MIT CSAIL 1 Problem: 3G/LTE is a battery hog Up to 14 hours on 2G - - PowerPoint PPT Presentation
Shuo Deng, Hari Balakrishnan MIT CSAIL 1 Problem: 3G/LTE is a battery hog Up to 14 hours on 2G Up to 6.5 hours on 3G 100% 90% 80% 70% Energy Percentage 60% 50% 40% 30% 20% 10% 0% Goal: Reduce Energy News IM
1
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% News IM Micro-blog Game Email Social Network Finance Energy Percentage Data Active, No Data High PowerIdle State Switch
“Up to 14 hours on 2G” “Up to 6.5 hours on 3G”
2
3
Cell_DCH (Active) Cell_FACH (High Power Idle) Cell_PCH/IDLE (Idle) Inactivity Timer t1 Inactivity Timer t2
4
IDLE
Active
High Power Idle
IDLE
t1 t2
t1 t2 IDLE
5
Active Idle Inactivity Timer t1 Inactivity Timer t2 High Power Idle Active Idle
Switching between states takes time(1~3 seconds), and
Signaling overhead
6
IDLE
t1
IDLE
t2 Active Idle
A traffic-aware design to control radio state transitions
MakeIdle: when to switch to Idle MakeActive: when to switch to Active
Experimental evaluation on real usage data
Energy reduction up to 75% across different carriers
7
Active Idle MakeIdle MakeActive
8
Control Module App1 App2 App3 Modified Socket Layer Fast Dormancy Interface 3G Radio Socket Calls Packet/Socket Call Information Trigger Fast Dormancy Data Switch to Idle Switch to Active Idle Active
9
Power time
IDLE
t1
IDLE
t2
10
time Power > If , should switch to Idle mode to minimize the energy consumption. IAT time Power If IAT > threshold, should switch to Idle mode. Idle Active MakeIdle Inter Arrival Time
Predict whether the IAT will be greater than threshold Wait for a short period of time twait, if no packet comes,
Why: the longer the network is idle, the longer it is
11
Idle Active MakeIdle
12
… Previous N packets
) (
_ _ _ switch state no switch state
Energy Energy
Idle Active MakeIdle
P(IAT>t+threshold | IAT>t)
13
time Power time Power time Power Current Scheme With MakeIdle With MakeActive
Reduce the number of state switches by introducing a
How much delay for each request?
Fixed delay bound Learning algorithm
14
Idle Active MakeActive
Energy profiling
Power consumption profiles for 4 US major carriers:
AT&T, Verizon, T-Mobile, Sprint
Trace driven simulation
Tcpdump traces for real usage data, collected from 9
users, 28 days in total
15
1 2 3
16
10 20 30 40 50 60 70 80
1 2 3
Energy Saved (%) User ID Increased Signaling Overhead User ID 40x 35x 30x 25x 20x 15x 10x 5x 95% IAT across users 95% IAT per user MakeIdle Prior knowledge of IAT (Oracle)
5 10 15 20 25 30
1 2 3
17
Prediction Accuracy
Idle Active MakeIdle False Positive (FP) or False Negative (FN) 95% IAT across users FP 95% IAT per user FN MakeIdle FP 95% IAT across users FN 95% IAT per user FP MakeIdle FN User ID
18
10 20 30 40 50 60 70 80 T-Mobile AT&T Verizon 3G Verizon LTE Energy Saved (%) Carriers 95% IAT across users 95% IAT per user MakeIdle MakeIdle+MakeActive
Inactivity timer reconfiguration
Statistical method [Falaki et al, 2010]: 95 percentile
packet inter arrival time
Applications-Involved Design
TailEnder [Balasubramanian et al, 2009]: each
application specifies its delay tolerance
TOP [Qian et al, 2010]: application predict the gap
between its own traffic transmissions
TailTheft [Liu et al, 2011]: application specifies delay
tolerance and predicts transmission duration
19
A traffic-aware design to control state transitions of
Require no modifications of the applications Save 3G/LTE energy consumption by up to 75% across
20