The small-cell technology is promising for spectral-efficiency enhancement. However, it usually requires a huge amount of energy consumption. In this paper, queue state information and channel state information are jointly utilized to minimize the time average of overall energy consumption for a multi-carrier small-cell network, where the inter-cell interference is an intractable problem. Based on the Lyapunov optimization theory, the problem could be solved by dynamically optimizing the problem of user assignment, carrier allocation and power allocation in each time slot. As the optimization problem is NP-hard, we propose a heuristic iteration algorithm to solve it. Numerical results verify that the heuristic algorithm offers an approximate performance as the brute-force algorithm. Moreover, it could bring down the overall energy consumption to different degrees according to the variation of traffic load. Meanwhile, it could achieve the same sum rate as the algorithm which focuses on maximizing system sum rate.
The characteristic of group delay is analyzed based on an electronic circuit, and its time-domain nature is studied with time-domain simulation and experiment. The time-domain simulations and experimental results show that group delay is the delay of the energy center of the amplitude-modulated pulse, rather than the propagation delay of the electromagnetic field. As group velocity originates from the definition of group delay and group delay is different from the propagation delay, the superluminaiity or negativity of group velocity does not mean the supeduminal or negative propagation of the electromagnetic field.
