Training-based channel estimation for multiple-antennas

A method comprising:

• developing n symbols streams from an information bit stream, where n is an integer greater than one;
• mapping symbols of each of said n symbol streams into a standard signal constellation to create n mapped streams;
• applying respective ones of the n mapped streams to a different one of n antennas at a first time period;
• generating a sequence of symbols;
• creating n training sequences from said sequence of symbols;
• mapping said n training sequences onto a standard constellation to create n mapped training sequences, where said n training sequences have a common length Nt, that is related to transmitted symbols memory, L-1, of a communication medium between said antennas and a receiver, Nt being greater than or equal to 2L-1 and
• applying respective ones of the n mapped training sequences to a different one of said n antennas during a second time period.

A space-time diversity transmitter comprising:

• n transmitting antennas, where n is an integer greater than one;
• a first encoder responsive to an applied information bit stream, said first encoder developing n symbol streams;
• a constellation mapper responsive to said n symbol streams, to map symbols of each of said n symbol streams into a standard signal constellation to create n mapped streams, said constellation mapper configured to apply respective ones of the n mapped streams to said n antennas in blocks of Nt mapped symbols that are synchronized in time to each other;
• a training generator for generating either n sequences of signals or a sequence of signals that contains n subsequences of signal;
• a second encoder for creating n training symbol sequences from said signals created by said training generator, each containing N symbols;
• and
• said constellation mapper is configured to map said n training symbol sequences onto a standard constellation to develop n mapped training sequences, and to apply respective ones of the n mapped training sequences to said n antennas at times when said constellation mapper is not applying said n mapped streams to said n antennas;
• where said n training symbol sequences have an impulse-like autocorrelation function and zero cross correlation, said training generator creates a sequence s, and said second encoder creates a first training sequence that is equal to -s concatenated with s, and a second training sequence that is equal to s concatenated with s.