An optimal precoder design is conceived for the decentralized estimation of an unknown spatially as well as temporally correlated parameter vector in a multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) based wireless sensor network (WSN). Furthermore, exploiting the temporal correlation present in the parameter vector, a rate-distortion theory based framework is developed for the optimal quantization of the sensor observations so that the resultant distortion is minimized for a given bit-budget. Subsequently, optimal precoders are also developed that minimize the sum-MSE (SMSE) for the scenario of transmitting quantized observations. In order to reduce the computational complexity of the decentralized framework, distributed precoder design algorithms are also developed which design precoders using the consensus based alternating direction method of multipliers (ADMM), wherein each SN determines its precoders without any central coordination by the fusion center. Finally, new robust MIMO precoder designs are proposed for practical scenarios operating in the face of channel state information (CSI) uncertainty. Our simulation results demonstrate the improved performance of the proposed schemes and corroborate our analytical formulations.

In this paper, we consider interference channel model in which transmissions from multiple users are partially correlated. This correlation arises in wireless sensor network (WSN) scenarios and temporally correlated models. Considering this model, two minimum mean squared error (MSE) based precoding methods are derived. With these formulations, an iterative convergent procedure is formulated similar to a typical interference alignment (IA) algorithm. Simulations show that the second method provides the best sum rates for different correlation values.

This paper presents two analog precoder feedback (PFB) schemes namely the simple PFB (SPFB) scheme and the reduced PFB (RPFB) scheme for interference channels. In the first scheme, each destination transmits full precoder, while in the second one, the size of the precoder to be fed back is reduced by removing the redundant rows of the precoder matrix. This reduced feedback is particularly useful in the systems, where the number of transmit antennas are more than the number of receive antennas. These schemes have been analyzed for the cases where time slots in the feedback duration are independent of the number of users and where the time slots scale linearly with the number of users. For both these schemes, the precoder reconstruction methods are investigated using the chordal distance. For interference alignment, the rate loss upper bound analysis shows that the rate loss increases with signal-to-noise-ratio (SNR) at low SNR, while it remains constant in high SNR range. Simulation results verify that the linear sum rate scaling is preserved at high SNRs. The full precoder feedback dominates, when the source cooperation is assumed, while RPFB yields improvements over SPFB scheme, when no source cooperation is assumed and$Kd$time slots are utilized in either orthogonal or independent transmission case.

Interference Alignment is a promising technique for achieving higher rates by aligning interference at the receiver. To design such a system, the global channel state information at the transmitter (CSIT) as well as at the receiver is necessary. But in practice, it is hard to obtain this information, therefore, limited feedback is used to provide CSIT or the precoder design information to the transmitter. Conventionally, precoders are quantized at receiver by finding its best match in the codebook using chordal distance and its index is fedback to the transmitter. In this paper, instead of minimizing chordal distance, we propose algorithms with objectives that are derived from subspace alignment method, SINR maximization, or minimization of leakage interference power to measure the “goodness” of quantized vector. These algorithms achieve higher rates for small size codebooks. The rate loss has been analyzed for precoder quantization. We also find less computational intensive solution to find the desired vectors in the codebook. The simulation results show that for small codebooks, significant sumrate gains can be achieved for (2 × 2,1)3for 2–6 bits of feedback per user, compared to quantization based on chordal distance, while for large codebooks, the chordal distance based quantization performs better.