The 1-bit compressed sensing framework enables the recovery of a sparse
vector x from the sign information of each entry of its linear transformation.
Discarding the amplitude information can significantly reduce the amount of
data, which is highly beneficial in practical applications.