The documentation of gr_single_pole_avg_filter_ff is not coherent to its code.

The equation presented in the doc does not reflect the current code:

y[n] - (1-alpha) y[n-1] = alpha x[n]

output = d_alpha * input - d_one_minus_alpha * d_prev_input;

And in gr_single_pole_rec.h:

computes y(i) = alpha * x(i) - (1-alpha) * y(i-1)

output = d_alpha * input + d_one_minus_alpha * d_prev_output;