In article <41EBCC47.7050801@sbcglobal.net>,
Alfred A. Aburto Jr. <aburto@sbcglobal.net> wrote:

I suspected this because the subbands are apparently divided into 
approximate 10KHz bands which implies that the subband sample frequency 
  should be about 20KHz (2*9765.62 = 19531.24 Hz).

The sample rate is correct.  The signal data is supplied as complex
numbers so both positive and negative frequency components (upper and
lower sidebands) can be computed.

As the samples are only two bits overall, they effectively have no amplitude
information but simply encode nominal phases of 45, 135, 225 and 315 degrees.