The QPSK modulation scheme employs the alphabets {+/1 +/-j}. The simulation model generates the QPSK alphabets, passes through additive white gaussian noise and decodes the received symbol.

The symbol error rate computed through simulations is closely matching the theoretical symbol error rate. For description of the theoretical derivation of QPSK (4-QAM) symbol error rate, kindly refer the post:
http://www.dsplog.com/2007/11/06/symbol-error-rate-for-4-qam/