Quadrature imbalance estimation using unbiased training sequences

A system for communication, the system comprising:

• a receiver configured to receive a training signal, the training signal representing a plurality of predetermined reference values at a frequency +f or a frequency adjacent to the frequency +f and a plurality of corresponding predetermined mirror values at a frequency −f or a frequency adjacent to the frequency −f, wherein the sum of the products of each predetermined reference value and the corresponding predetermined mirror value is zero;
• and
• a processor configured to:
• generate a plurality of received values based on the received portion of the training signal representing the predetermined reference values at the frequency +f or the frequency adjacent to the frequency +f,
• generate a plurality of derotation values by multiplying each received value by the corresponding predetermined mirror value, and
• average the derotation values so as to obtain a quadrature imbalance estimate for the frequency −f or the frequency adjacent to the frequency −f.

A system for communication, the system comprising:

• means for receiving a training signal at a quadrature demodulation receiver, the training signal representing a plurality of predetermined reference values at a frequency +f or a frequency adjacent to the frequency +f and a plurality of corresponding predetermined mirror values at a frequency −f or a frequency adjacent to the frequency −f, wherein the sum of the products of each predetermined reference value and the corresponding predetermined mirror value is zero;
• means for generating a plurality of received values based on the received portion of the training signal representing the predetermined reference values at the frequency +f or the frequency adjacent to the frequency +f;
• means for generating a plurality of derotation values by multiplying each received value by the corresponding predetermined mirror value;
• and
• means for averaging the derotation values so as to obtain a quadrature imbalance estimate for the frequency −f or the frequency adjacent to the frequency −f.

A computer-readable non-transitory medium having instructions encoded thereon which, when executed by a computer, cause a system to perform a method of communication, the method comprising:

• receiving a training signal at a quadrature demodulation receiver, the training signal representing a plurality of predetermined reference values at a frequency +f or a frequency adjacent to the frequency +f and a plurality of corresponding predetermined mirror values at a frequency −f or a frequency adjacent to the frequency −f, wherein the sum of the products of each predetermined reference value and the corresponding predetermined mirror value is zero;
• generating a plurality of received values based on the received portion of the training signal representing the predetermined reference values at the frequency +f or the frequency adjacent to the frequency +f;
• generating a plurality of derotation values by multiplying each received value by the corresponding predetermined mirror value;
• and
• averaging the derotation values so as to obtain a quadrature imbalance estimate for the frequency −f or the frequency adjacent to the frequency −f.