Hi Everyone!

I have a question about Lab 2. When it asks for the symbol error, what does it mean? I think that for 16-QAPSK, each “symbol” represents 2 bits so the bit error may be different from the symbol error. But for QPSK, the symbol error would be equal to the bit error since every symbol is one bit.

But there is another way to think about it where we can consider the output of the r and s matched filters as one symbol. So if you send one bit and erroneously get a 0 on the r matched filter but get a correct value for the s matched filter, the symbol error would be 100% but the bit error would be 50%. Any thoughts?

Thanks! Hassan

Hi Hassan,

The definition of a symbol is what you can transmit through the channel at a given instance… In both cases in our lab our Baud rate is 1 symbol/second. I think you are confusing the physical make up of the signal (such as sine waves and matched filters) with the more abstract “Digital symbol”. In the case of QPSK you transmit two digital bits at any given moment, thus you are sending 2 bits per symbol. In the case of 16-APSK you are sending 4 bits per symbol (analogous to sending a single hexadecimal digit per second). Now in both cases the bit error rate may be the same, but in the case symbol error rate, you are expecting to have more symbol mismatches in the case where each symbol carries more bits.

Example:

Source data:

1 0 1 0 0 0 1 1 1 0 1 0 0 1 0 1

1 0 1 0 1 0 1 1 1 1 1 0 1 1 0 1

Where the error rate is 3/16 errors per bit

Now in the first case each symbol is two consecutive bits:

The error rate is 3/8 errors per symbol.

Now in our second case each symbol is four consecutive bits:

The error rate is 3/4 errors per symbol.

Hope it helps, and hope someone will correct me if I am wrong :)

Regards,

Ron