In broadcast channels, network utility maximization formulates a scheduling problem in order to maximize a sum of given utility functions. It is known that the network utility maximization is achieved if a gradient method is used. In this paper, however, we show that the network utility maximization is achieved only if a gradient method is used. That proves the equivalence between a problem formulated by the network utility maximization and a problem with a gradient method. The gradient method simplifies the object function of a scheduling problem by modifying utility functions to a gradient form, so that it makes easy to deal with the problem. We apply the gradient method to a problem with utility functions given by generalized proportional fairness. It is revealed that using the gradient method for the generalized proportional fairness is equivalent to applying a binomial approximation. Simulation results are presented with various scheduling parameters.

This letter investigates the effect of the transmit correlation on the sum-rate capacity of a two-user broadcast channel (BC) with the use of equal-power allocation in high signal-to-noise ratio (SNR) environments. It is shown using an upper-bound analysis that the sum-rate capacity of a correlated two-user BC at high SNR highly depends on the phase difference between the transmit correlation coefficients of two users as well as the magnitude of the transmit correlation coefficients, and it is maximized (or minimized) when the principal eigenvectors of the transmit correlation matrix of two users are orthogonal (or parallel) to each other. The analytic results are verified by computer simulation.

In this paper, we propose an adaptive beamforming scheme that generates the beam weight using dominant eigenvectors of the spatial covariance matrix. The number of eigenvectors used for the generation of beam weight is determined to maximize the signal-to-noise ratio (SNR) for given feedback constraints (e.g., the amount of feedback information and feedback delay). It is shown that the conventional limited feedback beamforming and eigen-beamforming are special cases of the proposed scheme. Simulation results show that the proposed beamforming scheme can effectively be applied to spatially correlated channels.

In this paper, we propose a new beamforming scheme that utilizes both long-term and instantaneous channel state information (CSI). First, the dimension of the channel is effectively reduced by considering dominant eigen-dimensions of the channel covariance matrix. Then, the beam weight is generated using the instantaneous CSI of the effectively dimension-reduced channel. The reduction of the channel dimension is determined to maximize the signal-to-noise ratio (SNR) in an average sense considering the amount of feedback signaling burden and spatial correlation of the channel. It is analytically shown and verified by computer simulation that the proposed scheme outperforms conventional beamforming schemes when the same amount of feedback signaling burden is utilized.

In this paper, we propose a new beamforming scheme that generates beams by exploiting the spatial correlation in multiuser environments. The proposed scheme generates random beams using eigenvectors corresponding to dominant eigenvalues of the channel covariance matrix, increasing the multiuser diversity gain. Moreover, it can obtain an additional SNR gain by transmitting the signal with a beam weight coherent to the effective channel corresponding to dominant eigenvectors of the channel covariance matrix of a selected user. Simulation results show that the proposed scheme is quite effective in correlated channel environments.

This paper considers a transmit beamforming scheme that works with reduced feedback information in OFDM based multiple-input single-output (MISO) wireless systems. The proposed scheme generates the beamforming weight using information on the previous beamforming weights and channel correlation, significantly reducing the amount of feedback signaling burden. The feedback signaling overhead can further be reduced with the use of clustering and interpolation techniques. Simulation results show that the proposed scheme outperforms conventional beamforming techniques, while using the same amount of feedback signaling overhead.

In this paper, we propose a new beamforming scheme that can work with partial channel information. The proposed beamforming scheme can simultaneously achieve multiuser diversity (MUD) gain and beamforming gain by utilizing the beam weight information of the selected user in addition to the signal-to-noise ratio (SNR) information of all users. To fully exploit spatial multiplexing gain, we further consider the use of multiple beams. Multiple beams are generated in a sequential manner to reduce the interference among users, while virtually increasing the desired signal power. Finally, we analyze the performance of the proposed scheme and verify it by computer simulation. Although the proposed scheme is somewhat affected by the user mobility, numerical results show that the proposed scheme is quite effective in low mobility environments.

This paper considers opportunistic packet scheduling with the use of reduced channel information in multiuser OFDM wireless systems. The amount of channel information can be reduced by exploiting the correlation characteristics in the frequency domain. For a given amount of feedback information, the subcarriers are selected to maximize the system capacity. A simplified algorithm is proposed for the selection of subcarriers, making the proposed scheme realizable. Simulation results show that the proposed scheme provides significant performance improvement over conventional schemes especially when the channel is highly correlated in the frequency domain and/or the amount of allowable feedback information is small.

In this paper, we consider fast cell site selection (FCS) in the downlink of a packet based orthogonal frequency division multiplexing cellular system. Mobile stations near the cell boundary can have opportunities to encounter a link that can provide better condition than the link in service due to the nature of fading effect. Conventional FCS schemes can track the variation of channel fading, exploiting a site selection diversity gain. However, mobile stations near the cell boundary may suffer from other cell interference (OCI) which is highly time-varying and unpredictable. To mitigate the OCI effect, we propose a new FCS scheme that exploits cell selection diversity while preventing adjacent cells from using the same sub-channels. We also consider the selection of candidate cells. The performance of the proposed FSC scheme is analyzed in terms of the capacity using an upper bound. Finally, analytic results are verified by computer simulation.

In this paper, we propose a beamforming scheme that exploits the advantages of opportunistic beamforming and coherent beamforming in multi-user environment. It is analytically shown that the proposed scheme achieves multi-user diversity gain and beamforming gain simultaneously, providing much better performance over the conventional ones. The performance of the proposed scheme is analyzed using an upper bound method. Although the proposed scheme involves an additional feedback delay, the analytical result implies that the use of the proposed scheme is quite effective unless the user mobility is too high. Finally, the analytic results are verified by computer simulation.

In this paper, we propose a multi-beam multiplexing scheme that can simultaneously achieve spatial multiplexing gain and multi-user diversity (MUD) gain by generating coherent multiple beams in the multi-user domain. Multiple beams are generated to provide multiple channels in parallel, making it possible to achieve the MUD gain through each channel. Since the transmission power is spilt into multiple channels, the signal-to-noise power ratio (SNR) of each channel is reduced inversely proportional to the number of beams. However, multiple beams are utilized to make the multiplexing gain much larger than the decrease of SNR, increasing the overall system capacity. The proposed scheme is applicable to both multi-input multi-output (MIMO) and multi-input single-output (MISO) schemes, enabling the use of flexible antenna structures in the receiver.

Extensive use of the Internet and huge demands for multimedia services via portable devices require the development of packet-based radio access systems with high transmission efficiency. Advanced radio transmission technologies have recently been proposed to achieve this challenging task. However, few researches have been reported on the design of an integrated system that can efficiently exploit the advantages of these transmission technologies. This paper considers the design of a packet-based cellular system for next-generation radio access. We propose a novel system framework that can incorporate various advanced transmission technologies such as link adaptation, opportunistic packet scheduling, channel coding, and multiantenna techniques. For efficient use of these technologies together, we first investigate the interoperability between these technologies by proposing a so-called cause and effect analysis. Based on this investigation, we design a differentiated-segments-based orthogonal frequency-division multiplexing system, called DiffSeg, to accommodate heterogeneous operating conditions in a seamless manner. Simulation results show that the proposed DiffSeg system can provide a nearly optimum performance with flexible configuration in a wide range of wireless channel conditions.