant of the PCB substrate. Controlled impedance is mandatory for high-speed digital interfaces (USB, HDMI, DDR, PCIe) and RF circuits to prevent signal reflections that cause data corruption. Requires close coordination with the PCB fabricator, who must hold specified impedance within ±10% tolerance.

→ Related: PCB Design | High-Speed PCB Guide

EMI/EMC (Electromagnetic Interference / Compatibility)

EMI is unwanted electromagnetic energy emitted by a circuit that can disrupt nearby electronics. EMC is the ability of a device to function correctly in its electromagnetic environment without causing or suffering interference. PCB-level EMC design techniques include: uninterrupted ground planes, stitching vias along board edges, guard traces around sensitive signals, ferrite beads on power inputs, and keeping high-speed traces short with proper return paths. Regulatory compliance (FCC, CE, BIS in India) requires passing radiated and conducted emissions tests.

→ Related: PCB Services | Signal Integrity Guide

IPC Standards (IPC-2221/2222)

Industry standards published by the Association Connecting Electronics Industries (IPC) governing PCB design, fabrication, and assembly. IPC-2221 is the generic standard on printed board design. IPC-2222 covers rigid organic printed boards. These standards define design rules for conductor spacing, hole sizes, annular rings, and thermal relief. Fabrication to IPC Class 2 (dedicated service electronics) or Class 3 (high-reliability electronics) determines inspection criteria and acceptance quality levels. All Hexcode Plus PCB designs are produced to IPC-2221/2222 Class 2 minimum.

→ Related: PCB Design & Fabrication

Differential Pair Routing

A PCB routing technique where two traces carry equal and opposite signals, routed parallel to each other with tight coupling. Differential signalling (USB, HDMI, Ethernet, LVDS, CAN) rejects common-mode noise, reducing EMI susceptibility. Key rules: maintain equal length between the pair (length matching within 5 mils for high-speed), keep constant spacing along the entire route, and ensure both traces reference the same ground plane. The differential impedance must match the specification (typically 90Ω for USB, 100Ω for Ethernet).

→ Related: PCB Engineering | High-Speed Design

DFM (Design for Manufacturability)

The practice of designing PCBs so that they can be reliably and cost-effectively manufactured at scale. DFM checks include: minimum trace width and spacing (typically 5–6 mil for standard fabs), adequate annular rings around vias, solder mask sliver prevention, proper component-to-edge clearance, thermal relief on pads connected to large copper pours, and silkscreen legibility. A design that passes electrical DRC but fails DFM review will face fabrication delays, yield issues, or costly respins.

→ Related: PCB Fabrication | Firmware for Manufactured Boards

Gerber Files

The universal file format for PCB fabrication data. Each Gerber file represents one layer of the board: copper layers (top, inner, bottom), solder mask (top/bottom), silkscreen (top/bottom), solder paste (top/bottom), and board outline. Gerber RS-274X is the current standard. A complete fabrication package includes Gerber files, an NC drill file (for hole locations and sizes), a fabrication drawing (PDF), and a bill of materials (BOM). These are the deliverables Hexcode Plus provides to clients and fabricators.

→ Related: PCB Design Services

Glossary

Robotics Terms

SLAM (Simultaneous Localisation and Mapping)

An algorithmic technique that enables a robot to build a map of an unknown environment while simultaneously tracking its own location within that map. SLAM fuses data from LiDAR, cameras, IMUs, and wheel odometry using probabilistic filters (Extended Kalman Filter, particle filter) or graph-based optimisation. In modern robotics, SLAM is implemented through ROS 2 packages like slam_toolbox and cartographer. SLAM is foundational for autonomous mobile robots (AMRs) operating in warehouses, hospitals, and factories.

→ Related: Robotics Engineering | AMR with ROS 2

ROS 2 (Robot Operating System 2)

The second generation of the Robot Operating System — an open-source middleware framework for building robot applications. ROS 2 uses DDS (Data Distribution Service) for communication, providing real-time publish-subscribe messaging between nodes. Key improvements over ROS 1: real-time support, multi-robot coordination, security (DDS-Security), and production-grade reliability. The Nav2 stack builds on ROS 2 for autonomous navigation, and MoveIt 2 provides motion planning for robotic manipulators.

→ Related: Robotics Services | ROS 2 AMR Guide

Sensor Fusion

The process of combining data from multiple sensors (LiDAR, camera, IMU, GPS, wheel encoders) to produce more accurate and reliable state estimation than any single sensor could provide alone. Sensor fusion algorithms — typically Extended Kalman Filters (EKF) or Unscented Kalman Filters (UKF) — compensate for individual sensor weaknesses: LiDAR is accurate but slow, IMU is fast but drifts, GPS is absolute but imprecise. The robot_localization package in ROS 2 is a widely used EKF-based sensor fusion implementation.

→ Related: Robotics Engineering | Sensor Integration

Path Planning & Navigation (Nav2)

The computational pipeline that enables an autonomous robot to move from its current position to a goal location while avoiding obstacles. Nav2 (Navigation 2) is the ROS 2 framework that ties together: a global planner (computes the optimal path through the known map), a local planner/controller (generates velocity commands to follow the path while avoiding dynamic obstacles), a costmap (represents the environment as a grid of traversability costs), and behaviour trees (orchestrates recovery behaviours when the robot gets stuck).

→ Related: AMR Development | Autonomous Navigation Guide

LiDAR (Light Detection and Ranging)

A sensing technology that measures distance by illuminating a target with laser light and measuring the reflected signal. In robotics, 2D LiDAR (e.g., RPLIDAR, YDLIDAR, Slamtec) provides a planar scan of the environment at ranges of 12–40 metres, while 3D LiDAR (e.g., Velodyne, Ouster) captures full point clouds. 2D LiDAR is the primary perception sensor for warehouse AMRs; it provides the distance measurements that feed into SLAM and obstacle avoidance pipelines. Typical cost for a usable 2D LiDAR: ₹15,000–₹80,000 depending on range and sample rate.

→ Related: Robotics Hardware | AMR Sensor Stack

Odometry

The use of data from motion sensors (wheel encoders, IMU, visual odometry) to estimate a robot's change in position over time. Wheel odometry integrates encoder counts to estimate distance travelled and heading change — but it drifts due to wheel slip, uneven floors, and encoder resolution limits. Visual odometry uses camera feeds to track features between frames. In practice, odometry is fused with absolute references (LiDAR scan matching, GPS) through sensor fusion to produce a drift-corrected pose estimate. Bad odometry is the #1 reason SLAM fails in real-world deployments.

→ Related: Robotics Engineering | Odometry Deep Dive

Glossary

Firmware Engineering Terms

Bootloader

A small programme that runs at microcontroller startup, responsible for initialising hardware and deciding whether to launch the main application or enter firmware-update mode. A secure bootloader cryptographically verifies the application firmware's signature before executing it, preventing unauthorised or corrupted code from running. Dual-bank bootloaders store two firmware images in separate flash partitions — the new image is written and verified in the inactive bank, then atomically swapped on reboot. If the new image fails, the bootloader falls back to the known-good image.

→ Related: Firmware Development | Embedded Systems

OTA (Over-the-Air) Update

The mechanism by which embedded device firmware is updated remotely over a wireless connection — Wi-Fi, cellular, or LoRaWAN. A robust OTA pipeline includes: firmware image compression and encryption, differential/delta updates (sending only changed bytes to minimise bandwidth), integrity verification (CRC-32 or SHA-256 checksums), cryptographic signature verification, dual-bank flash swapping for atomic commit, and rollback protection. OTA is critical for deployed IoT fleets where physical access for manual firmware updates is impractical or impossible.

→ Related: Firmware Services | IoT Infrastructure

Bare-Metal Programming

Writing firmware that runs directly on the microcontroller hardware without an operating system abstraction layer. The developer manipulates hardware registers, manages interrupts, and implements the main control loop manually. Bare-metal programming offers the lowest possible latency (interrupt response in single-digit clock cycles), minimal flash/RAM footprint (a blinky can fit in under 1 KB), and full determinism. It is the right choice for simple, hard-real-time applications. For complex multi-task systems, an RTOS is generally more maintainable.

→ Related: Firmware Development | Embedded Programming

HAL (Hardware Abstraction Layer)

A software layer that provides a consistent API for interacting with microcontroller peripherals (GPIO, UART, SPI, I2C, ADC, timers) regardless of the underlying chip family or vendor. STM32Cube HAL, for example, abstracts the register-level differences between STM32F4 and STM32L0 into a uniform set of functions. A well-designed HAL makes firmware portable across MCU families and simplifies code maintenance. At Hexcode Plus, we write HAL-conformant driver code that can be migrated between STM32, ESP32, and other ARM Cortex-M platforms with minimal rework.

→ Related: Driver Development | Embedded Systems

MISRA-C

A set of software development guidelines for the C programming language developed by the Motor Industry Software Reliability Association (MISRA). MISRA-C:2012 defines 143 rules and 16 directives that restrict C language usage to a safer subset — forbidding constructs prone to undefined behaviour, memory corruption, and security vulnerabilities. Compliance is mandatory in automotive (ISO 26262), medical (IEC 62304), and industrial safety (IEC 61508) firmware. Hexcode Plus firmware is developed to MISRA-C:2012 guidelines with static analysis validation.

→ Related: Firmware Standards | Engineering Services

Watchdog Timer

A hardware timer that resets the microcontroller if the firmware fails to periodically reset (or "kick") the timer within a configured interval. The watchdog provides a last-resort recovery mechanism against firmware hangs caused by infinite loops, deadlocks, memory corruption, or unhandled exceptions. Independent watchdog (IWDG) runs from its own RC oscillator, so it continues counting even if the system clock fails. Window watchdog (WWDG) requires the kick to occur within a specific time window — kicking too early or too late both trigger a reset, catching runaway code that executes faster than expected.

→ Related: Firmware Reliability | Embedded Design

Need Engineering Expertise?

This glossary is maintained by the engineering team at Hexcode Plus R&D. If you need hands-on development in embedded systems, IoT, PCB design, robotics, or firmware — we can help.

DISCUSS YOUR PROJECT