Resource Allocation for Cost Minimization in Limited Feedback MU-MIMO Systems With Delay Guarantee

In this paper, we design a resource allocation framework for the delay-sensitive multiuser multiple-input multiple-output (MU-MIMO) broadcast system with limited feedback. Considering the scarcity and interrelation of the transmit power and feedback bandwidth, it is imperative to optimize the two resources in a joint and efficient manner while meeting the delay-QoS requirement. Based on the effective bandwidth theory, we first obtain a closed-form expression of average violation probability with respect to a given delay requirement as a function of the transmit power and codebook size of the feedback channel. By minimizing the total resource cost, we derive an optimal joint resource allocation scheme, which can flexibly adjust the transmit power and the feedback bandwidth according to the characteristics of the system. Moreover, through an asymptotic analysis, some simple resource allocation schemes are presented. Finally, the theoretical claims are validated by numerical results.