Search

Usually, centralized channel state information (CSI) is assumed to exploit the multiuser diversity with a smart transmission scheduler. However, such centralized CSI can be impractical for a broadband wireless communication system with a large number of mobile users (MUs). In this paper, we propose a decentralized method to exploit the multiuser diversity in a single cell scenario with orthogonal frequency-division multiplexing (OFDM) based downlink. The central part of our approach is the channel-aware ALOHA-based OFDM subcarrier assignment. According to it, each MU measures the channel at all OFDM subcarriers and tries to obtain proper ones by sending a service-request packet through the corresponding orthogonal uplink subchannel. This packet is sent when the measured channel-fading level exceeds a predetermined threshold [xi]. The base station processes these request packets with a collision-reception model, and assigns the corresponding subcarrier(s) to the MU whose request packet has been successfully received. Two implementation algorithms are developed, by solving the problem of optimization of [xi] under different system configurations. Computer simulations show that in comparison with the standard round-robin method, the proposed algorithms offer a substantial data-rate improvement, especially when the correlation property of the OFDM subcarriers is properly exploited. Index Terms--Access protocols, diversity methods, orthogonal frequency-division multiplexing (OFDM).

In the orthogonal frequency-division multiplexing (OFDM) scheme, some subcarriers may be subject to a deep fading. Adaptive techniques can be applied to mitigate this effect if the channel-state information (CSI) is available at the transmitter. In this paper, we study the performance of an OFDM-based communication system whose transmitter has only one bit of CSI per subcarrier, obtained through a low-rate feedback. Three adaptive approaches are considered to exploit such a CSI feedback: adaptive subcarrier selection; adaptive power allocation (APA); and adaptive modulation selection (AMS). Under the conditions of a constant raw data rate and perfect feedback channel, the performance of these approaches are analyzed and compared in terms of raw bit-error rate. It is shown that one-bit CSI feedback can greatly enhance the system performance. Moreover, imperfections of the feedback channel are considered, and their impact on the performance of these techniques is studied. It is shown that by exploiting the knowledge that the feedback channel is imperfect, the performance of the APA and AMS techniques can be substantially improved. Index Terms--Channel-state information (CSI), feedback channel, orthogonal frequency-division multiplexing (OFDM).

Exact expressions are obtained for the symbol-error probability of orthogonal space-time block codes at the output of the coherent maximum-likelihood decoder in the general case of arbitrary input signal constellation and code. Such expressions are derived for the cases of both deterministic (fixed) and random Rayleigh/Ricean fading channels, and both the two- and higher dimensional constellations. Index Terms--Exact probability of error, multidimensional constellations, orthogonal space-time block codes (OSTBCs).

In the orthogonal frequency-division multiplexing (OFDM) scheme, some subcarriers may be subject to a deep fading. Adaptive techniques can be applied to mitigate this effect if the channel-state information (CSI) is available at the transmitter. In this paper, we study the performance of an OFDM-based communication system whose transmitter has only one bit of CSI per subcarrier, obtained through a low-rate feedback. Three adaptive approaches are considered to exploit such a CSI feedback: adaptive subcarrier selection; adaptive power allocation (APA); and adaptive modulation selection (AMS). Under the conditions of a constant raw data rate and perfect feedback channel, the performance of these approaches are analyzed and compared in terms of raw bit-error rate. It is shown that one-bit CSI feedback can greatly enhance the system performance. Moreover, imperfections of the feedback channel are considered, and their impact on the performance of these techniques is studied. It is shown that by exploiting the knowledge that the feedback channel is imperfect, the performance of the APA and AMS techniques can be substantially improved. [ABSTRACT FROM AUTHOR]