Software Development for Atom based Safe and Sustainable BSN

IMPACT LAB

Project Report on “Safe, Secure and Sustainable Body Area

Networks using Intel Atom” funded by Intel

Impact Lab Research GoalEnvironmentally Aware

(physical), Performance aware (cyber),

Criticality Aware, Safe and Secure

Cyber-Physical Systems

Body-Area Sensor NetworksObjectives Minimize energy consumption Minimize health risks (safety) Ensure security and privacy Allow complex applications

Approach Infuse energy awareness Introduce high computation AADL modeling

Taxi Cab SchedulingObjectives Compute optimal route

with targets Minimize fuel

consumption Evaluate solutions

Approach Design proactive, spatio-

temporal schedules AADL modeling

Data CentersObjectives Minimize Energy

Consumption Maintain Performance

Approach Integrated Management Infuse Proactive Spatio-

Temporal Scheduling AADL Modeling

Project Goal

Body Sensor Networks (BSNs) – network of medical devices on human body are small scale cyber-physical system Critical infrastructures – used in medical applications Require to support life saving applications Involvement of human users require BSNs to be safe (reduce medical hazards)

and sustainable (provide seamless operation)

Complex application requirements (especially security protocols) demand powerful processors in BSN nodes

Atom is used as the BSN node processor to provide required computational capabilities

However, higher power dissipation of Atom, hampers the safe and sustainable operation of BSN nodes

Software Design of Computationally capable Safe and Sustainable Atom based BSN

Traditional Body Sensor Network

Present

Salient Features

Computationally incapable set of nodes

Heterogeneous hardware and software configuration

Constrained in energy – battery operated

No energy scavenging

Application requirements

Monitoring and Feedback Online detection of freezing of gait [1] in

Parkinson’s patients from on-body sensors Feedback through on-body actuators

Requirements Response within a small time window Fast Computation of windowed FFT and

associated signal processing

Security Physiological Value based Security [3] Combines signal processing with security

algorithms

Requirements

Hogs up 80 % of total RAM

Continuous Monitoring Seamless 24 hrs medical monitoring [2]

Requirements Increased lifetime of the sensors Battery less non-intrusive operation

Figure explains PVS Implementation

References1. M. B¨achlin et al. Online Detection of Freezing of Gait in Parkinson’s Disease Patients: A Performance Characterization. In Proc. of the 4th Intl. Conference on Body Area Networks, Apr. 20092. K. Venkatasubramanian et al. Ayushman: A Wireless Sensor Network Based Health Monitoring Infrastructure and Testbed. In Distributed Computing in Sensor Systems, pages 406–407, July 20053. K. Venkatasubramanian et al. Plethysmogram-based secure inter-sensor communication in body area networks. Military Communications Conference, 2008. IEEE, pages 1–7, Nov.

Proposed BSN System

Future

Computationally capable sensors Use Intel Atom as the sensor processor Addresses the computational requirements

of the present day applications

Homogeneous hardware and software platform Sensors running intel atom can have

stripped down versions of the same OS kernel

Resolves software compatibility issues

Energy Scavenging Incorporate energy scavenging hardware in

the network to sustain operation of the sensors

Supplement battery power Makes the BSN system greener

Challenges of Atom based BSNAtom can provide a uniform platform with highly capable BSN processors

Challenges with Atom Energy Efficiency

Relatively higher power footprint of Atom

Thermal Safety Possible high thermal footprint of Atom

Sustainability How long can energy scavenged from human sources sustain Atom operation ?

Atom background

Ultra low power processor for embedded applications However, order of magnitude higher power dissipation than the state-of-art BSN

node

IA-32 microarchitecture helps in easy application development Can use high level programming languages to develop applications

Six low power sleep states with ultra low power deep sleep state Sleep scheduling can be employed to reduce power consumption

Intel Speed Step technology enables seven different operating frequency levels Clock frequency control to reduce operating power

Sleep state and frequency control performed through easy ACPI support (through Model Specific Register (MSR) accesses)

Ayushman [2] health monitoring application is considered as the workload Ayushman has three phases of operation –

Sensing Phase – Sensing of physiological values (Plethysmogram signals) from the sensors and storing it in the local memory

Transmission Phase – Send the stored data to the base station in a single burst Security Phase – Perform network wide key agreement for secure inter-sensor communication using

Physiological value based Key Agreement Scheme (PKA) [3].

The Security phase occurs once in a day The Sensing phase and Transmission phase alternate forming a sleep cycle

Typical BSN Workload

Ayushman Phase Rate of Execution

Sensing 60 samples per second

Data Transmission every 5 seconds

PKA execution Once in 24 hrs

(the processor can sleep during sensing phase while it can be active during the transmission phase)

1/s sf t

Txt

Sensor CPU Utilization

Time

Sensing Phase

Transmission Phase

Security PhaseSleep Cycle

Ayushman Workload

Enables Sleep Scheduling

Frequency Throttling during security phase

4. K. Venkatasubramanian et al. Green and sustainable cyber-physicalsecurity solutions for body area networks. In BSN ’09: Proc. of the SixthIntl. Workshop on Wearable and Implantable Body Sensor Networks,pages 240–245, Washington, DC, USA

Management Strategies for Safety and Sustainability Challenge - Atom’s high TDP (2.2 W) with respect to present day

sensor nodes (~ 80 mW [4]) Remedy – Power budgeting through sleep scheduling and clock frequency control Road Blocks –

In a sleep mode the processor cannot compute Decrease in clock frequency increases computation time

Challenge - Increase lifetime of operation Remedy – include scavenging nodes in the BSN that will charge the Atom nodes

wirelessly and supplement battery Road Blocks –

The operation of scavenging sources are intermittent depending on the stochastic behavior of the host Often the scavenging nodes fail to provide appropriate power levels to the nodes

Intelligent design is required to achieve safety and sustainability while respecting the real time requirements of the applications

The strategies are closely related to the applications real time requirements.

Software Design Methodology

BSN Hardware model

Base Station

BSN Node BSN node Intel N270 single core processor

1.6 GHz clock frequency, 1 GB RAM Intel SpeedStep frequency control technology – useful for power

management 6 sleep states including one ultra low power sleep state (C6) – sleep

scheduling

Chipcon 2420 radio 2.45 GHz, 802.15.4 wireless standard Maximum Power dissipation (58 mW [4])

Scavenging Sources Body Heat, Ambulation, Respiration and Sun Light Wireless charging of BSN nodes from scavenging sources is

assumed Each source has a specified range upto which it can charge

nodesScavenging Sources

Wireless Charging

Base Station Atom based mobile phone

BSN node Software

The power consumption of Atom processor depends on the Operating System used Mobile Intel 945 GMCH board power consumption

Open Suse Linux = 11.7 W Moblin OS = 10.4 W

ACPI support required for accessing Intel SpeedStep frequency control and sleep states Moblin provides ACPI through which one can write to or read appropriate MSR

registers to – Control clock frequency Sleep States Measure core temperature

The BSN workload considered is the Ayushman application

Profiling Requirements

Thermal Safety – The maximum temperature of the skin in contact with the node should not exceed 39 ºC for 24 Hrs of operation

Thermal behavior of Atom under the given workload has to be evaluated

Sustainability – The available power from the scavenging sources should be able to meet the power demands of Atom node under the given workload

Power profiling of Atom processors during execution of Ayushman

Thermal Profiling Requires core temperature measurements for different operating points

of the Atom processor The Mobile Intel 945 GSE development platform (GMCH) provided by

Intel has digital thermal sensors The board thermal sensors were read from Model Specific Registers

The maximum core temperature (43 ºC) was observed during PKA execution

C6 Sleep State Run Ayushman

Turn On GMCH board

Read MSR

Log Temperature Data

Set OperatingFrequency

Thermal profiling methodology

PKA is the most power consuming computation in Ayushman [3] The difference between idle power and power during PKA execution

was measured using the GMCH board Idle power of Atom N270 processor was added to it to obtain PKA

power consumption

Power Profiling

AC Mains

Power Meter

Intel Atom N270 on Mobile Intel Chipset 945 GSE

Operating Mode (Percent throttling)

Power (W)

0 0.191

13 0.1864

25 0.17

37, 50, 62, 75 0.167

87 0.164

Power Measurement Set up Table showing Atom power consumption for PKA execution at different operating frequencies

Board Power Lead

Resource Consumption

Platform RAM Usage

Power (mW)

Computation Time (ms)

TelosB 80% 66 2186

Atom 0.006% 164 41

Resource Consumption for PKA execution

PKA computation in Ayushman involves signal processing of physiological signals as well as execution of security algorithms

Resource footprint of PKA is evaluated in terms of – RAM usage, Power Consumption, and Computation Time

Atom compared to TelosB provides very low RAM usage and computation time

However as expected it has around thrice the power consumption

Modeling Phase Industry Standard AADL language is used for modeling

Safety Analysis The temperature rise of human skin due to contact with Atom based

BSN node has to be evaluated The temperature rise occurs due to several physical phenomenon

and is modeled using the Penne’s bioheat equation -

2 4 4( ) ( )p b c r

dTC K T b T T SAR P A T Tdt

Sustainability Analysis Duty Cycling of Atom operation during Ayushman execution

Sleep mode (C6) during Sensing Phase Power consumption = Psleep for time ts

Active mode during data transmission phase Power consumption = Pactive + radio power Pradio for data transmission time ttx

Active mode during PKA execution Power Consumption = Pactive + PKA execution power PPKA for time tPKA

PKA involves transmission of security related information (vault) between two sensors The Atom processor must be in active state with the radio on. Power Consumption = Pactive + Pradio for

time tvault

Total energy consumption for n BSN nodes

x is the number of sleep cycles required in a day

[{( ) ( )} ( 1) / 2 ( )( 1) / 2]BSN s sleep Tx radio active PKA PKA Vault active radioE n t P t P P x t P n t P P n

Total EnergySensing Energy Transmission Energy PKA communication Energy (Pair wise

PKA)

PKA Computation Energy (Pair wise

PKA)

( ) ( 1) / 2 24 3600s Tx PKA Vaultt x t nx t t n n Total Sleep Cycle Time Pair wise PKA Execution Time

Sustainability Analysis Results

Four energy scavenging sources were considered Body Heat, Ambulation, Respiration and Sun Light

PM – Processor and radio

sleep scheduling NPM – Radio sleep sche-

ling NPNM – no sleep schedule

Scavenging Source

Available Power (W)

Scavenge Time (Hrs)

Body Heat 0.1 – 0.15 24

Ambulation 1.5 2

Respiration 0.42 6

Sun Light 0.1 3

PM NPM NPNM0

10

20

30

40

50

60

70

80

90

100

110

120Number of nodes sustained through scavenging for different design decisions

All FourBody Heat + Ambulation (Long term monitoring)Respiration + Ambulation(Athletes in training)Body Heat + Respiration(Patient Monitoring in Hospital)Ambulation + Sunlight(Perfomance Monitoring for outdoor sports)

Conclusions

Proper Sleep scheduling and Frequency throttling can be used to bring Atom’s power consumption to safe and sustainable levels

Atom based BSNs with upto 25 nodes can be sustained using scavenged energy from body heat and respiration

A model based engineering tool has also been developed in this process It uses industry standard AADL to model Analysis of Model is performed through an eclipse interface by developing java

based plug-ins

Problems Faced

Inaccuracies in Thermal Profiling Presence of heat sink on the Atom processor can cause additional thermal effects

which are not accounted for in the analysis Reliability of thermal sensors not known

Inaccuracies in Power Profiling Available board was used to determine the power consumption of Atom The power consumption may include several other components in the board The sense resistors across which power can be measured were not found due to

lack of documentation

Require stripped down version of Atom based development boards

Options to turn off components of the board has to evaluated

Thank You