The sum-rate capacity and partial capacity boundary of the parallel Gaussian Z interference channels are shown to be achieved by decoding part of the interference and treating the remaining interference as noise if a certain condition is satisfied.

The optimal signaling for K-user multiple-antenna Gaussian interference channels with single-user detection has been studied. It is shown that to achieve the boundary of the rate region, the rank of the optimal input covariance matrix is no greater than the rank of the direct link channel matrix as long as this transmitter has perfect channel state information of its direct link. Based on this, interference mitigation and signal enhancement are derived and shown to be necessary conditions for the optimal signaling. Different power constraints and channel availability scenarios are also considered.

For a transceiver pair in a multiuser Gaussian interference channel with multiple antennas, if the number of transmit antennas is greater than the number of receive antennas, then in an equivalent virtual channel this transceiver pair has redundant transmit antennas. These redundant antennas do not contribute to the transmission to the intended receiver, however, may generate interference to other receivers. The optimal way to use these redundant antennas to mitigate interference and enhance the useful signal is studied, under the assumption of perfect channel state information, channel distribution information, and imperfect channel state information at the transmitter side, respectively.

The capacity of a network in which a multiple access channel generates interference to a single-user channel is studied. An achievable rate region based on superposition coding and joint decoding is established for the discrete case. If the interference is very strong, the capacity region is obtained for both the discrete memoryless channel and the Gaussian channel. For the strong interference case, the capacity region is established for the discrete memoryless channel; for the Gaussian case, a line segment on the boundary of the capacity region is attained. Moreover, the capacity region for the Gaussian channel is identified for the case in which one interference link is strong, and the other is very strong. For a subclass of Gaussian channels with mixed interference, a boundary point of the capacity region is determined. Finally, for the Gaussian channel with weak interference, sum capacities are obtained under various channel coefficient and power constraint conditions.

New sufficient conditions for a vector Gaussian interference channel to achieve the sum-rate capacity by treating interference as noise are derived, which generalize and extend the existing results. More concise conditions for multiple-input single-output, and single-input multiple-output scenarios are obtained.

There exist vector Gaussian interference channels that may not satisfy the strong interference condition but for which the capacity can still be achieved by jointly decoding the signal and the interference. This kind of interference is called generally strong interference. Sufficient conditions for a vector Gaussian interference channel to have generally strong interference are derived. The sum-rate capacity and the boundary points of the capacity region are also determined. Furthermore, new sufficient conditions for a vector Gaussian interference channel to achieve the sum-rate capacity by treating interference as noise are derived, which generalize all the existing results. Some boundary points on the capacity region of a vector Z interference channel under noisy interference are also determined to be achieved by treating interference as noise.

An interference channel is said to have strong interference if a certain pair of mutual information inequalities are satisfied for all input distributions. These inequalities assure that the capacity of the interference channel with strong interference is achieved by jointly decoding the signal and the interference. This definition of strong interference applies to discrete memoryless, scalar and vector Gaussian interference channels. However, there exist vector Gaussian interference channels that may not satisfy the strong interference condition but for which the capacity can still be achieved by jointly decoding the signal and the interference. This kind of interference is called generally strong interference. Sufficient conditions for a vector Gaussian interference channel to have generally strong interference are derived. The sum-rate capacity and the boundary points of the capacity region are also determined.

This paper considers an increase in the achievable rate of beamforming systems in two-way communication, resulting from bi-directional use of the spatial resources. The concept of bi-directional use of the spatial resources is first introduced. Then, a condition is derived under which bi-directional beamforming (BB) has larger achievable rate than that of conventional beamforming operating via frequency division duplexing (UB-FDD). In addition, the variation in the achievable rate of bi-directional beamforming is investigated when echo interference is pre-cancelled in the air. Simulation results show that the achievable rate of BB is always larger when the echo is ideally eliminated, and is larger over a certain level of SNR and below a certain level of fading correlation than that of UB-FDD.

The capacity of a network in which a multiple access channel (MAC) generates interference to a single-user channel is studied. An achievable rate region based on superposition coding and joint decoding is established for the discrete case. If the interference is strong, the capacity region is obtained for both the discrete memoryless channel and Gaussian channel. A boundary point of the capacity region is determined for a subclass of Gaussian channels with mixed interference.

The sum-capacity for specific sub-classes of ergodic fading Gaussian two-user interference channels (IFCs) is developed under the assumption of perfect channel state information at all transmitters and receivers. For the sub-classes of uniformly strong (every fading state is strong) and ergodic very strong two-sided IFCs (a mix of strong and weak fading states satisfying specific fading averaged conditions) the optimality of completely decoding the interference, i.e., converting the IFC to a compound multiple access channel (C-MAC), is proved. It is also shown that this capacity-achieving scheme requires encoding and decoding jointly across all fading states. As an achievable scheme and also as a topic of independent interest, the capacity region and the corresponding optimal power policies for an ergodic fading C-MAC are developed. For the sub-class of uniformly weak IFCs (every fading state is weak), genie-aided outer bounds are developed. The bounds are shown to be achieved by treating interference as noise and by separable coding for one-sided fading IFCs. Finally, for the sub-class of one-sided hybrid IFCs (a mix of weak and strong states that do not satisfy ergodic very strong conditions), an achievable scheme involving rate splitting and joint coding across all fading states is developed and is shown to perform at least as well as a separable coding scheme.

The sum-rate capacity of the parallel Gaussian interference channel is shown to be achieved by independent transmission across subchannels and treating interference as noise if the channel coefficients and power constraints satisfy a certain condition. The condition requires the interference to be weak, a situation commonly encountered, e.g., in digital subscriber line transmission. The optimal power allocation is characterized by using the concavity of the sum-rate capacity as a function of the power constraints.

For a multiuser interference channel with multiantenna transmitters and single-antenna receivers, by restricting each transmitter to a Gaussian input and each receiver to a single-user detector, computing the largest achievable rate region amounts to solving a family of nonconvex optimization problems. Recognizing the intrinsic connection between the signal power at the intended receiver and the interference power at the unintended receiver, the original family of nonconvex optimization problems is converted into a new family of convex optimization problems. It is shown that, for such interference channels with each receiver implementing single-user detection, transmitter beamforming can achieve all boundary points of the achievable rate region.

The capacity of a network in which a degraded broadcast channel generates interference to a single-user channel is studied. An achievable rate region based on superposition coding and joint decoding is derived for the discrete case. Outer bounds on the capacity region are obtained for different degradedness conditions. For the Gaussian case, the capacity regions are obtained for very strong and slightly strong interference cases. When the interference is moderate or weak, some boundary points of the capacity regions are determined. For all the obtained capacity results, it is shown that to achieve the capacity, the interference should be fully decoded when it is strong, partially decoded when it is moderate, and treated as noise when it is weak.

In multiple antenna systems, spatial fading correlation causes a significant degradation in performance. In this work, a method for reducing the sensitivity to spatial fading correlation through the bi-directional use of spatial resources is investigated. Firstly, the concept of bi-directional use of spatial resources is introduced and the fading characteristics of the proposed bi-directional multiple-input multiple-output (B-MIMO) system are investigated. The achievable rates of the B-MIMO system are then investigated and compared with those of conventional MIMO systems for beamforming and full rank transmission. Based on those results, conditions needed for the B-MIMO system to demonstrate a larger average achievable rate are evaluated. B-MIMO systems are found to efficiently reduce the sensitivity to spatial fading correlation, while supporting comparable achievable rates in spatially white environments.

The capacity of a network in which a broadcast channel generates interference to a single-user channel is studied. The capacity region is achieved by fully decoding the interference when it is very strong or slightly strong. When the interference is moderate or weak, the weighted sum-rate capacities are achieved by partially decoding the interference or by treating the interference as noise, respectively.

The capacity regions of vector, or multiple-input multiple-output, Gaussian interference channels are established for very strong interference and aligned strong interference. Furthermore, the sum-rate capacities are established for Z interference, noisy interference, and mixed (aligned weak/intermediate and aligned strong) interference. These results generalize known results for scalar Gaussian interference channels.

For a multi-user interference channel with multi-antenna transmitters and single-antenna receivers, by restricting each receiver to a single-user detector, computing the largest achievable rate region amounts to solving a family of nonconvex optimization problems. Recognizing the intrinsic connection between the signal power at the intended receiver and the interference power at the unintended receiver, the original family of non-convex optimization problems is converted into a new family of convex optimization problems. It is shown that, for such interference channels with each receiver implementing single-user detection, transmitter beamforming can achieve all boundary points of the achievable rate region.

The sum-rate capacity of the parallel Gaussian interference channel is studied. Sufficient conditions are derived in terms of channel coefficients and power constraints such that the sum-rate capacity can be achieved by: 1) independent transmission across sub-channels, and 2) treating interference as noise in each sub-channel. The optimal power allocation is characterized for such parallel channels.

A new outer bound on the capacity region of Gaussian interference channels is developed. The bound combines and improves existing genie-aided methods and is shown to give the sum-rate capacity for noisy interference as defined in this paper. Specifically, it is shown that if the channel crosstalk coefficient magnitudes lie below thresholds defined by the power constraints then single-user detection at each receiver is sum-rate optimal, i.e., treating the interference as noise incurs no loss in performance. This is the first capacity result for the Gaussian interference channel with weak to moderate interference. Furthermore, for certain mixed (weak and strong) interference scenarios, the new outer bounds give a corner point of the capacity region.

In this paper, we study the optimal transceiver structure for achieving sum-rate capacity of a parallel Gaussian symmetric interference channel. Specifically, we derive a set of channel and power constraint conditions such that the total sum-rate capacity can be achieved by treating interference as noise in each sub-channel. The property of the optimal power allocation is also characterized for such parallel channels.

An outer bound on the capacity region of the two-user Gaussian interference channel is derived. The bound shows that for low power and small crosstalk coefficients the sum-rate capacity is achieved by treating interference as noise. The results are generalized to multiuser channels.

We study in this paper the sum capacity achievability of orthogonal transmissions in vector Gaussian multiple access channels (MAC). Specifically, we derive sufficient and necessary conditions, in terms of channel matrices and transmitter power constraints, for orthogonal transmissions to achieve the sum capacity of a vector Gaussian MAC. The obtained conditions provide a unified framework that helps explain many intuitive and known results as well as explore cases that have not been addressed. In the cases when these conditions are violated, our results enable us to quantify the suboptimality of orthogonal transmission when the sum capacity can only be achieved by overlay transmission.

We study in this paper the sum capacity achievability of orthogonal transmissions in vector Gaussian multiple access channels (MAC). Specifically, we derive the sufficient and necessary conditions, in terms of channel matrices and transmitter power constraints, for orthogonal transmission to achieve the sum capacity of a vector Gaussian MAC. The obtained conditions provide a unified framework in explaining many of the results that are intuitively true. They also enable us to explore cases that have not been addressed to determine the (sub)optimality of orthogonal transmissions compared with the overlay transmission.

A modified FDM/TDM method was first used by Sato to study the achievable rate region of degraded Gaussian interference channel. Recently Sason used this method to get an achievable rate region of Gaussian interference channel. In this paper, we generalize this method in terms of the transmission mode and frequency band allocation, and obtain an achievable rate region for Gaussian interference channel, which improves the existing computable rate regions for both the weak and moderate interferences. We note that this modified FDM/TDM is different to time sharing and the difference depends on the activeness and concavity of the constraints. The achievable sum rate of symmetric Gaussian interference channel is derived for a subregion of the classical Han-Kobayashi region, and subsequently, for the proposed new achievable rate region.

An inner bound of capacity region for the Gaussian interference channel is derived using Sato's modified frequency division multiplexing idea (Sato, 1978) and a special case of Han and Kobayashi's rate region (denoted by G' in Han and Kobayashi (1981)). We show that the new inner bound includes G', Sason's rate region D (Sason, 2004), as well as the achievable region via TDM/FDM (Carleial, 1978), as its subsets. The advantage of this improved inner bound over G' arises due to its inherent ability to utilize the whole transmit power range on the real line without violating the power constraint. We also provide analysis to show that the new achievable region strictly extends G' if max(R1, R2)isinG'{R1+ kR2} is nonconcave for some k isin 0, +infin, where R1and R2are the rates of the two users.

We develop a procedure to compute a lower bound of the sum capacity for the vector Gaussian interference channel (IFC). The obtained lower bound is shown to be tight for the very strong interference case and the procedure can be used to construct an inner bound for the vector Gaussian IFC. Alternatively, orthogonal transmission via frequency division multiplexing is considered and we establish the concavity of the sum rate as a function of the bandwidth allocation factor for the vector channel case. Numerical examples indicate that the achievable sum rate via the superposition code compares favorably with orthogonal transmission. This improvement holds for all interference power levels, a sharp contrast to that of the scalar counterpart

In this correspondence, we study some information theoretical characteristics of vector Gaussian interference channels. Resorting to the superposition code technique, a lower bound of the sum capacity for the vector Gaussian interference channel is obtained. Alternatively, orthogonal transmission via frequency division multiplexing is considered and we establish the concavity of sum rate as the bandwidth allocation factor for the vector channel case. Numerical examples indicate that the achievable sum rate via the superposition code compares favorably with orthogonal transmission: the lower bound obtained via the superposition code dominates the best achievable sum rate through orthogonal transmission. This improvement holds for all interference power levels, a sharp contrast to that of the scalar counterpart