Channel estimation circuits and methods for estimating communication channels

A telecommunication device comprising:

• a channel estimation circuit, comprising
• processing circuitry configured
• to generate a channel autocorrelation matrix and at least one channel cross-correlation vector based on predetermined statistical information on a communication channel,
• to generate a subspace mapping for a subspace transformation based on the channel autocorrelation matrix,
• to generate a subspace transformed channel autocorrelation matrix, at least one subspace transformed channel cross-correlation vector, and a plurality of subspace transformed pilot symbols, by applying the subspace mapping to the channel autocorrelation matrix, the channel cross-correlation vector, and to a plurality of received pilot symbols,
• to generate a plurality of subspace channel estimation filter coefficients based on the subspace transformed channel autocorrelation matrix and the subspace transformed channel cross-correlation vector, and
• to generate an estimate of at least one channel coefficient of the communication channel based on the subspace transformed pilot symbols and the subspace channel estimation filter coefficients;
• and
• a receiver configured to receive signals from the communication channel and to provide the received signals to the channel estimation circuit.

A method for estimating a communication channel, comprising:

• receiving a plurality of pilot symbols from the communication channel;
• generating a channel autocorrelation matrix and at least one channel cross-correlation vector associated with the communication channel and based on predetermined statistical information on the communication channel;
• generating a subspace mapping for a subspace transformation based on the channel autocorrelation matrix;
• applying the subspace mapping to the channel autocorrelation matrix, the channel cross-correlation vector, and to the plurality of pilot symbols to generate a subspace transformed channel autocorrelation matrix, at least one subspace transformed channel cross-correlation vector, and a plurality of subspace transformed pilot symbols;
• generating a plurality of subspace channel estimation filter coefficients based on the subspace transformed channel autocorrelation matrix and the subspace transformed channel cross-correlation vector;
• and
• estimating at least one channel coefficient of the communication channel based on the subspace transformed pilot symbols and the subspace channel estimation filter coefficients.