UART Protocol

Master the UART (Universal Asynchronous Receiver/Transmitter) protocol and its VHDL implementation

What is UART?

UART (Universal Asynchronous Receiver/Transmitter) is an asynchronous serial protocol: unlike SPI and I2C, there is no shared clock between transmitter and receiver. Synchronization is achieved by a predefined communication rate called the baud rate (bits per second).

Key characteristics:

• Asynchronous: no shared clock
• Full-duplex: TX and RX are independent lines
• Point-to-point: 2 devices only (no multi-slave bus)
• Baud rate: both sides must be configured to the same rate

UART Signals

At rest, the TX line is high (logic '1', called "mark").

UART Frame

A standard UART frame (8N1 — 8 data bits, no parity, 1 stop bit):

1. Start bit: low level for 1 bit period — signals the beginning of the frame
2. Data bits: 5 to 9 bits, LSB first (standard convention)
3. Parity bit (optional): even, odd, or none
4. Stop bit(s): high level for 1 or 2 bit periods — end of frame

Baud Rate and Clock Divider

The baud rate determines bit duration: T_bit = 1 / baud_rate

To generate the baud rate from a system clock, calculate a divider:

DIVIDER = SYS_CLK / BAUD_RATE
Example: 100 MHz / 115200 ≈ 868

VHDL Implementation

The UART transmitter uses a state machine with 4 phases: IDLE, START, DATA and STOP. A clock divider (CLK_SYS / BAUD_RATE) generates the timing for each bit. Data is sent LSB first, following the standard.

UART exercises will be available soon on the platform to practice implementing a complete UART transmitter and receiver.

UART Receiver — 16x Oversampling

The receiver must synchronize on the Start bit, then sample each data bit at the center of its period to maximize noise immunity.

The classic technique is 16x oversampling: the frequency divider generates 16 ticks per bit period. The receiver:

1. Waits for the falling edge of the Start bit
2. Waits 8 ticks (half period) to position at the center of the Start bit
3. Verifies the line is still low (validates the Start bit)
4. Every 16 ticks, samples a data bit

constant c_DIV16 : positive := g_CLK_HZ / (g_BAUD * 16);

Key Points

• Same baud rate on both sides: a difference of more than 3-5% causes decoding errors
• Logic inversion: RS-232 UART uses ±12 V levels. A MAX232 or adapter is required for 3.3 V or 5 V interfaces
• No error detection: except with the optional parity bit — higher-level protocols (e.g., Modbus) add checksums
• Idle level: the TX line must be high at rest. A disconnected TX will be seen as a continuous data stream

Advantages and Disadvantages