Sparse channel estimation for MIMO OFDM systems

A method for sparse channel estimation in Multiple Input Multiple Output Orthogonal Frequency Division Multiplexing (MIMO OFDM) systems with a plurality of subchannels having same sparsity structure, said method comprising steps of:

• initializing a plurality of residual vectors modeling a plurality of subcarriers;
• initializing observation generating matrices modeling the channel;
• sending at least one pilot signal for each of a plurality of subcarriers;
• converting the pilot signals to tap positions in the residual vectors;
• updating the plurality of residual vectors by removing all of the residual vectors exceeding a correlation threshold;
• updating the observation generating matrices in accordance with the residual vector having the optimal tap position;
• calculating, using a computer, weighted residuals based on the updated residual vectors;
• and repeating the steps of sending, converting, detecting, updating the residual vectors, updating the observation generating matrices, and calculating the weighted residuals, until a stopping condition is met, wherein the updated observation generating matrices estimate the sparse channel.

A method for sparse channel estimation in Multiple Input Multiple Output Orthogonal Frequency Division Multiplexing (MIMO OFDM) systems with a plurality of subchannels having same sparsity structure, said method comprising steps of:

• initializing a plurality of residual vectors modeling a plurality of subcarriers;
• initializing an observation generating matrix modeling the channel;
• sending at least one pilot signal for each of a plurality of subcarriers;
• converting the pilot signals to tap positions in the residual vectors;
• detecting an optimal tap position in one of the residual vectors;
• updating the plurality of residual vectors by removing the one residual vector having the optimal tap position;
• updating the observation generating matrix in accordance with the residual vector having the optimal tap position;
• calculating, using a computer, weighted residuals based on the updated residual vectors;
• and repeating the steps of sending, converting, detecting, updating the residual vectors, and calculating the weighted residuals, until a stopping condition is met, wherein the updated observation generation matrix estimates the sparse channel.