ASTRI Proprietary Wireless Body Sensor Network – Opportunities and Challenges Andy Wu Principal Engineer IC Design, ASTRI Portable Analog Mixed Signal Design Nov 08, 2013
ASTRI Proprietary
Wireless Body Sensor Network – Opportunities and Challenges
Andy Wu
Principal Engineer IC Design, ASTRI
Portable Analog Mixed Signal Design
Nov 08, 2013
ASTRI Proprietary
Outline
• Wireless Body Sensor Network Overview
• Background
• WBSN Market Opportunities
• Sensor Node Architecture & Requirements
• Sample WBSN application
• Challenges of WBSN
• Future Development opportunities
• Conclusion
2
ASTRI Proprietary
Wireless Sensor Network Overview
What is Sensor Network?
• Clusters of communicating sensor nodes which collect and process
information
• Share information with neighboring or surrounding nodes or nearby data
collection stations.
Smart Building Transportation & Logistics
Industrial & Factory
Automation
Environmental Monitoring Healthcare
Security & Surveillance 3
ASTRI Proprietary
WBSN consists of multiple interconnected nodes, on, near, or within
a human body to provide sensing, processing and communication
capabilities.
Wireless Body Sensor Network
4
ASTRI Proprietary
Outline
• Wireless Body Sensor Network Overview
• Background
• WBSN Market Opportunities
• Sensor Node Architecture & Requirements
• Sample WBSN application
• Challenges of WBSN
• Future Development opportunities
• Conclusion
5
ASTRI Proprietary
• Moore’s Law is making sufficient
CPU performance available with
decreasing size and total power.
Why Wireless Body Sensors now?
• Advances in Materials Science
has resulted in novel sensing
materials for many Chemical,
Biological, and Physical sensing.
•Wireless Transceivers
are becoming smaller,
less expensive, and less
power hungry.
• Power supply improvements in
batteries and energy scavenging, are
expanding application options.
6
ASTRI Proprietary
Applications Enabled By WBSN
Vital Signs Monitoring (EMG, ECG, SPO2, Blood Pressure,
Glucose level, etc.)
Fitness & Gaming (Calorie Burnt, Heart Rate Monitor, Sleep
analysis, etc.)
Electronic Drug Delivery (Swallowable pill, Implanted
device, etc.) WBSN
7
ASTRI Proprietary
Outline
• Wireless Body Sensor Network Overview
• Background
• WBSN Market Opportunities
• Sensor Node Architecture & Requirements
• Sample WBSN application
• Challenges of WBSN
• Future Development opportunities
• Conclusion
8
ASTRI Proprietary
Wearable Technology Market
Wearable Technology Market
has potential to grow to 171
million units and exceed $6
billion by 2016
Source – IHS IMS Research
9
ASTRI Proprietary
10
Wireless Body Sensor Market
ONworld projects 515M units
In 2017.
Chip set ASP = $3
World wide market potential
@ US$1.5B by 2017 for
WBSN IC
“Advances in low power wireless communications, MEMS and multi-
sensor arrays have resulted in viable body area network applications for
clinical patient monitoring, assisted care, at-home chronic disease
management and general wellness,” says Mareca Hatler, ON World’s
research director.
10
ASTRI Proprietary
Outline
• Wireless Body Sensor Network Overview
• Background
• WBSN Market Opportunities
• Sensor Node Architecture & Requirements
• Sample WBSN application
• Challenges of WBSN
• Future Development opportunities
• Conclusion
11
ASTRI Proprietary
Typical WBSN Architecture
Master Node
• Portable device for central control, processing and interface
to the user
Slave Node
• Sensor nodes for biomedical information acquisition, signal
processing, wireless transmission.
Master Node
Slave Node Internet / LTE …
• ECG
• EMG
• Heart Rate
• SPO2
• Glucose
• Temperature
12
ASTRI Proprietary
Wireless Body Sensor Node
Sensor Node Architecture:
• Sensing module
• ADC / Analog Front End
• Microprocessor
• Power supply
• Communication Unit
Sensors
Power supply
ADC
MCU
An
tenn
a
Tx
Rx
13
ASTRI Proprietary
WBSN Sensors Requirement
Physiological Parameter Specifications
Electrocardiogram (ECG) Frequency: 0.3Hz – 100Hz
Amplitude: 0.25 – 1mV
Electromyogram (EMG) Frequency: 10Hz – 500Hz
Amplitude: 50uV – 6mV
Electroencephalogram (EEG) Frequency: 0.5Hz – 100Hz
Amplitude: 1uV – 100uV
Blood Pressure (BP) Systolic: 60-200mmHg
Diastolic: 50-110mmHg
Body Temperature 320C – 400C
Galvanic Skin Response (GSR) 0 – 100KW
Saturation of peripheral oxygen
(SPO2)
80-100%
Heart Rate (HR) 40-220 Beats per minute
Respiratory Rate (RR) 2-50 breaths per minute
Frequency 0.1 – 10Hz
Sensor
specifications
Physiological Sensors -> ECG/EMG/EEG/BP/SpO2/temperature monitoring
Biokinetic Sensors -> Acceleration and angular rate of rotation
Ambient Sensor -> Humidity, light, temperature
14
ASTRI Proprietary
Sensor How it works Data rate
ECG/EEG/EMG Measures the potential difference across
electrodes placed on different parts of the body
High
Pulse Oximetry Measures the absorbance of the red and
infrared light passing from one side to the other
side of a selected body’s anatomy (e.g. finger)
Low
Blood Pressure Measures the systolic and diastolic pressure
Low
Blood Glucose Measure the total charge passing through the
electrode proportional to the amount of glucose
in the blood.
Low
Accelerometer Measures the acceleration relative to freefall in
three axes
High
Gyroscope Measures the orientation based on the
principles of angular momentum.
High
Temperature Detect the temperature changes by measuring
the resistance changes of a silicon integrated
circuit
Very low
Typical WBSN Sensors
15
ASTRI Proprietary
WBSN Sensors Requirement
Sensitivity & Calibration
• Sensitivity and electrical performance of body-worn sensor can be
affected due to changes in environmental conditions and may require
recalibration
Placement specific
• Unintended placement can significantly degrade the data’s quality
Noise Reduction
• Signal processing required to handle motion artifacts, placement errors
and coupling from other noise sources
Low signal frequency & amplitude
• Relatively low sampling frequency and low data rate for signal
transmission
16
ASTRI Proprietary
WBSN Data Rate Requirement
Electrocardiogram (ECG)
• 72 kb/s
(500 Hz sample, 12-bit ADC, 12
channels)
Electromyogram (EMG)
• 768 kb/s
(4 kHz sample, 16-bit
ADC, 12 channels)
Electroencephalogram (EEG)
• 115.2 kb/s
(300Hz sample, 16-bit ADC, 24 channels)
Blood Pressure (BP)
• <10 kb/s
Body Temperature
• <10 kb/s
Pulse Oximetry (SPO2)
• 16 kb/s
(500Hz sample, 16-bit ADC, 2 channels)
17
ASTRI Proprietary
Comparison for Different Wireless Options
Standard BLE Zigbee ANT+ UWB WiFi
Peak Current 12-16mA 30-40mA 17mA 227mA 116mA
Frequency
Band
2.4GHz 2.4GHz 2.4GHz 3.1~10.6
GHz
2.4GHz;
5GHz
Data/Chip rate 1-Mbps 2-Mbps 1-Mbps 110Mbps 54-Mbps
Range 50m 100-300m 10m 10m 150m
Latency 2.5ms 20ms,
30ms from
sleep
~zero ~zero 1.5ms
Network
Topology
Star Mesh Mesh Peer-to-
Peer
Star
Interference
Immunity
Good Good Worse Good Good
Suitability for
WBSN
18
ASTRI Proprietary
WBSN Communication Unit Requirement
Reliable Data Transfer
• Low transmission power and small sized antenna can reduces signal to
noise ratios yet reliable data transfer is crucial
Compatibility and Interference
• Multiple sensory devices might interfere with each other, especially in the
unlicensed industrial, scientific and medical band
Robust against transmitted signals attenuation
• Signal attenuation from body shadowing, line of sight absorption of RF
signal
QOS & Security
• Vital sign data should be guaranteed to delivered within reasonable delay
• Encrypted data transmission to ensure confidentiality and privacy
19
ASTRI Proprietary
WBSN Signal Processing Platform
• Signal Processing
• Amplify weak signals
detected from sensors
• Filter noise injected from
other sources
• Extract signal of interest
with sufficient resolution
• Identify specific events as
a trigger for notification
• Store or transmit the
processed data based on
applications
20
ASTRI Proprietary
• Low noise, high gain analog front end to process the acquired signal
• High resolution low power analog to digital converter
• Low power microcontroller to process the data for storage or transmission
WBSN Signal Processing Platform
Sensor
Signals
Bluetooth RFIC options
RF in/out Digital
in/out
Power Management IP..
Inst
Amp
IN
SEL
A
IN
SEL
B SW CAP
Filter
VREF
Multi-
channel
12 Bit
ADC
Setup
REG
Data
REG .
.
21
ASTRI Proprietary
22
SpO2 Oxygen Saturation Percentage In Blood
Oxygen saturation percentage in blood can be
measured by placing finger between a dual light
(read and infrared) source and a photodetector, the
photodetector records the relative amount of each
color absorbed by arterial blood and transmitted
the data for calculation.
SpO2 signal specification:
Dark current: 5nA
Photodiode current: 0.05mA~37mA
LED Driver
Voltage Reference LED
Current Control
12bit DAC
14bit ADC
AFE
SPI I2C
PGA
Sensor
Input
RED
Amb(RED)
IR
Amb(IR)
TIA
OSC
PWM
22
ASTRI Proprietary
23
ECG Electrocardiogram
The ECG works mostly by detecting and amplifying
the tiny electrical changes on the skin that are
caused when the heart muscle "depolarises"
during each heart beat.
ECG signal specification:
Frequency range: 0.5Hz ~ 100Hz
Amplitude: 0.25mV ~ 1mV
23
ASTRI Proprietary
WBSN Signal Processing Unit Requirement
Low noise, high precision Analog Front End
• Instrumentation amplifier, high resolution ADC for signal
acquisition
Ultra low power microcontroller
• Minimization of both peak & average power consumption
• Deep sleep mode to shutdown the circuit when idle
• Dynamic voltage / frequency scaling to preserve power
Configurable Analog / Digital Circuit
• Programmable amplifier, configurable DAC/ADC to
interface with sensors
• Algorithm for noise reduction and signal processing in
digital domain
24
ASTRI Proprietary
Outline
• Wireless Body Sensor Network Overview
• Background
• WBSN Market Opportunities
• Sensor Node Architecture & Requirements
• ASTRI’s Solution for WBSN
• Challenges of WBSN
• Future Development opportunities
• Conclusion
25
ASTRI Proprietary
WBSN Signal Processing Platform
Key Features
• Chopper Stabilized Instrumentation Amplifier with Input Referred Noise <1uVrms (0.5Hz – 150Hz)
• Proprietary Switched Capacitor Programmable Gain Amplifier
• ARM Cortex-M0
• 12bit 200ksps/14bit 1Msps SAR ADC
• 12bit 20ksps Sigma Delta DAC
• TSMC 0.18um/SMIC 0.13um technologies
• ECG demos available
ASTRI’s BSN SOC Architecture TSMC 0.18um SOC Die Photo SMIC 0.13um SOC Die Photo
ASTRI’s ECG Demo
26
ASTRI Proprietary
WBSN ECG Demo …
ASTRI Proprietary
BLE Digital Baseband and RF
SEU’s
Design
GFSK Modem, Packet Synchronization, Whitening, CRC, AES
Encryption
ASTRI’s
Design
RF Controller with Channel Selection, FSM, interfaces to MCU,
Packet Processing
Electrical Specifications
Frequency Band: 2.400GHz ~ 2.4835GHz
No. of Channels: 40
Channel Width: 2MHz
Tx Power: -20dBm ~ 10dBm
Rx Sensitivity: -85dBm
Modulation: GFSK
Modulation Index: 0.45 ~ 0.55
Data Rate: 1Mbps
Robustness: Adaptive Frequency Hopping
Encryption: 128bit - AES
Packet Length: 80bit ~ 376bit
28
ASTRI Proprietary
Conventional RF Transceiver
ASTRI’s Re-configurable Digital RF
Re-configurable Digital RF
Conventional RF Transceiver ASTRI’s Re-configurable Digital RF Transceiver Key Specification
Cartesian Transmitter with RF Mixers &
Quadrature LO Power Hungry
Direct Digital Phase Conversion with Polar
Transmitter Low Power
TX Power: 16.6mW
Linear Power Amplifier Power Hungry Digital Amplitude Modulated non-linear Power
Amplifier Low Power
PA Power: 4mW
PA Efficiency:18%
Analog Controlled PLL Power Hungry
Fixed Frequency
All Digital Fractional-N PLL Low Power
Tunable from 2.4-2.4835GHz
PLL Power : 5.7mW
Frequency:
2.4GHz(BLE)
Fixed A/D,D/A, LPF & digital
modulation/demodulation Fixed Modulation
Scheme & Data Rate
Reconfigurable A/D, D/A, LPF & digital
modulation/demodulation Adaptive Modulation
Scheme & Data Rate
Data Rate: 1Msps(BLE)
Modulation Scheme:
GFSK
Hard to scale and integrate in nanometer process
Easy to port and integrate in lower geometry
65nm Process
ASTRI Proprietary
S1
EXT
Sensors
S2
S3
SAW
Wireless Body Sensor Signal Processing Platform
(ART 133)
RF Transceiver
TX
RX
Ultra Low Power RF IPs & SOC Platform for WBSN
Digital
Baseband
FPGA
Complete set of IPs for Wireless Body Sensor Network
30
ASTRI Proprietary
Outline
• Wireless Body Sensor Network Overview
• Background
• WBSN Market Opportunities
• Sensor Node Architecture & Requirements
• ASTRI’s Solution for WBSN
• Challenges of WBSN
• Future Development opportunities
• Conclusion
31
ASTRI Proprietary
Challenges of WBSN
Energy-efficient
• Power minimization technique for the sensor, controller and RF transmitter
• High energy density battery technology
• Power Scavenging methods
Increasing sensor integration
• Detection of human’s motion, temperature, heart rate and blood pressure simultaneously
Sensitivity and Accuracy
• Sensor and signal processing chain must truthfully capture the biological signals
32
ASTRI Proprietary
Challenges of WBSN
Higher processing performance
• Sophisticated signal processing algorithm and complex application may require higher computational performance
Robustness and Security
• Reliable operation of WBSN
• Secure transmission of data to ensure privacy
Value and Ease of use
• Perceived value must be higher than the price paid
• Small, unobtrusive, ergonomic, easy and intuitive to use
Safety and Biocompatibility
• Compliance to regulatory requirements
• No Harmful effects of long term continuous use
33
ASTRI Proprietary
Outline
• Wireless Body Sensor Network Overview
• Background
• WBSN Market Opportunities
• Sensor Node Architecture & Requirements
• ASTRI’s Solution for WBSN
• Challenges of WBSN
• Future Development opportunities
• Conclusion
34
ASTRI Proprietary
Future Development Opportunities
Physical characteristics of sensor materials and electronic circuits
• Flexible electronics
• Biocompatibility
Power Source issues • Remote battery charging
• Energy harvesting
Security, authentication and privacy issues
Ultra-low power signal processing
and RF transceiver
Network resource management
35
ASTRI Proprietary
Outline
• Wireless Body Sensor Network Overview
• Background
• WBSN Market Opportunities
• Sensor Node Architecture & Requirements
• ASTRI’s Solution for WBSN
• Challenges of WBSN
• Future Development opportunities
• Conclusion
36
ASTRI Proprietary
Conclusion
Wireless Body Sensor Network presents a billion dollar market of consumer applications
Sensor Node must be unobtrusive, easy to put on while operating at extremely low power, with reliable, secure wireless data transmission
Pretty of open research area in materials, packaging, energy supply, signal processing and wireless networking
Design trade off between energy consumption, processing, communication resources and the fidelity, throughput in the smallest possible package
Wireless Body Sensor Network could be the next life saving technology you need!
37
ASTRI Proprietary
End of Presentation Thank you. Questions are welcome.
Andy Wu [email protected]
Our corporate website: www.astri.org