Obsidian/Zotero/001_artiklid/kassaenergyefficientcellularnetwork2017.md

---

category: literaturenote citekey: kassaenergyefficientcellularnetwork2017 title: ENERGY EFFICIENT CELLULAR NETWORK USING ADAPTIVE USER CLUSTERING ALGORITHM FOR SPARSELY POPULATED AREA authors: "Kassa, Hailu B." year: 2017 date: 2017-10-00 2017-10 url: "https://repository.arizona.edu/handle/10150/627026" zotero_key: 8DVBUK4G zotero_storage: XMBJVJM7 collections: imporditud folder: 001_artiklid firstAuthor: "Kassa, Hailu B."

status: converted

ENERGY EFFICIENT CELLULAR NETWORK USING ADAPTIVE USER CLUSTERING ALGORITHM FOR SPARSELY POPULATED AREA

| Item Type | text; Proceedings | | |---------------|--------------------------------------------------------------------------------------------------|--| | Authors | Kassa, Hailu B. | | | Publisher | International Foundation for Telemetering | | | Journal | International Telemetering Conference Proceedings | | | Rights | Copyright © held by the author; distribution rights International
Foundation for Telemetering | | | Download date | 06/05/2026 17:14:40 | | | Item License | http://rightsstatements.org/vocab/InC/1.0/ | | | Link to Item | http://hdl.handle.net/10150/627026 | |

ENERGY EFFICIENT CELLULAR NETWORK USING ADAPTIVE USER CLUSTERING ALGORITHM FOR SPARSELY POPULATED AREA

Hailu B Kassa

Doctorate of Engineering Candidate, Department of Electrical and Computer Engineering, Morgan State University, Baltimore, USA, hailu.kassa@morgan.edu

Faculty Advisor: Farzad Moazzami, Yacob Astatke

ABSTRACT

In this paper, an algorithm for distance aware energy efficient Base Stations (BSs) is proposed, which exploits the knowledge of the distance between the Mobile User (MUs) and the BS. The proposed algorithm changes the cell radius depending on user distribution in the cell. In this case adaptive (dynamically changing) concentric circles are virtually drawn so that the UEs in the same concentric circle can get constant power from the BS. It means that the mobile users(MUs) are clustered based on the distance from the BS. The energy consumption has been evaluated without sacrificing significant offered Quality of Service (QoS) on the cellular networks. The proposed scheme aids to achieve energy saving as a result of reduced transmit power based on reducing the radius of the concentric. As the radius reduces the transmit power is also reduced logarithmically. The system uses Omni directional antenna which covers all 360 degree of the cell at once. The simulation result shows that as the distance from the BS decreases, the transmission power decreases and the energy also decreases. The distance or the radius variation dynamically is based on the number of users in that track comparing to the minimum required number of threshold users. The result shows that an average of 0.1762 dB or 1.04 Watt which is 9.45% of the maximum transmit power can be saved by dynamically varying the base station the radius at the area of low traffic load for a single scan.

INTRODUCTION

Recent analysis by manufacturers and network operators has shown that current wireless networks are not very energy efficient, particularly the base stations by which terminals access services from the network. Green radio technology describes one of the most promising research directions in reducing the energy consumption as well as the carbon emissions of future base stations. Given the worldwide growth in the number of mobile subscribers, the move to higherdata-rate mobile broadband, and the increasing contribution of information technology to the overall energy consumption of the world, there is a need on environmental grounds to reduce the energy requirements of radio access networks [3]. The Green Radio program sets the aspiration

of achieving a hundredfold reduction in power consumption over current designs for wireless communication networks. This challenge is rendered nontrivial by the requirement to achieve this reduction without significantly compromising the quality of service (QoS) experienced by the network's users.

Traditional traffic engineering in cellular networks was focused on traffic measurement, characterization and control, aiming to carry the largest amount of traffic load while satisfying the required quality of service (QoS) using limited radio resources, e.g., wireless channels [4]. The new dimension of energy consumption can be similarly added into this problem, since energy is also an important type of radio resource, except that the objective now becomes to minimize energy consumption per traffic bit, for energy efficiency maximization under given QoS constraints [5]. Typically, cellular network operators allocate their radio resources based on the worst case principle, which is to allocate available resources based on usage requirements at peak traffic load [6]. This leads to significant waste of resources, including energy consumption, during periods with low traffic load [7]. To save energy, a shutdown or switch-off under-utilized BSs scheme was proposed at low traffic load conditions [8]. A method to accommodate peak traffic load with secondary (micro-, pico- or femto-) BSs that have smaller coverages was presented in [9]. Furthermore, combining these two methods was explored by, for example, employing the shut-down strategy to smaller secondary BSs in [10]. In [12], the authors explored the tradeoff between the operating power and the embodied power contained in the manufacturing process of infrastructure equipment from a life-cycle perspective. However, most of the above traffic studies in wireless networks neglect realistic transmission of a single base station based on the traffic density of virtual area clustering to divide the transmit power accordingly. To the best of our knowledge, there is no research work has been conducted using this technique to bring adaptive transmit power adjustment on a single base station based on traffic density.

The rest of the paper is organized as follows. Section II introduces the system model and in section III the power model is described in detail. In Section IV simulation results are presented based on numerical values considered. Finally, Section V concludes this paper.

SYSTEM MODEL

In this paper, energy efficiency is computed for the downlink cellular system considering the density of the MUs which is found by counting and adding down those users and dividing by their respective area to find the specific density. Counting users is performed considering channel state information as signal to noise ratio (SNR). If the specific density is less than the average density, which is the threshold density (found from the two variables: total coverage and total number of MUs), then the cell radius shrinks to a new lower radius position and the transmit power is adjusted based on this radius. The proposed system model consists of one serving base station and multiple mobile users which create point to multipoint communication. The BS transmits signal with Omni directional antenna. The total transmit power is shared by all users in the range of five kilometers. The cell is divided by a number of virtual concentric circles in a given cell using an adaptive user clustering technique. The users in the cell are randomly distributed. The system power model estimates the total power supply where our focus is on the downlink communication, i.e., from BS to MU.

The proposed algorithm, which can perform the combination of users clustering and adaptive transmit power allocation is specified below. The parameters used here are basically number of users and the cell radius. Assume Rx, Rⁱ , Uo, Ux are radius of the concentric circle, minimum radius, number of users to be marginalized initially and the number of users between the consecutive concentric circles respectively.

• 1: Initialize; R=Rmax (max radius), P=Pmax,Uo=0
• 2: Find the no. of users, Ux between Rx and Rx-1
• 3: Uo=Uo+Ux
• 4: Compare Uo and Th
• 5: if Uo > Th then
• 6: Assign power based on R^x
• 7: T time interval
• 8: Go to step 1
• 9: Else
• 10: Rx=Rx-1
• 11: Compare R^x and Rⁱ
• 12: if R^x = Rⁱ
• 13: Go to Step 6
• 14: Else
• 15: Go to Step 2

POWER MODEL

The problem formulation for the overall system goes through the following procedures:

• Count and add the number of users until it is greater than or equal to the threshold value (Th) from outer to inner tracks. N=N1+N2+N3+…+Nx,
• If N=0 at the beginning, then the base station sleeps (uses minimum transmit power) and no need of clustering.
• The base station wakes up immediately when the users enter to the active zone (radius)
• Find the number of users N^x at each track
• If the number of users of a track N^x at the far distance from the base station is instantly falls below the threshold value, then R decreases to Rm-1(to the radius of the next concentric circle) and the new path loss will be calculated using Okumura Hatta models based on Rm-1 such that PL=247.41+35.22*log(Rm-1) and the new transmit power, P^T will change since it depends on the path loss effect.

$$P_T = P_R * L \tag{1}$$

Where PT, is transmitted power, PR, is minimum received power, and the path loss, L, from reduced Okumura Hatta propagation model can also be written as:

$$L=247.41+35.22*log(R_m), (2)$$

Where Rm,, is the mobile user's distance from the BS in the mth concentric circle.

NUMERICAL RESULTS

In this part, simulation results for power minimization techniques of a base station with clustered users in the cell have been shown. All the simulations are carried out using MATLAB software. The analysis for each result is discussed in detail after different system parameter considerations. Table1 provides system parameters that are used in the simulation and their corresponding values. The values for the different simulation parameters are selected in such a way that the whole system meets some standard regulations and have practical applicability.

Figure.1: BS with Full Coverage

TABLE 1: SIMULATION PARAMETERS AND THEIR CORRESPONDING VALUES

Parameters	Symbol	Value (if any)
Maximum Distance	R	2km,3km,5 km
frequency	f	900 Mhz
Base station Antenna Height	hte	200 meter
Mobil station Antenna Height	hre	30 meter
Wave length	λ	Depends on frequency
Antenna gain at the transmitter	Gte	-
Speed of light	C	3*108 m/s
Antenna gain at the receiver	GArea	1
Environment gain	GAREA	1
Number of users	n	Adaptive
Received power	pr	10mw
User threshold	thr	-
Minimum DSP power for each iteration	Pdsp	0.001
Omni directional Antenna	-	-

Figure.1 shows the random distribution of users over the cell's maximum coverage. In this case the cell radius is 5kms and the transmitted power reaches maximum. But in Figure.2, the cell is clustered in three regions depending on the number of users and minimum radius. By smoothly moving down to the center, the MEs are counted and the density is compared with the threshold users' density. The blue colored line is the position where the density of the users counted is less or equal to the threshold density and as a result the radius shrinks to 4.3 km from 5 km. Therefore, by doing this the BS saves 0.1762 dB or 1.04 Watt for the radius reduction. From Figure.1 and Figure.2, the Omnidirectional antenna transmits power for the whole of five kilometers radius. If the number of users at the last track is less than the threshold value (density), then the radius shrinks to the second outer most track as shown on Figure.2 and the transmit power is adjusted based on that reduced distance.

Fig.2: BS with Reduced Distance

We also compared the propagation models in Figure.3 based on the distance of randomly distributed MUs throughout the cell for the transmit power calculation as stated in equation (1). From this we have found that the Okumura Hatta model is better than the Cost 231 and Free Space for its lower power and practical feasibility. Using the Okumura Hatta Model we show the relationship between the transmit power and the distance between of MUs and the BS.

As it is shown on Figure.4, the transmit power logarithmically increases as the distance increases. Figure.5 depicts the transmit power pattern of different scan times. As the number of users and the distribution varies from time to time, the power pattern also varies with those parameters. At iteration three there are no any users between 0.5 and 2.4 kilometers and thus there is no power dissipation at this point however the middle area power is not the main concern.

Figure.3: Transmit Power at different Propagation Models

Figure.4:Transmit Power Vs Distance From BS

Here the main point is that as the minimum and maximum distances of each iteration vary the saved power also varies. From Figure 6 the bars represent, the maximum transmit power at maximum coverage, the power with respect to the reduced distance and the gain which is obtained from the difference of the two respectively. This result is of only one iteration or scan time and as the BS performs a number of scans within a second, the saved power will ultimately increase with time. Figure 6 is the graphical representation of Figure 2.

Figure.5: Transmit Power Pattern at Three iterations

Figure.6: Saved Power for one Scan Time

CONCLUSIONS

The simulation results discussed in section IV have shown that we have studied different approaches in order to reduce the transmit power. We have shown that the energy efficiency can be achieved at the BS based on several factors, including traffic load. Dynamically adjusting the cell radius based on traffic statistics in the specific area has brought greater energy savings. Our simulations show that from a single scan the BS has saved 0.1762 dB or 1.04 Watt which is 9.45% of the maximum transmit power. As a future work, it can be extended to the technique to which the power adjustment is sensitive to only the last user's distance.

REFERENCES

• [1] A. Fehske, G. Fettweis, J. Malmodin, and G. Biczok, "The global footprint of mobile communications: the ecological and economic perspective," IEEE Commun. Mag., vol. 49, no.8, pp. 55–62, Aug. 2011
• [2] C. Han, T. Harrold, S. Armour, I. Krikidis, et al., "Green radio: radio techniques to enable energy efficient wireless networks," IEEE Commun. Mag., vol. 49, no. 6, pp. 46–54, June 2011
• [3] D. E. Everitt, "Traffic engineering of the radio interface for cellular mobile networks," Proc. IEEE, vol. 82, no.9, pp. 1371–1382, Sep. 1994
• [4] H. Bogucka and A. Conti, "Degrees of freedom for energy savings in practical adaptive wireless systems," IEEE Commun. Mag., vol. 49, no. 6, pp. 38–45, June 2011
• [5] P. Trimintzios, G. Pavlou, P. Flegkas, P. Georgatsos, A. Asgari, and E. Mykoniati, "Servicedriven traffic engineering for intra-domain quality of service management," IEEE Netw., vol. 13, no. 3, pp. 29–36, May 2003
• [6] E. Hwang, K. J. Kim, J. J. Son, and B. D. Choi, "The power-saving mechanism with periodic traffic indications in the IEEE 802.16e/m," IEEE Trans. Veh. Technol., vol. 59, no. 1, pp. 319–334, Jan. 2010
• [7] E. Oh, B. Krishnamachari, X. Liu, and Z. Niu, "Toward dynamic energy efficient operation of cellular network infrastructure," IEEE Commun. Mag., vol. 49, no. 6, pp. 56–61, June 2011
• [8] W. Wei and G. Shen, "Energy efficiency of heterogeneous cellular network," in Proc. 2010 IEEE VTC – Fall, pp. 1–5
• [9] I. Ashraf, F. Boccardi, and L. Ho, "SLEEP mode techniques for small cell deployments," IEEE Commun. Mag., vol. 49, no. 8, pp. 72–79, Aug. 2011
• [10] R. Wang, J. Tsai, C. Maciocco, T. Y. Tai, and J. Wu, "Reducing power consumption for mobile platform via adaptive traffic coalescing," IEEE J. Sel. Areas Commun., vol. 29, no. 8, pp. 1618–1629, Sep. 2011
• [11] I. Humar, X. Ge, L. Xiang, J. Ho, and M. Chen, "Rethinking energy efficiency models of cellular networks with embodied energy," IEEE Netw., vol. 25, no. 2, pp. 40–49, Mar.–Apr. 2011
• [12] L. Xiang, F. Pantisano, R. Verdone, X. Ge, and M. Chen, "Adaptive traffic load-balancing in green cellular networks," in Proc. 2011 IEEE PIMRC, pp. 41–45