Abstract
Energy harvesting and sleeping strategy are the important methods to deal with the increasing grid power consumption problem. However, power supply solely from energy harvesting modules would be more likely to cause energy outage of base station (BS), and the load of sleeping BS would lead to coverage hole problem, which directly affect user’s quality of service (QoS). In this paper, to achieve the trade-off between grid power consumption and user’s QoS for each tier, optimal joint of ON/OFF mode of a BS and power source switching strategies are designed for two-tier femtocell networks, which are powered by the cooperation of the harvested energy and grid power. Then, exploiting stochastic geometry and random walk theory, BS availability and availability region are characterized with the proposed strategies. Moreover, the relationships among BS availability, battery capacity of grid power and cut-off energy level are exploited to verify the proposed strategies. By numerical simulations, we find that when energy storage capacity achieves a certain value, the effect of grid power battery capacity on availability almost vanishes especially in femto tier. So these redundant grid power can be saved when the stored energy in battery meets current traffic requirement.
Similar content being viewed by others
Notes
Femtocells are assumed to operate in open access mode at the beginning of this section, so these uncovered macrocell users can connect with adjacent FAP when some MBSs are shut down. Then we ignore the effect of sleeping MBS on system coverage in this paper, and uncoordinated strategy can be allowed in macrocell networks.
The figure only shows the sketch of changes trend in energy, and does not mean the actual change trend in every minute as shown in the figure.
These values refer to the reference [6], not represent the real value, only be used to verify our work.
References
Son, K., Kim, H., Yi, Y., & Krishnamachari, B. (2011). Base station operation and user association mechanisms for energy-delay tradeoffs in green cellular networks. IEEE Journal on Selected Areas in Communications, 29(8), 1525–1536.
Mohammed, H. A., Rosdiadee, N., & Mahamod, I. (2015). Cooperation management among base stations based on cells switch-off for a green LTE cellular network. Wireless Personal Communications, 81(1), 303–318.
Ismail, M., Zhuang, W. H., Serpedin, E., & Qaraqe, K. (2015). A survey on green mobile networking: from the perspectives of network operators and mobile users. IEEE Communications Surveys and Tutorials, 17(3), 1535–1536.
Huang, K. B. (2013). Spatial throughput of mobile ad hoc networks powered by energy harvesting. IEEE Transactions on Information Theory, 59(11), 7597–7612.
Chamola, V., & Sikdar, B. (2015). Outage estimation for solar powered cellular base stations. In Proceedings of IEEE ICC (pp. 172–177).
Dhillon, H. S., Li, Y., Nuggehalli, P., Pi, Z., & Andrews, J. G. (2014). Fundamentals of heterogeneous cellular networks with energy harvesting. IEEE Transactions on Wireless Communications, 13(5), 2782–2797.
Majharul Islam Rajib, M., & Nasipuri, A. (2015). Delay performance of intermittently connected wireless sensor networks with cooperative relays. In Proceedings of IEEE ICC (pp. 1994–1999).
Liu, D. T., Chen, Y., Chai, K. K., Zhang, T. K., & Pan, C. K. (2014). Adaptive user association in HetNets with renewable energy powered base stations. In Proceedings of IEEE ICT (pp. 93–97).
Mao, Y. Y., Yu, G. D., & Zhong, C. J. (2013). Energy consumption analysis of energy harvesting systems with power grid. IEEE Wireless Communications Letters, 2(6), 2162–2337.
Yu, P. S., Jemin, L., Quek, T. Q. S., & Hong, Y. W. P. (2015). Energy harvesting personal cells-traffic offloading and network throughput. In Proceedings of IEEE ICC (pp. 2184–2189).
Zhou, S., Gong, J., & Niu, Z. S. (2013). Sleep control for base station powered by heterogeneous energy sources. In Proceedings of IEEE ICTC (pp. 666–670).
Thuc, T. K., Tabassum, H., & Hossain, E. (2014). A stochastic power control game for two-tier cellular networks with energy harvesting small cells. In Proceedings of IEEE Globecom (pp. 2637–2643).
Gong, J., Zhou, S., & Niu, Z. S. (2013). Optimal power allocation for energy harvesting and power grid coexisting wireless communication systems. IEEE Transactions on Communications, 61(7), 3040–3049.
Gong, J., Thompson, J. S., Zhou, S., & Niu, Z. S. (2014). Base station sleeping and resource allocation in renewable energy powered cellular networks. IEEE Transactions on Communications, 62(11), 3801–3813.
Cui, Y., Vincent, K. N. Lau, & Zhang, F. (2015). Grid power-delay tradeoff for energy harvesting wireless communication systems with finite renewable energy storage. IEEE Journal on Selected Areas in Communications, 33(8), 1651–1666.
Zhao, J., Zhao, M., & Zhou, W. Y. (2014). Energy efficiency optimization in energy harvesting cooperative relay systems. In Proceedings of sixth WCSP (pp. 1–6).
Wang, Y., Zhang, Y., Chen, Y. C., & Wei, R. (2015). Energy-efficient design of two-tier femtocell networks. EURASIP Journal on Wireless Communications and Networking. doi:10.1186/s13638-015-0242-4.
Cheung, W. C., Quek, T. Q. S., & Kountouris, M. (2012). Throughput optimization, spectrum sharing, and femtocell access in two-tier femtocell networks. IEEE Journal on Selected Areas in Communications, 30(3), 561–574.
Jo, H. S., Sang, Y. J., Xia, P., & Andrews, J. G. (2012). Heterogeneous cellular networks with flexible cell association: A comprehensive downlink SINR analysis. IEEE Transactions on Wireless Communications, 11(10), 3484–3495.
Kijima, M. (1997). Markov processes for stochastic modeling. London: Chapman Hall.
Resnick, S. I. (2005). Adventures in stochastic processes. Boston: Birkhauser.
Acknowledgements
This work is supported by National 863 Project (2014AA01A701), National Nature Science Foundation of China (61372113, 61421061).
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1
From Eq. (3) we can obtain that \({\rho _m} = \frac{1}{{1 + {\varphi _{1m}} + {\varphi _{2m}}}}\), where \({\varphi _{1m}} = \frac{{{N_0}}}{{{\mu '_m}( {E[ {J_{{k_{11}}}^m} ] + E[ {J_{{k_{12}}}^m} ]} )}}\), \({\varphi _{2m}} = \frac{{{N_m} - {N_0}}}{{{\mu _m}( {E[ {J_{{k_{11}}}^m} ] + E[ {J_{{k_{12}}}^m} ]} )}}\). So the variation trend of \({\rho _m}\) depends on \({\varphi _{1m}}\) and \({\varphi _{2m}}\). Without loss of generality, we present the derivations of \({\varphi _{1m}}\) and \({\varphi _{2m}}\) in (16).
Then for the situation of \(E'[ {J_{{k_{11}}}^m} ]<0\) and \(E[ {J_{{k_{11}}}^m} ] + E[ {J_{{k_{12}}}^m} ] > ( {{N_m} - {N_0}} ){{E'[ {J_{{k_{11}}}^m} ] }}\), \({\varphi _{1m}}\) and \({\varphi _{2m}}\) are increasing functions of \(N_m\), so \({\rho _m}\) decreases with \(N_m\). For the situation of \(E'[ {J_{{k_{11}}}^m} ]> 0\) and \(E[ {J_{{k_{11}}}^m} ] + E[ {J_{{k_{12}}}^m} ] < ( {{N_m} - {N_0}} ){{E'[ {J_{{k_{11}}}^m} ] }}\), \({\varphi _{1m}}\) and \({\varphi _{2m}}\) are decreasing functions of \(N_m\), so \({\rho _m}\) increases with \(N_m\). For the situation of \(E'[ {J_{{k_{11}}}^m} ]>0\) and \(E[ {J_{{k_{11}}}^m} ] + E[ {J_{{k_{12}}}^m} ] > ( {{N_m} - {N_0}} ){{E'[ {J_{{k_{11}}}^m} ] }}\), the monotonicity of \({\rho _m}\) will depend on \(\left| \frac{{\varphi ' _{1m}}}{{\varphi ' _{2m}}} \right| = \frac{{{N_0}{\mu _m}E'\left[ {J_{{k_{11}}}^m} \right] }}{{{{\mu '}_m}\left[ {E\left[ {J_{{k_{11}}}^m} \right] + E\left[ {J_{{k_{22}}}^m} \right] - \left( {{N_m} - {N_0}} \right) E'\left[ {J_{{k_{11}}}^m} \right] } \right] }}\). When \(\left| \frac{{\varphi ' _{1m}}}{{\varphi ' _{2m}}} \right| >1\), \({\rho _m}\) will be an increasing function of \(N_m\). Otherwise, \({\rho _m}\) will decrease with \(N_m\). For the situation of \(E'[ {J_{{k_{11}}}^m} ]<0\) and \(E[ {J_{{k_{11}}}^m} ] + E[ {J_{{k_{12}}}^m} ] < ( {{N_m} - {N_0}} ){{E'[ {J_{{k_{11}}}^m} ] }}\), when \(\left| \frac{{\varphi ' _{1m}}}{{\varphi ' _{2m}}} \right| >1\), \({\rho _m}\) will be a decreasing function of \(N_m\). Otherwise, \({\rho _m}\) will increase with \(N_m\).
Appendix 2
Set \(g( {{k_1}} ) = \frac{{{\mu _f}E( {J_{{k_1}}^f} )}}{{{N_f}}}\), \(g( {{k_{11}}}) =\frac{{{\mu _f}E( {J_{{k_{11}}}^f} )}}{{{N_f}}}\), \(g( {{k_{12}}} ) = \frac{{{\mu _f}E( {J_{{k_{12}}}^f} )}}{{{N_f}}}\), \(g( {{k_{13}}} ) = \frac{{{\mu _f}E( {J_{{k_{13}}}^f} )}}{{{N_f}}}\), \(g( k ) = g( {{k_1}} ) + \frac{{1 - p}}{p}[ {g( {{k_{11}}} ) + g( {{k_{12}}} ) + g( {{k_{13}}} )} ]\), then (7) can be transformed into \({\rho _f} = \frac{1}{{1 + \frac{1}{{g\left( k \right) }}}}\). Because the value of \(\frac{{1 - p}}{p}\) is nonnegative, then to determine the value of \(g'\left( {{k}} \right)\), we need to ensure the value of \(g'\left( {{k_1}} \right)\), \({g'\left( {{k_{11}}} \right) }\), \({g'\left( {{k_{12}}} \right) }\) and \({g'\left( {{k_{13}}} \right) }\). Obviously, following the Remark 1 introduced in Sect. 3, function \(g\left( {{k_1}} \right)\) will increase with the increase in \({N_f}\). Moreover, from (10) we known that \(g\left( {{k_{11}}} \right)\) (or \(g\left( {{k_{13}}} \right)\)) decreases with the increase in \({N_f}\), and \(\frac{{{\mu _f}E[ {J_{{k_{12}}}^f} ]}}{{{N_f} - {N'_0}}}\) decreases with the increase in \({N_f}\) can also be concluded from the Remark 1. The relationship between \(g\left( {{k_{12}}} \right)\) and \({N_f}\) is analyzed in the following. The derivations of \(g\left( {{k_{12}}} \right)\) is presented as
For the case of \(g'\left( {{k_{12}}} \right) <0\) \(({N_f}E'[ {J_{{k_{12}}}^f} ] < E[ {J_{{k_{12}}}^f} ])\), when satisfy \(g'\left( {{k_1}} \right) > \frac{{1 - p}}{p}[g'\left( {{k_{11}}} \right) + g'\left( {{k_{12}}} \right) + g'\left( {{k_{13}}} \right) ]\), i.e., \(p[E'(J_{{k_1}}^f){N_f} - E(J_{{k_1}}^f)] > (1 - p)[E'(J_{{k_{12}}}^f){N_f} - E(J_{{k_{12}}}^f) - 2E(J_{{k_{11}}}^f)]\), \(g\left( {{k}} \right)\) is an increasing function of \(N_f\), and we can also obtain that \(\rho _f\) is an increasing function of \(N_f\). On the contrary, when \(g'\left( {{k_1}} \right) < \frac{{1 - p}}{p}[g'\left( {{k_{11}}} \right) + g'\left( {{k_{12}}} \right) + g'\left( {{k_{13}}} \right) ]\), i.e., \(p[E'(J_{{k_1}}^f){N_f} - E(J_{{k_1}}^f)] < (1 - p)[E'(J_{{k_{12}}}^f) {N_f}- E(J_{{k_{12}}}^f) - 2E(J_{{k_{11}}}^f)]\), \(\rho _f\) is a decreasing function of \(N_f\).
For the case of \(0< g'\left( {{k_{12}}} \right)\) \(\left( {N_f}E'\left[ {J_{{k_{12}}}^f} \right] > E\left[ {J_{{k_{12}}}^f} \right] \right)\), when \(g'\left( {{k_1}} \right) +\frac{{1 - p}}{p}g'\left( {{k_{12}}} \right) > \frac{{1 - p}}{p}[g'\left( {{k_{11}}} \right) + g'\left( {{k_{12}}} \right) + g'\left( {{k_{13}}} \right) ]\), i.e., \(p[E'(J_{{k_1}}^f){N_f} - E(J_{{k_1}}^f)] + (1 - p)[E'(J_{{k_{12}}}^f){N_f} - E(J_{{k_{12}}}^f)] > - 2(1 - p)E(J_{{k_{11}}}^f)\), \(\rho _f\) increases with the increase in \(N_f\). On the contrary, when \(g'\left( {{k_1}} \right) + \frac{{1 - p}}{p} g'\left( {{k_{12}}} \right) < \frac{{1 - p}}{p}[g'\left( {{k_{11}}} \right) + g'\left( {{k_{13}}} \right) ]\), i.e., \(p[E'(J_{{k_1}}^f){N_f} - E(J_{{k_1}}^f)] + (1 - p)[E'(J_{{k_{12}}}^f){N_f} - E(J_{{k_{12}}}^f)] < - 2(1 - p)E(J_{{k_{11}}}^f)\), \(\rho _f\) is a decreasing function of \(N_f\).
Rights and permissions
About this article
Cite this article
Zhang, Y., Wang, Y., Chen, Y. et al. Optimal Switching Strategy for Heterogeneous Energy Supplying Energy-efficient Two-tier Femtocell Networks. Wireless Pers Commun 95, 1635–1654 (2017). https://doi.org/10.1007/s11277-016-3872-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11277-016-3872-x