Show understanding of embedded systems

3.1 Computers and Their Components – Embedded Systems

1. Definition and Why Embedded Systems Are Needed

An embedded system is a computer system designed to perform a dedicated function (or a small set of functions) within a larger mechanical or electrical device. It is “embedded” because the hardware and software are built into the host product and are usually invisible to the end‑user. Embedded systems are required when a device must operate autonomously, reliably, and often in real‑time while meeting strict constraints on cost, size, power consumption, and durability.

2. Data Representation (Syllabus Requirement 1)

All digital computers, including embedded ones, store and manipulate data in binary form. The following representations are essential for the Cambridge International AS & A Level Computer Science syllabus.

Worked Example – Binary ↔ Decimal conversion

Binary 1101 1010₂ = (1·2⁷)+(1·2⁶)+(0·2⁵)+(1·2⁴)+(1·2³)+(0·2²)+(1·2¹)+(0·2⁰)
                = 128+64+0+16+8+0+2+0 = 218₁₀

3. Hardware & Processor Fundamentals (Syllabus Requirement 3)

This section expands the earlier hardware overview to cover the internal workings of a processor, which are required knowledge for the Cambridge syllabus.

• CPU Registers
  • General‑purpose registers (e.g., R0‑R15) hold temporary data and intermediate results.
  • Special‑purpose registers: Program Counter (PC), Stack Pointer (SP), Status/Flag register (Zero, Carry, Overflow, Interrupt Enable).
• Arithmetic Logic Unit (ALU)
  • Performs integer arithmetic, logical operations (AND, OR, XOR, NOT), shift/rotate.
  • Sets condition flags used by branch instructions.
• Control Unit
  • Generates control signals that orchestrate data movement between registers, ALU, memory, and I/O.
  • Implements the fetch‑decode‑execute cycle.
• Fetch‑Execute Cycle
  1. Fetch: PC provides address → memory read → instruction placed in Instruction Register (IR).
  2. Decode: Control unit interprets opcode, determines required operands.
  3. Execute: ALU or other functional unit carries out operation; result may be stored back.
  4. PC is incremented (or altered by branch/jump).
• Interrupts
  • Asynchronous signals that pause the current instruction stream, save context, and jump to an Interrupt Service Routine (ISR).
  • Essential for real‑time response to sensors, communication peripherals, or timer events.
• Bus Architecture
  • Address bus – carries memory/peripheral addresses (unidirectional).
  • Data bus – carries actual data (bidirectional).
  • Control bus – carries read/write, interrupt, clock signals.
• RISC vs. CISC
  • RISC (Reduced Instruction Set Computer): simple, fixed‑length instructions, many registers, typically pipelined (e.g., ARM Cortex‑M).
  • CISC (Complex Instruction Set Computer): variable‑length, many addressing modes, historically used in x86.
• Pipelining
  • Divides the fetch‑decode‑execute cycle into stages; multiple instructions overlap, increasing throughput.
  • Hazard handling (data, control, structural) is a design concern for deterministic embedded timing.
• Virtual Machines (VMs)
  • Software layer that emulates a hardware architecture (e.g., Java Virtual Machine, .NET CLR).
  • Used in some embedded platforms to provide portability at the cost of performance.
• Boolean Algebra & Logic Circuits
  • Fundamental gates: AND, OR, NOT, NAND, NOR, XOR, XNOR.
  • Truth tables define output for every input combination.
  • Combinational circuits (adders, multiplexers) and sequential circuits (flip‑flops, counters) are built from these gates.

4. Principal Hardware Components of an Embedded System

5. Memory Types – RAM vs. ROM and SRAM vs. DRAM

6. Buffers – Why They Are Needed

A buffer is a temporary storage area that holds data while it is transferred between two components that operate at different speeds or use different protocols. Buffers prevent data loss, smooth timing mismatches and allow the CPU to continue processing while I/O devices complete their operations.

7. Sensors, Actuators and Feedback (Monitoring & Control)

• Sensors (inputs) – Convert a physical quantity (temperature, pressure, light, motion, etc.) into an electrical signal that the CPU can read.
• Actuators (outputs) – Convert an electrical command from the CPU into a physical action (motor rotation, valve opening, LED illumination, etc.).
• Feedback loop – The sensor measures the effect of the actuator; the CPU compares this measurement with the desired set‑point and adjusts the control signal accordingly.
  Example: A thermostat reads a temperature sensor, compares it with the user‑set temperature, drives a heating element, and then re‑reads the temperature to maintain the set point.

8. Role of Secondary (Removable) Storage

Secondary storage provides non‑volatile space that can be added or replaced after manufacture. In embedded systems it is used for:

• Storing large firmware images or OS kernels that do not fit in internal flash.
• Logging data (event logs, sensor recordings) that must survive power loss.
• Enabling field upgrades – the system can load a new program from an SD card or USB stick without specialised programming equipment.

9. Communication & Networks (Syllabus Requirement 2)

Modern embedded devices increasingly communicate with each other or with cloud services. The following concepts satisfy the Cambridge syllabus.

10. Design Process – Step‑by‑Step

1. Define functional requirements and performance constraints (real‑time deadlines, power budget, cost).
2. Select hardware that satisfies the requirements (CPU core, RAM/Flash size, required peripherals, communication interfaces).
3. Choose a development environment and, if needed, an RTOS.
4. Write firmware:
   • Use interrupt‑driven I/O and buffers for deterministic timing.
   • Apply modular programming and, where appropriate, object‑oriented concepts.
5. Integrate sensors, actuators and feedback logic; verify the control loop with hardware‑in‑the‑loop testing.
6. Perform system‑level testing (environmental, reliability, power consumption) and verify compliance with relevant standards (e.g., IEC 60730 for safety‑critical devices).
7. Document the design, produce user manuals, and plan for future firmware upgrades.

11. Embedded vs. General‑Purpose Computers

12. Performance Modelling for Real‑Time Tasks

For each periodic task the worst‑case execution time (WCET) must not exceed its deadline \(D\):

When \(n\) periodic tasks share a single processor, Liu & Layland’s bound for Rate‑Monotonic Scheduling must be satisfied:

where \(C_i\) is the computation time and \(T_i\) the period of task \(i\). The utilisation \(U\) gives a quick check of whether the set of tasks can meet all deadlines.

13. Illustrative Examples of Embedded Systems

• Automotive Engine Control Unit (ECU) – reads multiple sensors, controls fuel injection and ignition timing.
• Digital washing‑machine controller – monitors door lock, water level, temperature; drives motor and valve actuators.
• Smart thermostat – temperature sensor → control algorithm → heating/cooling actuator; communicates with a cloud service for remote control.
• Medical infusion pump – precise flow sensor feedback drives a stepper motor that delivers medication at a programmed rate.
• Industrial PLC (Programmable Logic Controller) – I/O modules, real‑time ladder logic for factory automation.
• Wearable fitness tracker – accelerometer sensor, Bluetooth Low Energy (BLE) communication, on‑board ARM Cortex‑M0+ for activity analysis.

14. Future Trends

• Internet of Things (IoT) – billions of networked sensors and actuators, often powered by energy‑harvesting techniques.
• Edge AI – on‑device machine‑learning inference for vision, speech, predictive maintenance, reducing latency and bandwidth usage.
• Ultra‑low‑power architectures (ARM Cortex‑M0+, RISC‑V) for battery‑operated wearables and medical implants.
• Hardware‑based security – secure boot, TPM (Trusted Platform Module), cryptographic co‑processors to protect firmware integrity and data privacy.