UEs Power Reduction Evolution with Adaptive Mechanism over LTE Wireless Networks

1.Introduction

It has been reported now that cell phones are ubiquitous with a worldwide user subscription of over 4.55 billion, a significant number of mobile devices are being used to access the Internet which leads to the larger use of data and as a result increased power consumption[1],[2]. Most of the mobile devices presently support the majority of services like the 3rd generation telecommunication (3G), global system for mobile communications (GSM), Wi-Fi, and long term evolution (LTE)for data transfer. After 2009, LTE has effectively made its entry in the commercial markets and is available in more than ten countries, with a fast-growing user base. The targeted user throughput is 100 Mbps for downlink and 50 Mbps for uplink,which is larger than the existing 3G networks, with the userplane latency less than 5 ms[3].

Devices such as cell phones, email pagers, and moving picture experts group audio layer-3 (MP3) players have some specific functionality and hence are dedicated to the use of power leading to longer battery life. The design of such devices also contributes towards efficient power consumption which has resulted in the popularity of their usage. Therefore, a device for general purpose incorporating many other functionalities should have such design with minimal power consumption.Fig. 1 illustrates the setup of a mobile environment in which we assume connections from the Internet created and terminated at mobile end-points. The backbone of the mobile network is represented as a core network and also assumed as a full outfit network which connects the access points to the rest of the mobile network. The core network behaves like a wired multimedia environment with minimal bit error losses and congestion losses.

唐纳森作为全球过滤行业领导者之一，早在70年代就进入中国展开业务，并于1997年在无锡建厂。无锡现有工厂占地面积60 000平方米，拥有员工1200人，主要生产发动机过滤和工业过滤产品。作为一家扎根中国多年的企业，唐纳森始终坚持以本土制造服务于本土客户，随着唐纳森在中国乃至亚洲地区业务的不断增长，目前工厂的产能已接近饱和。在无锡政府的大力支持下，唐纳森决定在无锡投资建立新的工厂。

Fig. 1. Mobile environment with LTE networks.

Batteries provide the necessary energy required for the operations of mobile devices. The capacity of a battery employed in a device is limited which indicates an optimal power management system to be implemented in such devices.Internet streaming applications on mobile devices are popular due to faster broadband speeds and higher bandwidth provided in advanced LTE and 5G networks but constrained in longer usage time due to the limited battery capacity[4]. Battery power consumption can be slightly reduced by lowering the transmission rate of streamed data (reducing the streaming quality), as the power consumption depends on the data transmission rates[5],[6]. Thus, we propose an energy-efficient rate adaptation (RA) mechanism that can reduce the mobile battery power consumption during multimedia streaming.

The rest part of the paper is organized as follows. In Section 2, we review related literature. The literature study for the proposal is presented in Section 3, followed by the description of a system model which introduces the LTE user equipment (UE) power consumption method as a way to save battery power on mobile devices. Section 4 discusses the simulation results where UE’s energy reduction problem was solved by applying RA and without rate adaptation (WRA) based mechanism over LTE networks. Finally, Section 5 concludes the paper.

2.Related Work

The system model assumes that we have considered an LTE UE power consumption model. Traditionally, LTE has two radio resource control (RRC) states namely RRC_Connected and RRC_Idle according to the UE active or inactive status[18],[19]. As shown in Fig. 2, there are given three modes, respectively: Continuous reception (CR), short discontinuous reception (DRX), and long discontinuous reception (DRX). UE can be in one of these three modes at the RRC_Connected state, but UE is only in the DRX mode at the RRC_Idle state. In that situation, if UE is in the RRC_Idle state and also receives and sends the packet without considering the packet size, then the state will switch from RRC_Idle to RRC_Connected with a pretty substantial delay. Thus, UE enters CR after being endorsed to RRC_Connected by default and keeps observing the physical downlink control channel(PDCCH) to give a control message to UE[20].

In [5], a study presented for the transmission of data to reduce battery power consumption, which was much more complicated while choosing a high-quality version for end users and required server support in advance. Therefore, most of the service network interface cards are compatible by default with the IEEE 802.11 power saving mode (PSM). Recently, the battery power consumption during streaming services is directly minimized, when the power saving mode is on in the devices. For instance, authors in [10] provided a method that maximized the sleeping time of the wireless network interface card (WNIC) in the LTE network. Therefore, the power saving mode suggested a client-centric arrangement to allow buffering on the server side so that streaming data could be able to send in bursts.

Moreover, to measure the power consumption for LTE UE,a study showed a standard model and parameters[11]. In this model, the power consumption is entirely dependent on the network component of UE based on the carrier aggregation.And in [12], it was shown that decoding multimedia contents consumes a large part of the energy and that the power required to decode audio or video depends on the computational complexity of the codec. Though, the authors had not defined the exact amount of energy consumed by the decoding part in mobile devices. Efficiently managing power on mobile platforms has always been a complex and challenging research problem so far, because of the several running applications and internal configurations, such as possible hardware configuration options,power states, and the interdependencies between computing,sensing, and networking resources, and their quality of experience (QoE)[13]-[15].

研究区硅质岩的化学成分以SiO2为主，含量为70.02×10-2～93.72×10-2，平均为82.12×10-2，其次为TFe2O3、Al2O3和MgO，含量分别为2.53×10-2～6.89×10-2、1.50×10-2～7.81×10-2和0.30×10-2～5.14×10-2，其余成分含量均很低(表2)。

Based on Fig. 2, note that Ttaii coexists with Ti and Tis, so the UE overall power consumption model in LTE is defined as(1) with the unit of Watt:

3.Power Consumption Model

Power saving schemes during video streaming have engaged ample research recently[7]. Some related studies proposed several streaming techniques used by commercial video providers, for instance, ON-OFF, throttling, bitrate streaming, and compared them in the context of power,bandwidth, and delay[8]. The authors in [9] measured the video streaming’s energy while using Wi-Fi and 3G. Nevertheless,they did not offer any useful solution to improve the energy efficiency regarding the long life battery consumption.

油田从毕业生引进工作抓起，通过设置量化测评、专业知识考试、面试等环节，全面考察毕业生的专业水平、能力素质、工作责任心等方面，提升毕业生引进质量。自1995年以来，油田连续27年在见习期满大学生中开展“优秀见习毕业生”评选和“优秀见习论文”竞赛，鼓励见习期间年轻人才主动学习，脱颖而出。对表现优秀的青年人才纳入后备人才库，有针对性地进行重点扶持、跟踪培养。

Fig. 2. State transitions in LTE network.

The DRX inactivity timer Tin also starts by an individual UE, which will be reset whenever UE receives and sends a packet to/from the server. Based on Tin’s expiration UE enters the short DRX mode without seeing any data activity at the starting point in each DRX cycle. An ON-period will follow, in which the UE turns its receiver on, to receive packets from the eNB (i.e., base station) and monitor whether there is any pending downlink transmission or not while UE is in short DRX.

Thus when the short cycle timer starts, UE switches to long DRX if there is no data activity. Else, UE goes back into CR. If there is any data transfer, then every time UE enters CR. And then UE starts the tail timer, which is reset whenever a packet is sent or received by UEs. When expires, UE breaks from RRC_Connected to RRC_Idle and the allocated radio resource is released.

The researchers in [16] introduced the mechanism with the supporting of carrier aggregation (CA) with a complexity in UEs design regarding the receiver and transmitter. Some work has been presented on how this can be affected on the UE battery life after CA was standardized in the 3rd generation partnership project (3GPP). In [17], some contribution has been discussed with the effect on UEs complexity and battery life but the current power consumption was not considered. Our paper has considered saving the power consumption based on the running application of Internet streaming.

where V is the binary variable expressing whether the UE is in RRC_Connected; VC, Vid, and VD are binary variables indicating that UE is in connected mode, idle mode, and DRX mode, respectively. Similarly, PC, Pid, and PD are the power consumption in the connected mode, idle mode, and DRX mode, respectively.

In this section, the performance evaluation results of the proposed algorithm are presented. At first, we have analyzed a power level comparison between RA and WRA based schemes.We also showed the energy level with increasing and decreasing the throughput with and without recharging. Finally,this paper conducted average power consumption over picture resolution in two different scenarios such as online streaming and offline playback. Table 1 lists the simulation parameters which we have considered in this paper.

Fig. 3. LTE cellular system power model.

The function of the block can be evaluated empirically by varying the function-specific parameter like RTX of the TX BB,at the same time observing the other parameters, such as LTX,RRX, and LRX and a level where they influence the measurement at least. Based on Fig. 3, the RRC_Connected mode power consumption is defined as:

Based on our experiment and results, total power consumption during streaming over LTE in scenarios using RA and WRA is shown in Fig. 4 and Fig. 5. The LTE power consumption is calculated based on (1) and (4) as described in Section 3.

4.Results and Discussion

The future UE model depends on the transmits (TX) and receive (RX) power levels, downlink (DL) and uplink (UL)data rates, and RRC mode[21]. The RRC_Connected mode is divided into baseband (BB) and functional blocks. BB block includes the TX baseband (TXBB) and RX baseband (RXBB).The radio frequency (RF) functional blocks which define the power consumption as a function of either TX or RX power level (L) and data rate (R). Fig. 3 shows the model which is divided into blocks, where each block has a different parameter. For instance, the transmit power LTX in the TXRF initially affects the power consumption of that block.

Table 1: Simulation parameters

?

where PC is the power consumption in the connected mode;Pon is the power consumed when UE is ON, PRX is the power consumed when the receiver is actively receiving;PTX is the power consumed when the transmitter is actively transmitting. And in (2) and (3), PRXBB is the power consumed of RXBB; PRXRF is that of RXRF; PTXBB is that of TXBB; PTXRF is the power consumed in TXRF. During the simulation, the UE energy level is decreased by using RA mechanism based on calculations.

如果你随时做到尊重事实，实话实说，你还需要精心设计、周密安排自己的语言吗？是什么就是什么，有什么就说什么，心底无假心自安，哪怕是说过的话很快被自己忘了也不要紧，因为客观事实还在，它不但不需要你去花工夫“论证”它的存在，甚至还可以帮忙证明你的本真面目。

Fig. 4. Energy level with power-ON.

Fig. 5. Energy level due to power in sleep mode.

We have proposed the RA mechanism to save the power for the mobile phone as a mean to battery power reduction in two conditions: Power-ON and without power-ON means in the sleep mode in a mobile phone individually. We proposed the mechanism based on a power level comparison. Fig. 4 shows the power level comparison between RA and WRA in the power-ON scenario where WRA mechanism consumes 99.5% of the overall power and the proposed RA mechanism consumes 94.5% of the overall power within 5 hours. As the RA mechanism saves almost 5% of power consumption compared with the WRA method. So we can say that the applied WRA mechanism is less efficient than the RA method.And the without power-ON scenario means the sleep-mode condition. As shown in Fig. 5, the WRA mechanism consumes 98% of the overall power and the RA mechanism consumes 88% within 5 hours. So the RA mechanism almost saves 10%power compared with the WRA method. Hence, the RA mechanism has better power consumption in comparison with the WRA mechanism, and this saves high power consumption in the without power-ON means sleep mode in a mobile phone than that with power-ON.

深圳“图书馆之城”也是一个典范。深圳图书馆课题组编纂出版了《图书馆之城建设指标体系研究》与《深圳模式——深圳“图书馆之城”探索与创新》，两书为深圳“图书馆之城”的服务品牌建设作了一个系统性的总结，从探索、创新、成就与展望等维度出发，体现出深圳图书馆人创建服务品牌的文化自觉和开创中国特色中国城市图书馆发展道路的文化自信。

节职能，加快推进基本公共服务均等化，缩小收入分配差距。”[10]335 实现社会公平正义，调节初次分配和加大再分配力度，注重两者之间的协调性，鼓励支持慈善事业的发展，发挥第三次分配的作用。 然而，当前中国仍处于社会主义初级阶段，社会中还存在许多不公正、不合理的现象，依旧制约着社会的发展，我们应当以这些现象为治理目标，在制度和实践层面推进社会公平正义建设。

According to different UE’s energy level, as we can see in Fig. 6, every single UE throughput increases and decreases because their power level is changing. And also,it adjusts the quality of the streaming multimedia contents depending upon the UE’s energy level. The quality of the multimedia content is controlled by adjusting the sending packet size as the constant bit rate (CBR) at the server. The energy level in UE is monitored by every 1 minute and the packet size is changed according to the UE’s extra energy level. Thus, it achieves overall slightly high energy saving with our proposed mechanism.

5.Conclusions

Fig. 6. Throughput vs. time.

In this paper, we studied the energy-aware multimedia streaming on mobile phones. A bit-RA based approach was presented to optimize the power consumption in the context of the green technology. By applying the energy consumption model to a comprehensive dataset, we show that LTE is much less power efficient than the other previous technology.However, we did not show the mechanism comparison with previous technology, but we have shown the battery power consumption on mobile devices based on streaming multimedia contents over LTE networks. Moreover, we have proposed an RA mechanism based on the remnant mobile power level as a mean to save battery power consumption. By comparing the RA and WRA based mechanisms, our results had successfully achieved low power consumption in both power-ON and without power-ON means in the sleep mode during streaming in LTE networks.

References

[1]B. Rashidi, C. Fung, and T. Vu, “Recdroid: A resource access permission control portal and recommendation service for smartphone users,” in Proc. of the ACM MobiCom Workshop on Security and Privacy in Mobile Environments,2014, pp. 13-18.

[2]A. Sun, T. Wambach, A. G. Venkatesh, and D. A. Hall, “A low-cost smartphone-based electrochemical biosensor for point-of-care diagnostics,” in Proc. of 2014 IEEE Biomedical Circuits and Systems Conf., 2014, pp. 312-315.

[3]A. R. Jensen, M. Lauridsen, P. Mogensen, T. B. Sørensen,and P. Jensen, “LTE UE power consumption model: For system level energy and performance optimization,” in Proc.of 2012 IEEE Vehicular Technology Conf., 2012, pp. 1-5.

[4]A. Carroll and G. Heiser, “An analysis of power consumption in a smartphone,” in Proc. of USENIX Annual Technical Conf., 2010, p. 21.

[5]Y. Liu, L. Guo, F. Li, and S. Chen, “An empirical evaluation of battery power consumption for streaming data transmission to mobile devices,” in Proc. of the 19th ACM Intl. Conf. on Multimedia, 2011, pp. 473-482.

[6]R. Trestian, Q. T. Vien, P. Shah, and G. Mapp, “Exploring energy consumption issues for multimedia streaming in LTE HetNet small cells,” in Proc. of IEEE the 40th Conf. on Local Computer Networks, 2015, pp. 498-501.

[7]S. Aghaeinezhadfirouzja, H. Liu, B. Xia, and M. Tao,“Implementation and measurement of single user MIMO testbed for TD-LTE-A downlink channels,” in Proc. of IEEE the 8th Intl. Conf. on Communication Software and Networks, Beijing, 2016, pp. 211-215.

[8]M. A. Hoque, M. Siekkinen, J. K. Nurminen, and M. Aalto,“Dissecting mobile video services: An energy consumption perspective,” in Proc. of IEEE the 14th Intl. Symposium on A World of Wireless, Mobile and Multimedia Networks, 2013,pp. 1-11.

[9]P. M. Eittenberger, M. Hamatschek, M. Großmann, and U.R. Krieger, “Monitoring mobile video delivery to android devices,” in Proc. of the 4th ACM Multimedia Systems Conf.,2013, pp. 119-124

[10]E. Tan, L. Guo, S. Chen, and X. Zhang, “PSM-throttling:Minimizing energy consumption for bulk data communications in WLANs,” in Proc. of IEEE Intl. Conf. on Network Protocols, 2007, pp. 123-132.

[11]R. Sanchez-Mejias, Y. Guo, M. Lauridsen, P. Mogensen, and L. A. Maestro Ruiz de Temino, “Current consumption measurements with a carrier aggregation smartphone,” in Proc. of IEEE the 80th Vehicular Technology Conf., 2014,pp. 1-5.

[12]M. A. Hoque, M. Siekkinen, and J. K. Nurminen, “Energy efficient multimedia streaming to mobile devicesa survey,”IEEE Communications Surveys and Tutorials, vol. 16, no. 1,pp. 579-597, 2014.

[13]Q. Jiang, V. C. M. Leung, H. Tang, and H. S. Xi, “Energyefficient adaptive rate control for streaming media transmission over cognitive radio,” IEEE Trans. on Communications, vol. 63, no. 12, pp. 4682-4693, Dec. 2015.

[14]N. Khan and M. G. Martini, “Qoe-driven multi-user scheduling and rate adaptation with reduced cross-layer signaling for scalable video streaming over lte wireless systems,” EURASIP Journal on Wireless Communications and Networking, no. 1, pp. 93:1-23, 2016, DOI:10.1186/s13638-016-0584-6.

[15]M. Seufert, S. Egger, M. Slanina, T. Zinner, T. Hoßfeld, and P. Tran-Gia, “A survey on quality of experience of HTTP adaptive streaming,” IEEE Communications Surveys and Tutorials, vol. 17, no. 1, pp. 469-492, 2015.

[16]B. Dusza, P. Marwedel, O. Spinczyk, and C. Wietfeld, “A context aware battery lifetime model for carrier aggregation enabled LTE-A systems,” in Proc. of IEEE Consumer Communications and Networking Conf., 2014, pp. 13-19.

[17]NTT DoCoMo, Views on Downlink Reception Bandwidth Considering Power Saving Effect in LTE-Advanced, 3GPP R1-090310, 2009.

[18]W. D. Doyle, “Magnetization reversal in films with biaxial anisotropy,” IEEE Trans. on Magnetics, vol. 2, no. 2, pp. 68-73, 1966.

[19]Z. Shen, A. Papasakellariou, J. Montojo, D. Gerstenberger,and F. Xu, “Overview of 3GPP LTE-advanced carrier aggregation for 4G wireless communications,” IEEE Communications Magazine, vol. 50, no. 2, pp. 122-130,2012.

[20]D. Musiige, L. Vincent, F. Anton, and D. Mioc, “LTE RF subsystem power consumption modeling,” in Proc. of the 1st IEEE Global Conf. on Consumer Electronics, Tokyo, 2012,pp. 645-649.

[21]J. Huang, F. Qian, A. Gerber, Z. M. Mao, S. Sen, and O.Spatscheck, “A close examination of performance and power characteristics of 4G LTE networks,” in Proc. of the 10th Intl. Conf. on ACM Mobile Systems, Applications, and Services, 2012, pp. 225-238.

[22]R. Kwan and C. Leung, “A survey of scheduling and interference mitigation in LTE,” Journal of Electrical and Computer Engineering, no. 1, pp. 1-10, 2010, DOI:10.1155/2010/273486