Let us consider any video signal. Bit rate of a video signal is the no of bits per frame per second. Also, data rate is the no of bits per second. Baud rate is the number of signal changes per second.

But I really confused between these 3 terms. They look similar to each other. So can anyone tell me whether these terms are same or different with the help any example?


Bit rate and data rate are somewhat ambiguous, and their exact definitions vary from one field or application to another.

Bit rate is often used to measure the information rate, since information is measured in bits. Sometimes it is used to measure the number of actual 1s and 0s transmitted per second. When the information is compressed (source-coded), the two definitions are equal.

Data rate can be used to mean the same as bit rate. Sometimes it is used to measure the rate at which data, as opposed to overhead, is transmitted. For instance, an audio compact disc system has a data rate of $2\cdot44100\cdot16=1411200$ bits (zeros and ones) per second, but the actual transmission rate is approximately double that, once you include framing, metadata, and FEC.

The baud rate (or symbol rate, or pulse rate) is the number of actual pulses (symbols) transmitted per second. If a pulse "carries" several bits, then the baud rate can be significantly smaller than the bit rate. This is important because the bandwidth required to transmit the information is determined by the baud rate.

So, consider an encoded video stream with 1000 bits per frame, at 24 frames per second, with 10% overhead per frame, and 4 bits per symbol:

  • From an information-theory perspective, the bit rate is $24\cdot 1000=24000$ (information) bits/sec.
  • You can also say, from a more practical perspective, that the bit rate required is $1.1\cdot 24 \cdot 1000 = 26400$ (ones and zeros) bits per second, but you could still define the data rate as $24000$ bits per second, since the remaining 2400 bits are overhead.
  • The pulse rate on the transmission link will be $26400/4=6,600$ pulses per second.
  • $\begingroup$ Is there a term inbetween information rate and gross bitrate that describes the rate of the bitstream that includes error coding (but doesn't include other layer 1 overheads i.e. guard carriers and cyclic prefices)? $\endgroup$ – Lewis Kelsey Oct 13 '20 at 3:07
  • $\begingroup$ I'm going to go with coded data bit rate, which seems to be being denoted as Rc (and Ec/N0 for its SNR, whereas Eb/N0 is with respect to information rate), although Ec/N0 means something else in CDMA (per chip) $\endgroup$ – Lewis Kelsey Oct 13 '20 at 6:26
  • 1
    $\begingroup$ @LewisKelsey What you describe is fine, as long as you define the terms precisely. I've also seen "bit rate" to describe the actual channel rate, and "information rate" to describe actual user data rate. $\endgroup$ – MBaz Oct 13 '20 at 12:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.