IoT Protocols Compared: MQTT, CoAP, HTTP/2, WebSocket - Performance, Overhead, and Use Cases | Embedded Systems Hub

Introduction Technical comparison of MQTT 3.1.1/5.0, CoAP, HTTP/2, and WebSocket for IoT: message overhead, QoS, connection patterns, and protocol selection matrix. This article provides a comprehensive technical deep-dive into the subject, covering theoretical foundations, practical implementation strategies, and real-world engineering trade-offs encountered in production embedded systems. Core Concepts Fundamental Principles Understanding the fundamental principles is essential before diving into implementation. This section covers the theoretical underpinnings that drive design decisions in embedded systems engineering. Key Terminology MCU/MPU: Microcontroller Unit / Microprocessor Unit - the compute engine RTOS: Real-Time Operating System - deterministic task scheduling HAL: Hardware Abstraction Layer - portable hardware interface BSP: Board Support Package - hardware-specific initialization OTA: Over-The-Air - remote firmware updates System Constraints Embedded systems operate under strict constraints that fundamentally shape architecture: Constraint Typical Range Impact on Design Flash Memory 64 KB - 2 MB Code size optimization, compression RAM 8 KB - 512 KB Static allocation, pool allocators CPU Frequency 16 MHz - 400 MHz Algorithm complexity limits Power Budget µA - mA range Sleep modes, duty cycling Real-time Deadlines µs - ms Deterministic scheduling required Implementation Details Architecture Selection Choosing the right architecture depends on system complexity, team expertise, and long-term maintenance requirements. Code Example: Minimal RTOS Task // FreeRTOS task template void sensor_task(void *pvParameters) { TickType_t xLastWakeTime = xTaskGetTickCount(); const TickType_t xFrequency = pdMS_TO_TICKS(100); // 10 Hz for (;;) { // Read sensor data sensor_data_t data = read_sensor(); // Process with filter filtered_data_t filtered = apply_filter(data); // Send to queue for logging task xQueueSend(sensor_queue, &filtered, 0); // Precise periodic timing vTaskDelayUntil(&xLastWakeTime, xFrequency); } } // Task creation xTaskCreate(sensor_task, "Sensor", 1024, NULL, tskIDLE_PRIORITY + 2, NULL); Memory Management Strategies Static allocation preferred over dynamic to avoid fragmentation and enable compile-time verification: // Static memory pool for zero-copy networking #define POOL_SIZE 16 #define BUF_SIZE 1536 static uint8_t tx_pool[POOL_SIZE][BUF_SIZE] __attribute__((aligned(4))); static uint8_t rx_pool[POOL_SIZE][BUF_SIZE] __attribute__((aligned(4))); typedef struct { uint8_t *buf; size_t len; bool in_use; } buffer_t; static buffer_t tx_buffers[POOL_SIZE]; static buffer_t rx_buffers[POOL_SIZE]; Best Practices Development Workflow Version Control: Git with conventional commits for traceability Static Analysis: MISRA C, Cppcheck, Clang-Tidy in CI Code Coverage: Target >80% branch coverage for safety-critical Documentation: Doxygen for API, architecture decision records (ADRs) Testing Strategy Unit tests on host (Unity/CMock) - fast feedback Integration tests on target hardware - real peripherals HIL simulation for fault injection - edge cases Long-running stress tests - memory leaks, watchdog Production Readiness Before shipping: enable watchdog, configure brownout detection, implement health monitoring, verify OTA rollback, test power-loss recovery, and validate all error paths. Conclusion This article covered the essential concepts, implementation patterns, and best practices for building robust embedded systems. The key takeaway is that successful embedded engineering requires balancing competing constraints—performance, power, memory, and determinism—while maintaining code quality through disciplined processes. Continue exploring related topics in our articles catalog or dive deeper into specific areas like RTOS internals and IoT connectivity .