2.1 The Central Role of the Processor

The processor is the heart of an embedded system. It is the fundamental unit responsible for taking inputs, processing data according to a set of instructions, and producing outputs. For any embedded systems designer, a thorough understanding of processor fundamentals is non-negotiable, as it dictates the system’s capabilities, performance, and overall efficiency.

A processor contains two essential units that work in tandem:

• Program Flow Control Unit (CU): This unit, often simply called the Control Unit, manages the flow of the program. It includes a fetch unit that is responsible for retrieving instructions from memory.
• Execution Unit (EU): This unit carries out the instructions fetched by the CU. Its circuits implement data transfer operations, convert data from one form to another, and execute program control tasks such as jumps and interrupts. The EU contains the Arithmetic and Logical Unit (ALU), which performs all mathematical and logical calculations.

The processor operates on a fundamental “fetch and execute” cycle, continuously retrieving the next instruction from memory via the CU and executing it with the EU.

2.2 A Taxonomy of Processors

The term “processor” encompasses a wide range of devices, each tailored for different applications. They can be broadly classified as follows:

• General Purpose Processor (GPP)
  • Microprocessor
  • Microcontroller
  • Embedded Processor
  • Digital Signal Processor
  • Media Processor
• Application Specific System Processor (ASSP)
• Application Specific Instruction Processors (ASIPs)
• GPP or ASIP cores on an Application Specific Integrated Circuit (ASIC) or a Very Large Scale Integration (VLSI) circuit.

For the purpose of introductory embedded systems, the most important distinction is between the Microprocessor (a single VLSI chip containing a CPU) and the Microcontroller (a single-chip VLSI unit with a CPU, memory, and I/O integrated).

2.3 Comparative Analysis: Microprocessor vs. Microcontroller

While related, microprocessors and microcontrollers are designed for fundamentally different purposes. The choice between them is a critical design decision.

2.4 System Architecture: Memory Organization

The way a system organizes its memory for storing program instructions and data defines its fundamental architecture. There are two primary models.

Von Neumann Architecture

Proposed by the computer scientist John von Neumann, this architecture is defined by its use of a single data path or bus for both instructions and data. Because they share a common bus, the CPU can only perform one operation at a time: it either fetches an instruction from memory or performs a read/write operation on data. It cannot do both simultaneously. This design supports simpler hardware and allows for the use of a single, sequential memory space.

Harvard Architecture

The Harvard architecture features separate storage and signal buses for instructions and data. This physical separation is its key advantage, as it allows the CPU to access both instructions and data simultaneously. In this model, data storage is entirely contained within the CPU, and there is generally no way to access the instruction storage area as if it were data.

Architectural Comparison

(A note from the professor: The speed comparison in the table above is reproduced faithfully from the source material, but it is a common point of confusion and likely an error in the source. In practice, the Harvard architecture’s parallel buses for instructions and data allow for simultaneous access, which generally leads to higher performance by eliminating the “Von Neumann bottleneck” where data and instruction fetches compete for the same bus.)

2.5 Instruction Set Architecture: CISC vs. RISC

The final architectural consideration is the design philosophy behind the processor’s instruction set—the set of commands it can execute.

• Complex Instruction Set Computer (CISC): This architecture can address a large and complex number of instructions. The emphasis is on having powerful instructions that can accomplish multi-step operations in a single command.
• Reduced Instruction Set Computer (RISC): Developed in the early 1980s, the RISC philosophy proposed that computers should use fewer, simpler instructions that could be executed much faster within the CPU, without needing frequent memory access.

Architectural Comparison

Having reviewed the general theories of processors and architectures, we will now shift our focus to a specific, historically significant microcontroller that serves as an excellent platform for learning: the 8051.