On the energy efficiency of rate and transmission power control in 802.11
Introduction
In recent years, along with the growth in mobile data applications and the corresponding traffic volume demand, we have witnessed an increased attention towards “green operation” of networks, which is required to support a sustainable growth of the communication infrastructures. For the case of wireless communications, there is the added motivation of a limited energy supply (i.e., batteries), which has triggered a relatively large amount of work on energy efficiency [1]. It turns out, though, that energy efficiency and performance do not necessarily come hand in hand, as some previous research has pointed out [2], [3], and that a criterion may be required to set a proper balance between them.
This paper is devoted to the problem of rate adaptation (RA) and transmission power control (TPC) in 802.11 WLANs from the energy consumption’s perspective. RA algorithms are responsible for selecting the most appropriate modulation and coding scheme (MCS) to use, given an estimation of the link conditions, and have received a vast amount of attention from the research community (see e.g. [4], [5] and references therein). In general, the challenge lies in distinguishing between those loses due to collisions and those due to poor radio conditions, because they should trigger different reactions. In addition, the performance figure to optimise is commonly the throughput or a related one such as, e.g., the time required to deliver a frame.
On the other hand, network densification is becoming a common tool to provide better coverage and capacity. However, densification brings new problems, especially for 802.11, given the limited amount of orthogonal channels available, which leads to performance and reliability issues due to RF interference. In consequence, some RA schemes also incorporate TPC, which tries to minimise the transmission power (TXP) with the purpose of reducing interference between nearby networks. As in the case of “vanilla” RA, the main performance figure to optimise is also throughput.
It is generally assumed that optimality in terms of throughput also implies optimality in terms of energy efficiency. However, some previous work [6], [7] has shown that throughput maximisation does not result in energy efficiency maximisation, at least for 802.11n. However, we still lack a proper understanding of the causes behind this “non-duality”, as it may be caused by the specific design of the algorithms studied, the extra consumption caused by the complexity of MIMO techniques, or any other reason. In fact, it could be an inherent trade-off given by the power consumption characteristics of 802.11 interfaces, and, if so, RA-TPC techniques should not be agnostic to this case.
This work tackles the latter question from a formal standpoint. A question which, to the best of the authors’ knowledge, has never been addressed in the literature. For this purpose, and with the aim of isolating the variables of interest, we present a joint goodput (i.e., the throughtput delivered on top of 802.11) and energy consumption model for single 802.11 spatial streams in the absence of interfering traffic. Packet losses occur due to poor channel conditions and RA-TPC can tune only two variables: MCS and TXP.
Building on this model, we provide the following contributions: (i) we demonstrate through an extensive numerical evaluation that energy consumption and throughput performance are different optimisation objectives in 802.11, and not only an effect of MIMO or certain algorithms’ suboptimalities; (ii) we analyse the relative impact of each energy consumption component on the resulting performance of RA-TPC, which serves to identify the critical factors to consider for the design of RA-TPC algorithms; (iii) we experimentally validate our numerical results; and (iv) we assess the performance of several representative RA-TPC algorithms from the energy consumption’s perspective.
The rest of this paper is organised as follows. In Section 2, we develop the theoretical framework: a joint goodput-energy model built around separate previous models. In Section 3, we provide a detailed analysis of the trade-off between energy efficiency and maximum goodput, including a discussion of the role of the different energy parameters involved. We support our numerical analysis with experimental results in Section 4. Section 5 explores the performance of RA-TPC algorithms from the energy consumption’s perspective. Finally, Section 6 summarises the paper.
Section snippets
Joint goodput-energy model
In this section, we develop a joint goodput-energy model for a single 802.11 spatial stream and the absence of interfering traffic. It is based on previous studies about goodput and energy consumption of wireless devices. As stated in the introduction, the aim of this model is the isolation of the relevant variables (MCS and TXP) to let us delve in the relationship between goodput and energy consumption optimality in the absence of other effects such as collisions or MIMO.
Beyond this primary
Numerical results
Building on the joint model presented in the previous section, here we explore the relationship between optimal goodput and energy efficiency in 802.11a. More specifically, our objective is to understand the behaviour of the energy efficiency of a single spatial stream as the MCS and TXP change following our model to meet the optimal goodput.
Experimental validation
This section is devoted to experimentally validate the results from the numerical analysis and, therefore, the resulting conclusions. To this aim, we describe our experimental setup and the validation procedure, first specifying the methodology and then the results achieved.
On the performance of RA-TPC algorithms
So far, we have demonstrated through numerical analysis, and validated experimentally, the existence of a trade-off between two competing performance figures, namely, goodput and energy efficiency. This issue arises even for a single spatial stream in absence of interference. Furthermore, we have discussed in Section 3.6 some ideas about the kind of mechanisms that energy-aware RA-TPC algorithms may incorporate, to leverage the behaviour that we have identified in our analysis in these
Conclusions
In this paper, we have revisited 802.11 rate adaptation and transmission power control by taking energy consumption into account. While some previous studies pointed out that MIMO rate adaptation is not energy efficient, we have demonstrated through numerical analysis that, even for single spatial streams without interfering traffic, energy consumption and throughput performance are different optimisation objectives. Furthermore, we have validated our results via experimentation.
Our findings
Acknowledgements
This work has been performed in the framework of the H2020-ICT-2014-2 projects 5GNORMA (grant agreement no. 671584) and Flex5Gware (grant agreement no. 671563). The authors would like to acknowledge the contributions of their colleagues. This information reflects the consortium’s view, but the consortium is not liable for any use that may be made of any of the information contained therein.
References (17)
- et al.
Greening wireless communications: status and future directions
Comput. Commun.
(2012) - et al.
Rate adaptation algorithms for IEEE 802.11 networks: A survey and comparison
2008 IEEE Symposium on Computers and Communications
(2008) - et al.
Energy efficiency is a subtle concept: fundamental trade-offs for cognitive radio networks
Commun. Mag., IEEE
(2014) - et al.
Balancing energy efficiency and throughput fairness in IEEE 802.11 WLANs
Pervasive Mob. Comput.
(2012) - et al.
H-RCA: 802.11 collision-aware rate control
Networking, IEEE/ACM Trans.
(2013) - et al.
Energy-based rate adaptation for 802.11n
Proceedings of the 18th Annual International Conference on Mobile Computing and Networking. Mobicom ’12
(2012) - et al.
Model-driven energy-aware rate adaptation
Proceedings of the Fourteenth ACM International Symposium on Mobile Ad Hoc Networking and Computing. MobiHoc ’13
(2013) - et al.
Goodput analysis and link adaptation for IEEE 802.11a wireless LANs
Mob. Comput., IEEE Trans.
(2002)
Cited by (5)
NDN-MMRA: Multi-Stage Multicast Rate Adaptation in Named Data Networking WLAN
2021, IEEE Transactions on MultimediaCharacterization and performance of a thermal camera communication system
2020, Sensors (Switzerland)Link adaptation algorithm for IEEE 802.11 wireless local area netwoks in fading channel
2018, Indonesian Journal of Electrical Engineering and Computer Science