how does USB2.0 protocol work (good read)

from someone good at EE on the internet:

There are four wires (ignoring USB3.x for the moment). Two for power (+5V and GND), and two for signalling (D+ and D-).

The key thing to note about the signal wires is their name, note the + and -. These typically indicate in electronics that something is differential. What this means is that the 1's and 0's are indicated by the polarity of the voltage between each cable. This is as opposed to single ended where the 1's and 0's are carried as a voltage relative to GND.

What do I mean by polarity? Well imagine the D+ cable is driven to +3.3V + 3.3 V , and the D- is driven to 0V 0 V . The difference between the two is 𝑉𝐷+−𝑉𝐷−=3.3−0=3.3V V D + − V D − = 3.3 − 0 = 3.3 V . Now if instead the D+ cable was driven to 0V 0 V and the D- driven to +3.3V + 3.3 V , the difference becomes 𝑉𝐷+−𝑉𝐷−=0−3.3=−3.3V V D + − V D − = 0 − 3.3 = − 3.3 V . Notice the minus sign, indicating the opposite polarity.

For this to work then, the two data cables must be the complement of each other (when one is high, the other is low) to transfer data, thus must operate at the same frequency. You may think why bother, just use one cable. The thing is the world is a rather noisy place, a single ended (common mode) wire is very prone to noise which at high speeds (even at low speeds in harsh environments) which can corrupt the data (make a 1 a 0). In differential signalling, both cables are exposed to the same noise, so it cancels out