Spectrally efficient communication for wireless sensor networks using a cooperative MIMO technique

The multi-input multi-output (MIMO) communication framework is adopted for wireless sensor networks by having multiple sensors equipped with single-element antennas cooperate in transmission. A power method-based iterative algorithm is developed that computes the optimal transmit and receive eigen-filters distributively among the sensors while transferring most of the computational burden to the central collector node. Since the proposed algorithm implicitly exploits the channel state information (CSI) both at the receiver and the transmitter, it is expected that the resulting spectral efficiency is higher than what can be achieved by receive CSI-only space-time coding. This intuition is confirmed by employing a variable-rate adaptive modulation scheme for the eigen-transmission and comparing its spectral efficiency with that of orthogonal space time block codes (OSTBCs) at specific target bit error rates. The performance is also evaluated using realistic channel estimation as well as the least mean square (LMS) and recursive least square (RLS) algorithms for iterative eigencoding.