Performance analysis of stable matching based resource unit scheduling and energy-efficient power allocation scheme in dense WLAN

Abstract

It is prospected that the number of access points (APs) as well as that of stations (STAs) greatly increases. An environment in which both APs and STAs are densely existing is called dense wireless local area network (WLAN). In this paper, we propose an effective resource unit (RU) scheduling based on stable matching algorithm introduced in economics firstly, to improve the efficiencies of dense WLAN which follows IEEE 802.11ax standard. An RU is assigned to a client based on an orthogonal frequency- division multiple access (OFDMA). There are some following processes to apply the stable matching algorithm to RU scheduling. First, every STA makes a preference list for every RU based on channel condition. Similarly, every RU makes a preference list for every STA based on channel condition. Second, all STAs request RUs to AP in accordance with their preference lists. If more than 2 STAs request the same RU, then the AP allocates the requested RU to an STA which has a higher ranking in the RU’s preference list. Since the STA did not receive RU from the AP requests another RU having the next highest ranking, every STA does not request the same RU more than one time. Additionally, we extend this concept to multi-user multiple-input and multiple-output (MU-MIMO) system which allows to allocate a single RU to more than 1 STA.

Meanwhile, for reducing the greenhouse gas emission and increasing the battery lifetime, energy efficiency (EE) has been addressed as an important indicator to design recent communication networks. The objective function of EE is expressed as the form of the throughput divided by the power consump- tion. In this thesis, we utilize the fractional programming to derive the optimal allocated powers to the subcarriers consisting of RU for maximizing the objective function of EE.

Finally, we analyze the performances of proposed RU scheduling and energy-efficient power allocation through simulation in dense WLAN. Compared with other conventional methods, simulation results show that the throughput and EE are both enhanced when applying our proposed method.