开发工具:
文件大小: 259kb
下载次数: 0
上传时间: 2019-01-12
详细说明:In recent media, much attention has been paid to the battery life of VoLTE enabled LTE
phones in comparison to existing 2G and 3G circuit switched telephony. This white paper
aims to address the confusion and uncertainty around the topic and demonstrate the true
potential of VoLTE.SSON LTE, Telephony and Battery Life
White Paper
Contents
Contents
About this document
1.1
Purpose
1.2
Audience
Revision information
44444
1.4
Reference list
Introduction
Key Power Consumption Factors
3.1
Software architecture
3.2
Radio hardware
56677
3.3
Transmission protocols
Summary
9
VoLtE Reva 2012-12-18
Prepared:
Bjorn ekelund
LTE, Telephony and battery life
White Paper
SSON
About this document
Purpose
The purpose of this document is to reduce the confusion in the marketplace around
expectations around battery life for VoLTE enabled LTE smartphones
Audience
The intended audience for the document is engineers, marketing and planning personnel
at device manufacturers and telecom operators
Revisⅰ on information
Table 1
Revision history
Date
Rev Comments
2012-12-17
PA1 Draft
2012-12-19
A
First release
1.4
Reference list
[13GPP specifications
www.3Gpp.oRg
[2gsMarecommendationshttp://www.gsma.com/newsroom/technical-
documents
VoltE Reva 2012-12-18
4(9
O Copyright ST-Ericsson 2012. All rights reserved
LTE, Telephony and battery life
White Paper
SSON
2
Introduction
Lately there has been much attention on battery life for devices supporting the new
telephony system for LtE known as VoLtE. They main source of the underlying data is a
measurement report by Metrico made in MetroPCS CDMA+LTE network in late
November. It has since then been quoted and analyzed by many media and analysts
worldwide
For obvious reasons, this white paper does not disclose detailed data about ST
Ericsson's products and will instead try to shed some light on the underlying mechanics
and also make some projections of what can be expected in well-designed, commercial
VoLtE devices compared to what is available in the market in the fall of 2012
First of all, without VoLtE CDMA/LTE systems suffer a severe penalty in that they use a
concept called simultaneous Voice and LtE-svLtE. This means that during a voice
call, both the Cdma radio and the lte radio are in full operation The consequence of
this is clearly visible in the Metrico data, which also shows the obvious fact that
VoLTE+LTE consume less power than CDMA+LTE
LTE networks with a 3GPP-based legacy do not suffer this. Neither do cdma+LTE
solutions based on fall-back schemes(such as the one used in iPhone 5)
When LtE and VoLtE where standardized years ago, great care was put into making the
Much of the media attention has been on first generation VoltE implementations based
on first generation LTE chipsets. Chipsets developed with a focus on time to market and
on plain data services rather than more complex ones
Given its long experience in wireless technology, ST-Ericsson has put its aim much higher
and is currently launching LtE solutions which will demonstrate a completely different
performance compared to what is currently on the market
This white paper will offer some simple back-of-an-envelope calculations indicating the
kind of performance you will soon see in LTE phones in a shop near you
Apart from the radio usage scheme used(single/double), there are really three major
factors influencing the battery life of a VoLtE enabled smartphone; software
architecture, radio hardware and transmission protocols
VoltE Reva 2012-12-18
O Copyright ST-Ericsson 2012. All rights reserved
LTE, Telephony and battery life
White Paper
SSON
Key power Consumption Factors
3.1
Software architecture
The software architecture of the handset ultimately defines how the computing load of the
IMS protocol stack and the audio processing is distributed across the various processors
in the handset. a smartphone typically has a number of processor clusters, the two major
ones being the Application processor(typically based on one or more ARM Cortex A-
class cores)and modem processor (typically based on or more proprietary solutions or
smaller ARM cores). Obviously, these two main processor clusters are optimized for
completely different purposes
The Application processor is optimized to run a high-level operating system such as
Android, whereas the Modem processor typically runs a real-time Os and is optimized for
decision making and number crunching
DB8540
ARM CortexA9
Video
Display
20Mtix shA
+a modem
Cun eHD e VUGA
cam 0
Graphics
IMG SGX544
Cw1250CIFD
This means that for a task like protocol/audio processing the modem processor is vastly
more efficient. For example, comparing the MIPS/milliwatt numbers( the metric telling you
how much computing power you get for your electrical power) for a typical dual core
application processor and an embedded modem processor you may see ratios between
2x and 4x. Of course these ratios are only valid for very specialized processing tasks. If
we tried to run a high level oS with large memory pools and heavy context switching on
the modem processor it would perform very poorly.
The huge power saving(2-4x)from using the most appropriate processor in the handset
is to some extent offset by the use of the multimedia accelerators in all application
processors. In reality this means that the ratio is closer to 1.5-2X for highly optimized
code
VoltE Reva 2012-12-18
O Copyright ST-Ericsson 2012. All rights reserved
LTE, Telephony and battery life
White Paper
SSON
3.2
Radio hardware
Another key factor for VolTE power consumption is the radio hardware. First of all, the
linearity requirements for an LTE radio are far higher than for a CDMA or WCDMA radio
This means that without changes to the radio circuitry, the operating point has to be
lowered, which in turn means lower energy efficiency and, consequently, a higher power
consumption to reach the same effective antenna radiation
The difference also includes passives. a 3G radio rarely have more than 3 or 4 bands. An
LTE radio can have up to 8 or even 10 bands. Losses related to the large number ot
switches and filters mean that, again without major architectural changes, the radiated
antenna power will be lower for the same battery power spent
Power drain for same antenna power
Conventional re
OFDM Optim ized RF
These are factors playing a key role in most first generation LtE devices. For more
advanced solutions, such as the thor tm M7400 modem from ST-Ericsson these
challenges are addressed with new circuits and new hardware architectures, giving no
effective penalty on battery power for LTE
Transmission protocols
Together with protocols for call establishment, codec negotiation etc, the VoLtE standard
comes with some new radio protocol enhancements aimed to reduce both spectrum
usage and battery power. Few if any of them are activated in networks today
The first one is Semi-persistent Scheduling, SPS. SPS aims to drastically reduce control
signaling overhead for services that require persistent radio scheduling like volP. This of
course has a direct effect on battery life. Simulations show between 10 and 20%
depending on coverage conditions and radio architecture
VoltE Reva 2012-12-18
O Copyright ST-Ericsson 2012. All rights reserved
LTE, Telephony and battery life
White Paper
SSON
20ms
30ms
40ms
BFN mod4=0
BFN mod4=1
BFN mod4=2
BEN mod4=3
HARQ
Dynamically Scheduled PRB
1a RE T
2nd ReTx
Example of Semi-Persistent Scheduling
Via: 3g4g. blogspot. com
TTI bundling is another one; however, this has less effect on battery life. Its main effect is
instead greatly improved service quality at the cell edge. Predictions of up to 4db have
been made which is dramatic the effect on battery life in real use is however difficult to
predict and so small that it can be ignored for the sake of this exercise
The third, and perhaps the most important one, is Discontinuous Reception, DRX DRX iS
a method to switch the smartphone transmitter on and off based on available transmission
data. The scheme is useful not only for voice but also for any intermittent traffic, such as
web browsing or online gaming
DRX benefits both the uplink(smartphone transmit) and downlink(smartphone receive)
but the gains in uplink drx greatly exceeds those of downlink dRX. So, for the purpose
of this paper, we will disregard downlink DRX
It is a well-established fact that a in a typical phone conversation the two parties speak for
around 80% of the time this means that the smartphone transmitter only needs to be
active about 40% of the time. Theoretically the power reduction from using this would be
60%(silence plus when the other party speaks) but due to factors like switching policies
and protocol overhead the real saving is closer to 30%. Nevertheless, a major saving
X10
5
VoltE Reva 2012-12-18
8(9
O Copyright ST-Ericsson 2012. All rights reserved
LTE, Telephony and battery life
White Paper
SSON
Summary
For"normal"output powers, the radio transmitter circuitry with its power amplifier
dominates a smartphone's power consumption during a voice call. Now, with conservative
savings of 20% due to OFDM/multiband optimized circuitry and passives, 15% from SPS
and 30% from DRX, we see a total reduction for the radio circuitry of around 50%
Assuming a conservative ratio of 1.5 between the power consumption of anapp-like
VoLtE client and a deeply embedded one, provides a baseband power saving of roughly
30%. Using realistic numbers from real products these savings give you a chart like this
Total Power Consumption
First generation
Baseband
Radio
Optimized
Baseband
Since we see first generation VoL TE enabled phones(not employing SVLTE) in the
market today with battery life just slightly worse than 3G phones, this very conservatively
drawn chart indicates a great potential for the more mature products that we will start
seeing in 2013
Comparing with the actual data processing power needs and the actual radio output
powers of today's 3G phones, there is definitely potential for VoLTE phones having
significantly longer use time
On top of this Moore's law just continues to help and help
Another final, albeit interesting, observation is that over-the-top services like skype are
forced to run in the power-hungry application processor(since they are apps)and have
no help from advanced protocols; consequently they will be increasingly disadvantaged
versus VOLTE. Something they are not today
VoltE Reva 2012-12-18
O Copyright ST-Ericsson 2012. All rights reserved
(系统自动生成,下载前可以参看下载内容)
下载文件列表
相关说明
- 本站资源为会员上传分享交流与学习,如有侵犯您的权益,请联系我们删除.
- 本站是交换下载平台,提供交流渠道,下载内容来自于网络,除下载问题外,其它问题请自行百度。
- 本站已设置防盗链,请勿用迅雷、QQ旋风等多线程下载软件下载资源,下载后用WinRAR最新版进行解压.
- 如果您发现内容无法下载,请稍后再次尝试;或者到消费记录里找到下载记录反馈给我们.
- 下载后发现下载的内容跟说明不相乎,请到消费记录里找到下载记录反馈给我们,经确认后退回积分.
- 如下载前有疑问,可以通过点击"提供者"的名字,查看对方的联系方式,联系对方咨询.