An Adaptive Bandwidth Allocation for Energy Efficient Wireless Communication Systems

Yung-Fa Huang, Che-Hao Li, Chuan-Bi Lin, and Chia-Chi Chang

An Adaptive Bandwidth Allocation for Energy Efficient Wireless Communication Systems

Yung-Fa Huang, Che-Hao Li, Chuan-Bi Lin, and Chia-Chi Chang

―In this paper, an energy efficient bandwidth allocation scheme is proposed for wireless communication systems. An optimal bandwidth expansion (OBE) scheme is proposed to assign the available system bandwidth for users. When the system bandwidth does not reach the full load, the remaining bandwidth can be energy-efficiently assigned to the other users. Simulation results show that the energy efficiency of the proposed OBE scheme outperforms the traditional same bandwidth expansion (SBE) scheme. Thus, the proposed OBE can effectively assign the system bandwidth and improve energy efficiency.

Index Terms―Bandwidth resource allocation, energy efficiency, optimal bandwidth expansion, same bandwidth expansion.

1. Introduction

The global warming issues become the most important. Thus, how to reduce Carbon oxide gas emissions attracted all over the world to work out[1]. Due to the global growth of mobile users, the transmission data traffic increases so much in wireless communication networks in recent years. Thus, it is need to provide a technology to reduce energy consumption in mobile communications[1].

Orthogonal frequency division multiplexing (OFDM) can provide higher data transmission[2], and has been widely used in mobile communication systems[3],[4]. In order to enhance the system capacity, the OFDM based orthogonal frequency division multiple access (OFDMA) technique[5]is adopted as the multiple access scheme in 4G mobile communication systems[6]-[8].

In OFDMA systems, different users use the same super frame assigned to the sub-channel. But the different usersoccupy different bandwidths. In the full load (bandwidth usage rate of 100%) or the low load (bandwidth utilization below 50%) cases, users expect to reduce the transmission signal power while maintaining transmission rate by providing higher bandwidth. But the bandwidth is limited, how to allocate the remaining bandwidth to the user and how to reduce the user’s energy consumption are the important issues studied in this paper.

2. Energy Efficiency with Bandwidth Extension

The capacity of a communication system can be obtained by Shannon-Hartley theory as[9]

where the required bandwidth of the user is denoted byB. The signal-to-noise ratio (SNR) of received signal is, whereEis the received signal energy andNois the power spectral density of the added white Gaussian noise (AWGN). To improve energy efficiency, the bandwidth of users is re-allocated to beαtimes of required bandwidth,α＞1. This is so-called bandwidth expansion mode (BEM)[10], whereαis the bandwidth expansion coefficient (BEC). After the bandwidth expansion, we obtain[10]-[12]

whereΓBEM=EBEM/No,EBEMis the required consumption energy. From (2), we obtainBEMΓby

Therefore, from (3), the required energyEBEMcan be obtained with the knownNo,E, andα.

After bandwidth expansion, the total energy reduction factor (ERF)forNusers can be obtained by

To investigate the relationship between the ERF and BEC,α, we define the ERF by

whereEis the energy consumption of a user andEBEMis the energy consumption for the user after bandwidth expansion. Fig. 1 shows that the relationship between ERF and BEC withNo=1 nJ for various SNR. From Fig. 1, it is observed that ERF is increasing with SNR. It is intuitive that the higher SNR users consume more energy, while the BEM will benefit more efficiency.

Fig. 1. Relationship betweenErand BEC withNo=1 nJ for various SNR.

Furthermore, in Fig. 1, the ERF areEr=156.4293 nJ andEr=181.5161 nJ withα=1.4 andα=1.8, respectively, for SNR=23dB. Moreover, the ERF areEr=181.5161 nJ andEr=189.3976 nJ withα=1.8, andα=2.2, respectively, for SNR=23 dB. Therefore, it means that the less energy efficiency the higher BEC.

In the BEM systems, there are totalNusers, the total bandwidth isB(Hz). With available bandwidth allocated to users, we can equal the BEC for users byαS, called same bandwidth expansion (SBE). Thus, theαS, can be obtained by

whereBiis the required bandwidth of theith user.

Thus, the allocated bandwidth of theith user will be. The total energy consumption for users of SBE is obtained by

Then, the ERF of SBE can be expressed by

Moreover, to perform more energy-efficiently, the available bandwidth can be adaptively allocated to users with different BECαifor theith user, respectively. We perform the optimal bandwidth allocation (OBA) with the optimal BEC for optimal energy efficiency, which is called optimal bandwidth expansion (OBE). The OBE optimization algorithms can be expressed by

where the total consumption energy can be obtained by

Theαiis the optimal BEC for theith user andBiis the allocated bandwith for theith user. Thus the ERF of OBE can be obtained by

The ERF ratio of OBE to SBE can be obtained by

In the two-user system, we develop the optimal BEC in this subsection. Assume that total bandwidth isB=20 MHz, The required bandwidth for user A and user B isBA=6 MHz andBB= 4MHz, respectively.

The BECs for the two users areAαandBα. Then, the allocated bandwidth can beMHz andMHz after BEM, whereandare the allocated bandwidth for user A and B, respectively. Thus with SNR=15 dB, by using SBE, the BECs can be obtained byαS= 2. Then the ERF can be obtained from (4) byEr=85.14% as shown in Table 1 (a). However, with OBE the BECs can be obtained from the OBE algorithms in (8) byαA=1.92, andαB=2.12. Thus, the allocated bandwidth of OBE isMHz =11.52 MHz andMHz = 8.48 MHz for user A and user B, respectively. The ERF obtained from (4) byEr=85.27% is shown in Table 1 (a). Therefore, it is easily seen that the OBE is more energy efficient than SBE.

We can compare the ERF for different BECs of user A as shown in Fig. 2. From Fig. 2, it is observed that with SBEαA=2,Eris increasing to the maximum of OBE,αA=1.92. Furthermore, the performance of OBE fordifferent total bandwidth environments is investigated as the results in Table 1 (b) and Table 1 (c). With the assumption ofB=15 MHz, SNR=15 dB,BA= 6 MHz , andBB= 4 MHz, the OBE outperforms the SBE with ERF=0.34%. From (12), in comparison between OBE and SBE, the energy reduction gain can be obtained by, as shown in Table 1 (b).

With the assumption ofB= 20 MHz , SNR=15 dB,BA= 8 MHz , andBB= 2 MHz , the OBE outperforms the SBE with ERF=1.64%. From (12), in comparison between OBE and SBE, the energy reduction gain can be obtained by, as shown in Table 1 (b). Thus, when the required bandwidth between the two users is higher, the improved ERF of OBE becomes higher.

Fig. 2. ERF comparison for different BECs for user A,αAwith SNR=15dB.

3. Simulation Results

From the investigation in the previous section, improving energy efficiency of the bandwidth re-allocation is related with the required bandwidth and SNR of users. Therefore, in this section, the effectiveness of OBE on the system environments is further studied.

At first, the simulation environment is shown in Table 2, whereM=BA/BB,M=1/4, 1/3, 1/2, 1, 2, 3, 4, andBA=1.2, 1.5 MHz, 2 MHz, 3 MHz, 4 MHz, 4.5 MHz, and 4.8 MHz. Then, the BECαA,Oof user A can be found by the proposed OBE of (9) for different total bandwidths. Because we fix the total required bandwidth of two usersBU=BA+BB, the BECs of SBE,αS, can be obtained in proportional to the total bandwidthB. With the reallocation algorithms, the BECs of user A,αA,O, vs. the BECs of SBE,αS, can be obtained as shown in Fig. 3. From Fig. 3, it is observed that withM=1,αA,Oare the same asαS, that is, no need to relocate the bandwidth when the required bandwidths of users are the same. Moreover, whenM＜1 and smaller, theαA,Obecomes higher. But on the contrary, whenM＞1 and larger, thebecomes lower.

Table 2: Simulation parameters ofM,BA, andBB

Fig. 3. Relationships betweenαA,OandSαwith differentMand SNR=15 dB.

The performance of the improvement of the proposed OBE over SBE is depicted in Fig. 4 forM=1/4, 1/3, 1/2, 1, 2, 3, 4, and SNR=15 dB. In Fig. 4, due toM=BA/BB, both results ofM=4 and 1/4 are the same. Similarly, the results of two pairs,M=3,1/3, andM=2,1/2 are also the same. From Fig. 4, it is obvious that the improvements of the proposed OBE over SBE,, are higher than 60% forαS＞ 1.5 exceptM=1. Moreover, whenαSis higher than 3, the improvementbecomes less and approaches tobe saturated. Moreover, the performance offor differentM, SNR=15 dB,αS=2, andB=12 MHz is shown in Table 3.

Fig. 4. Relationships betweenandwith differentMand SNR=15 dB.

Table 3: Performance offor differentM, SNR=15 dB,αS=2 (B=12 MHz)

Table 3: Performance offor differentM, SNR=15 dB,αS=2 (B=12 MHz)

MBBEMA(MHz)BBEMB(MHz)AαBα,,(%)EUr O S1/4 3.504 8.496 2.92 1.77 73.5486 1/3 3.945 8.055 2.63 1.79 72.5484 1/2 4.680 7.320 2.34 1.83 71.3783 1 6.000 6.000 2.00 2.00 0 2 7.320 4.680 1.83 2.34 71.3783 3 8.055 3.945 1.79 2.63 72.5484 4 8.496 3.504 1.77 2.92 73.5486

Previous results are performed with the same SNR for all users. However, when the SNRs of users are different, we define the SNR ratio of SNRAto SNRBby

where SNRAand SNRBare the SNRs of user A and user B, respectively. The simulation parameters ofn, SNRAand SNRBare shown in Table 4.

The simulation results of relationships ofαA,OtoSαwith differentnandM=1 (BA=BB=3 MHz) are depicted in Fig. 5. From Fig. 5, it is observed that whenn＞1, theαA,Ois becoming higher than that ofn=1. That is to say when SNRAis higher than SNRB, user A can have more energy efficiency gain than user B. On the contrary, whenn＜1, theαA,Ois becoming smaller than that ofn=1. The improvement of ERF of OBE to SBE,, is depicted in Fig. 6 fornwithM=1. In Fig. 6, due toM=1 and according to (13), both results ofn=4 and 1/4 are the same. Similarly, the results of two pairs,n=3, 1/3 andn=2, 1/2 are also the same. From Fig. 6, it is observed that withn=1/8 and 8, the improvement of ERF of OBE to SBE,, is the highest of 24% withSα=1.5. However, when then=1/4 or 4 and 1/2 or 2, the improvement of ERF,, becomes smaller.

Table 4: Simulation parameters ofn, SNRAand SNRB

Fig. 5. Relationships betweenandwith differentnandM=1.

Fig. 6. Improvement offornandwithM=1.

After we investigate the performance of energy efficiency improvement of the proposed OBE algorithms on the scenarios of the same SNR and the same required bandwidth for two users, we further combine the previous results as shown in Fig. 7 and Fig. 8. The optimal BECs for user A are depicted in Fig. 7 for differentMandnwithαS= 2. From Fig. 7, whenM=1,n＜1 (SNRA＜ SNRB), theoptimal BECs obtainαA,O＜ 2. However, whenMincreases, theαA,Obegins to increase. Moreover, whenn＞1,M=1, SNRA＞SNRB, andαA,Oincreases more. WhenMincreases,αA,Obegins to decrease andαA,O> 2.

Fig. 7. 3D relationship ofαA,O, withMandnforαS= 2.

Fig. 8. 3D relationship ofwithMandnforαS= 2.

The performance of energy efficiency improvement of the proposed OBE,, is shown in Fig. 8 with. From Fig. 8, it is seen that whileM>2 andn＜1,. Especially whenn=1/8 andM=4,approaches the highest 45%. Moreover,forM＜2. It is observed that with two regions: a)M≤1 andn≤ 1, and b) 2＜n≤8 andM＞2,is less than 10%.

4. Conclusions

In this study, we propose an optimal bandwidth allocation algorithm to improve the energy efficiency for wireless communication systems. Simulation results show that the OBE outperforms the SBE with different SNRs and different required bandwidths for users. In two users scenario, the ERF ratio of OBE to SBE,, can reach 45%. The ERF of OBE achieves 86.57%, which is 11.61% higher than the ERF of SBE,.

[1] T. Edler, “Green base stations—How to minimize CO2emission in operator networks,” presented at Ericsson seminar, Bath Base Station Conf., 2008.

[2] V. P. Ipatov,Spread Spectrum and CDMA: Principles and Applications.Hoboken: Wiley, 2005.

[3] J. S. Lee and L. E. Miller,CDMA System Engineering Handbook, Artech House, 1998.

[4] F. Adachi, M. Sawahashi, and H. Suda, “Wideband DS-CDMA for next-generation mobile communications systems,”IEEE Commun. Mag., vol. 36, no. 9, pp. 56-69, 1998.

[5]Standard for Local and Metropolitan Area Networks, Part16: Air Interface for Fixed Broadband Wireless Access Systems, IEEE Std. 802.16, 2004.

[6] X. Dai, “Carrier frequency offset estimation and correction for OFDMA uplink,”IET Communications, vol. 1, no. 2, pp. 273-281, 2007.

[7] A. M. Tonello, N. Laurenti, and S. Pupolin, “Analysis of the uplink of an asynchronous multi-user DMT OFDMA system impaired by time offsets, frequency offsets, and multi-path fading,” inProc. of IEEE VTC, 2000, pp. 1094–1099.

[8]Digital Video Broadcasting (DVB): Transmission System for Hand-Held Terminals (DVB-H), European Telecommunication Standards Institute (ETSI), ETSI EN 302 304 V1.4.1, Nov. 2004.

[9] J. G. Proakis,Digital Communications, 4th ed., McGraw Hill, 2001.

[10] S. Videv and H. Haas, “Energy-efficient scheduling and bandwidth-energy efficiency trade-off with low load,” inProc. of IEEE ICC, 2011, pp. 1-5.

[11] F. Meshka, V. Poor, and S. Schwartz, “Energy-efficient resource allocation in wireless networks,”IEEE SignalProcessing Magazine, vol. 24, no. 3, pp. 58–68, 2007.

[12] C. Han, T Harrold, and S. Armour, “Green radio: Radio techniques to enable energy efficient wireless networks,”IEEE Commun. Mag., vol. 49, no. 6, pp. 46–54, 2011.

Yung-Fa Huang received the Diplom-Eng. in electrical engineering from National Taipei University of Technology, Taipei in 1982, the M.Eng. degree in electrical engineering from National Tsing Hua University, Hsinchu in 1987, and the Ph.D. degree in electrical engineering from National Chung Cheng University, Chiayi in 2002. During 1982 to 1984, he joined the Air Forces for the military service. During 1987 to 2002, he was an instructor with Chung Chou Institute of Technology, Yuanlin. From February 2002 to July 2004, he was an associate professor with the Department of Electrical Engineering, ChungChou Institute of Technology. From August 2004 to July 2007, he was an associate professor with the Graduate Institute of Networking and Communication Engineering, Chaoyang University of Technology, Taichung. From August 2007 to July 2008, he was the Head of Department of Computer and Communication Engineering and the Chair of the Graduate Institute of Networking and Communication Engineering, Chaoyang University of Technology. From August 2008 to July 2010, he was the Head of Department of Information and Communication Engineering, Chaoyang University of Technology. Since September 2012, he has been a professor with the Department of Information and Communication Engineering, Chaoyang University of Technology. His current research interests include multiuser detection in OFDM-CDMA cellular mobile communication systems, communication signal processing, fuzzy systems, and wireless sensor networks. Dr. Huang is a member of IEEE and serves as a Co-Chair of IEEE SMC Society Technical Committee on Intelligent Internet Systems.

Che-Hao Li received his bachelor degree in the information and communication engineering from Chaoyang University of Technology, Taichung in 2011. He is currently pursuing his M.S. degree with the Department of Information and Communication Engineering, Chaoyang University of Technology.

Chuan-Bi Lin received his Ph.D. degree in the electrical and computer engineering from New Jersey Institute of Technology, Newark in 2008. He was a postdoctoral researcher with the Department of Electrical and Computer Engineering, New Jersey Institute of Technology in 2008 to 2009. Currently, he is an assistant professor with the Department of Information and Communication Engineering, Chaoyang University of Technology, Taichung. His current research interests include design and analysis of Internet of thing (IOT), software-defined networks, and apps of smartphone.

Chia-Chi Chang received the B.S. degree in mechanical engineering from Chung Yuan Christian University, Taoyuan in 1999, the M.S. degree in mechanical engineering from National Yunlin University of Science & Technology, Yunlin in 2001, and the Ph.D. degree from National

CentralUniversity,

Taoyuan. Currently, he is an associate professor with the Department of Information and Communication Engineering, Chaoyang University of Technology. His current research interests include wireless sensor networks and ultra-low embedded system design.

Manuscript received December 20, 2013; revised March 14, 2014. This work was supported by the NSC under Grant No. 101-2221-E-324-024.

Y.-F. Huang is with the Department of Information and Communication Engineering, Chaoyang University of Technology, Taichung (Corresponding author e-mail: yfahuang@cyut.edu.tw).

C.-H. Li, C.-B. Lin, and C.-C. Chang are with the Department of Information and Communication Engineering, Chaoyang University of Technology, Taichung (e-mail: jackbest8@yahoo.com.tw; cblin@cyut. edu.tw; ccchang@cyut.edu.tw).

Digital Object Identifier: 10.3969/j.issn.1674-862X.2015.01.006