Cooperative Algorithms for MIMO Interference Channels

TLDR

Interference alignment exploits all degrees of freedom in multi‑user interference channels, yet prior work ignores interference from non‑participating nodes. This study introduces three IA generalizations that incorporate colored noise to model uncoordinated interference. The authors present a minimum interference‑plus‑noise leakage algorithm equivalent to subspace methods under white noise, a joint MMSE design optimizing transmit precoders and receive filters, and a convergent maximum SINR algorithm, and evaluate their sum throughput in networks with uncoordinated cochannel interferers. The INL algorithm yields orthonormal precoders suitable for limited‑feedback systems, the maximum SINR algorithm converges, and simulations show that accounting for cochannel interference improves throughput except when a single receiver experiences very high interference.

Abstract

Interference alignment (IA) is a transmission technique for exploiting all available degrees of freedom in the frequency- or time-selective interference channel with an arbitrary number of users. Most prior work on IA, however, neglects interference from other nodes in the network that are not participating in the alignment operation. This paper proposes three generalizations of IA for the multiple-antenna interference channel with multiple users that account for colored noise, which models uncoordinated interference. First, a minimum interference-plus-noise leakage (INL) algorithm is presented and shown to be equivalent to previous subspace methods when noise is spatially white or negligible. This algorithm results in orthonormal precoders that are desirable for practical implementation with limited feedback. A joint minimum mean square error design that jointly optimizes the transmit precoders and receive spatial filters is then proposed, whereas previous designs neglect the receive spatial filter. Finally, a maximum signal-to-interference-plus-noise ratio (SINR) algorithm is developed and proven to converge, unlike previous maximum SINR algorithms. The sum throughput of these algorithms is simulated in the context of a network with uncoordinated cochannel interferers that are not participating in the alignment protocol. It is found that a network with cochannel interference can benefit from employing precoders that are designed to consider that interference, but in extreme cases, such as when only one receiver has a large amount of interference, ignoring that the cochannel interference is advantageous.