ABSTRACT
Where SRP(·) is the operation of total power of zero pilots and y0 is the received time-domain training signal. The choosing of Q (or ε∆ = 1 / Q) is trade-off between accuracy in CFO estimate and computational complexity. Fig. 4 shows the mean square error (MSE) for different values of Q. Obviously, the estimation error decreases as the value of Q increases. A large Q can achieve a better performance, but it induces high computational complexity.
