Abstract: In the heterogeneous network based on multi-source power supply, two user selection algorithms based on energy cost were proposed considering the influences of supply rate of renewable energy, peak-valley time-of-use tariff strategy of the traditional energy and the fairness among users. The processes of user selection can be divided into two stages which are allocating the users on the cell edge and determining the users who can get service, and a balanced load distribution was obtained. According to the relationships among power consumption, energy cost and the fairness among users, two exponential utility functions were constructed in the direct and the indirect method, respectively. Simulations show that the direct and indirect method can significantly reduce the power consumption and energy cost of base station system. Compared with the indirect method, the direct method employs an adaptive price factor, which changes with the electricity price of traditional price and the supply rate of renewable energy. Thus, it can reduce the power consumption at a lower level of energy cost.

Key words: heterogeneous network, multi-source supply, fairness, user selection