Receive Beamforming Design Based on a Multiple-State Interference Model

Abstract

Abstract-In this paper, we propose a multiple-state interference model, which takes into account the interference variability and uncertainty created by scheduling and other fast resource allocation adaptivity. This situation is modeled by a set of states, each described by a spatial covariance matrix and a probability. In order to illustrate the usefulness of such a model, we study two receive beamforming design problems, one maximizing the average data rate and one maximizing the worst-case data rate. We compare the resulting performance with what can be obtained when the state information is not available. We show theoretically and numerically that exploring the multiple-state interference structure can improve the receive beamforming design efficiency, especially for cell-edge users in a interference-limited system.

Optimal Resource Allocation in Coordinated Multi-Cell Systems

“…It is frequently assumed in multi-cell scenarios (but not necessary) that each power constraint only affects the signals from one of the base stations; for example, per-transmitter power constraints is represented by having L = K t and the constraint affecting BS l is 9 It is implicitly assumed that n k is an ergodic process, which is not necessarily satisfied if unknown communication systems with fast adaptive resource allocation strategies are creating the interference; a further discussion is available in [302].…”

“…An important difference is that the actual transmitted signals are D k s k (and not s k ), thus each weighting matrix Q lk should satisfy the additional condition that Q lk − D H k Q lk D k is diagonal for all l,k (e.g., being zero). This technical assumption makes sure that power cannot be allocated to unallowed subspaces for the purpose of reducing the (measured) power in the subspaces used for transmission -which is only possible when Q lk is nondiagonal.It is frequently assumed in multi-cell scenarios (but not necessary) that each power constraint only affects the signals from one of the base stations; for example, per-transmitter power constraints is represented by having L = K t and the constraint affecting BS l is 9 It is implicitly assumed that n k is an ergodic process, which is not necessarily satisfied if unknown communication systems with fast adaptive resource allocation strategies are creating the interference; a further discussion is available in [302]. …”

Robust receive beamforming with interference and channel uncertainty

“…Apart from the afore mentioned single-state interference model where the interference is modeled as colored noise, in [12], we proposed a multiple-state interference model, which can better reflect the bursty nature of the interference generated from modern fast resource allocation schemes. Here the interference is generated from one of the K states, and each state is characterized by a covariance matrix R ik associated with a probability of occurrence p k , k = 1, 2, ..., K, K being the number of states in the model.…”

“…Finally, we illustrate the performance on different interference models: traditional single-state interference model and the multiplestate interference model propose in [12].…”

“…Furthermore, we extend our design to a multiple-state interference model [12] and show the performance gains obtained by exploiting the interference structure. …”