Multiuser Scheduling Scheme Research Articles

Performance evaluation of threshold-based power allocation algorithms for down-link switched-based parallel scheduling

In this paper, we propose threshold-based power allocation algorithms for a recently proposed down-link parallel multiuser scheduling (SBS) scheme. The proposed algorithms re-allocate the extracted excess transmit power from some acceptable users to those unacceptable users among scheduled users. After the power allocation process, more unacceptable users will enjoy acceptable link reliability as the number of effective acceptable users with an acceptable receive SNR level among the scheduled users is increased without any additional down-link transmit power. Some selected numerical results, show that the proposed power allocation algorithms offer a certain improvement in average spectral efficiency (ASE) and an increase in the average number of effective acceptable users. Although the average bit error rate (BER) performance degrades, especially when the average SNR is close to the SNR threshold, this average BER performance still meets the average BER requirement.

Threshold-based parallel multiuser scheduling

In this paper, we propose two new threshold based parallel multiuser scheduling schemes, namely (i) an on-off based scheduling (OOBS) scheme and (ii) a switched based scheduling (SBS) scheme. The objective is to reduce the complexity of implementation without a considerable performance loss in comparison with conventional selection based scheduling. While the OOBS scheme schedules all the users with an SNR above a fixed preselected signal to noise ratio (SNR) threshold, the SBS scheme schedules only the K <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">s</sub> users with an acceptable SNR if there are enough acceptable users. Otherwise, the scheduler selects the best K <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">s</sub> users. We analyze statistical characteristics and performances of these proposed schemes. Numerical results show that the proposed schemes can offer some decrease in the average feedback load without an important degradation in average bit error rate and average spectral efficiency.

On the Capacity-Fairness Tradeoff in Multiuser Diversity Systems

Conventional multiuser diversity maximizes the system throughput but results in an unfair scheduling of the system resources across users. Proportional fair scheduling types of systems achieve strict fairness among different users, but this fairness comes at the cost of a significant system capacity penalty. In this paper, we propose two forms of a hybrid multiuser scheduling scheme that provides a flexible balance/tradeoff between the system achievable capacity and the fairness among users. Our results show that the capacity versus fairness tradeoff can be achieved by grouping users and then using a two-step selection process.