摘要 : We obtain the Shannon capacity region of the down-link (broadcast)channel in fading and additive white Gaussian noise (AWGN) fortime-division, frequency-division, and code-division. For all of thesetechniques, the maximum capacity... 展开 We obtain the Shannon capacity region of the down-link (broadcast)channel in fading and additive white Gaussian noise (AWGN) fortime-division, frequency-division, and code-division. For all of thesetechniques, the maximum capacity is achieved when the transmitter variesthe data rate sent to each user as their channels vary. This optimalscheme requires channel estimates at the transmitter; dynamic allocationof timeslots, bandwidth, or codes; and variable-rate and powertransmission. For both AWGN and fading channels, nonorthogonalcode-division with successive decoding has the largest capacity region,while time-division, frequency-division, and orthogonal code-divisionhave the same smaller region. However, when all users have the sameaverage received power, the capacity region for all these techniques isthe same. In addition, the optimal nonorthogonal code is amultiresolution code which does not increase the signal bandwidth.Spread-spectrum code-division with successive interference cancellationhas a similar rate region as this optimal technique, however, the regionis reduced due to bandwidth expansion. We also examine the capacityregion of nonorthogonal code-division without interference cancellationand of orthogonal code-division when multipath corrupts the codeorthogonality. Our results can be used to bound the spectral efficiencyof the downlink channel using time-division, frequency-division, andcode-division, both with and without multiuser detection 收起