Title: Fault Tolerant Sensor Network Routing for Patient Monitoring
1Fault Tolerant Sensor Network Routing for
Patient Monitoring
- Shanshan Jiang, Annarita Giani, Allen Yang, Yuan
Xue, and Ruzena Bajcsy
Vanderbilt University University of California at
Berkeley TRUST Autumn 2008 Conference November
11th, 2008
2Outline
- Motivation
- System and Network Architecture
- System Prototype and Implementation
- Network and Routing Model of the Backbone Network
- Optimization-based Routing Restoration of the
Backbone Network - Performance Evaluation
3Motivation
- Aging population
- According to the U.S. Census Bureau, the number
of people over the age of 65 is expected to hit
70 million by 2030, having doubled since 2000. - Health care expenditures
- Health care expenditures in the United States are
projected to rise to 15.9 of the GDP (2.6
trillion) by 2010. - The cost of health care for the nations aging
population has become a national concern.
4Motivation
- Wireless Sensor Networks
- Deploy wearable sensors on the bodies of patients
in a residential setting - Continuously monitor physiological signals (such
as ECG, blood oxygen levels) and other health
related information (such as physical activity) - Advantages
- Shift from a clinic-oriented, centralized
healthcare system to a patient-oriented,
distributed healthcare system - Reduce healthcare expenses through more efficient
use of clinical resources and earlier detection
of medical conditions - Challenges
- Performance, Reliability, Scalability, QoS,
Privacy, Security - More prone to failures, caused by power
exhaustion, software and hardware faults, natural
disasters, malicious attacks, and human errors
etc.
Provide fault-tolerant wireless communication
that can satisfy both the performance and
reliability requirements
5Outline
- Motivation
- System and Network Architecture
- System Prototype and Implementation
- Network and Routing Model of the Backbone Network
- Optimization-based Routing Restoration of the
Backbone Network - Performance Evaluation
6System and Network Architecture
7System and Network Architecture
Lower Tier Body Sensor Network Upper Tier
Multi-hop Wireless Backbone
Network
8Outline
- Motivation
- System and Network Architecture
- System Prototype and Implementation
- Network and Routing Model of the Backbone Network
- Optimization-based Routing Restoration of the
Backbone Network - Performance Evaluation
9System Prototype and Implementation
10System Prototype Experiment
11Outline
- Motivation
- System and Network Architecture
- System Prototype and Implementation
- Network and Routing Model of the Backbone Network
- Optimization-based Routing Restoration of the
Backbone Network - Performance Evaluation
12Network Model of the Backbone Network
- Backbone Network Performs Sensor Data Routing and
Forwarding - Network and Interference Model
- Topology G(V, E)
- All nodes have a uniform transmission range and
interference range - Two edges interfere with each other if they have
two nodes within the interference range of each
other - Traffic Demand Model
- df is the traffic demand of flow f, which is an
aggregation amount of all the sensor data
received at the sender of flow f - Be routed over multiple paths
- xf(e) denotes the amount of flow fs traffic
being routed on link e
Backbone Network
13Routing Model of the Backbone Network
- Metric for routing performance
- Minimum Flow Throughput Scaling Factor
- The minimum, over all flows, of the actual flow
throughput being routed divided by its throughput
demand - Optimal Routing Formulation
amount of traffic received at the destination
node rf
wireless channel constraint (necessary scheduling
condition)
flow conservation conditions
14Outline
- Motivation
- System and Network Architecture
- System Prototype and Implementation
- Network and Routing Model of the Backbone Network
- Optimization-based Routing Restoration of the
Backbone Network - Performance Evaluation
15Optimization-based Routing Restoration
- Discover Alternate Paths Bypassing the Failed
Nodes - Reactive Restoration
- Not reserve any network resource
- Deal with failures only when they occur through
network resource reallocation - Application
- Resource-limited System that allows performance
degradation upon failures - Proactive Restoration
- Reserve additional resources a priori
- Provide certain performance assurance for the
rerouted flows with a shorter restoration time - Application
- Life-critical System
- Admission Control
- Result in a lower network utilization before
failure occurs - Need to know the worst-case node failure
situations
16Optimization-based Routing Restoration
- Global Restoration
- All flows will be rerouted in order to get an
optimal utilization of the network - All flows have to be notified with the failure
information - End-to-end Restoration
- The flows from the failed path will be diverted
to a number of paths from its source to the
destination - Failure information has to be propagated to the
source nodes of the disrupted flows - Local Restoration
- Uses a set of bypaths to route around the failed
node locally - The restoration is locally activated
Increase
Increase
Repairing Time During Restoration
Network Performance after Restoration
17Optimization-based Routing Restoration
- End-to-end Restoration (1) Calculate Unaffected
Flow Truncations (2) Optimal Flow Augmentation
Restoration Formulation
- Local Restoration (1) Calculate Bypass Flows (2)
Optimal Bypass Restoration Formulation
18Outline
- Motivation
- System and Network Architecture
- System Prototype and Implementation
- Network and Routing Model of the Backbone Network
- Optimization-based Routing Restoration of the
Backbone Network - Performance Evaluation
19Performance Evaluation
Simulated Backbone Network
20Conclusion
- Three-Phase System Architecture
- Two-Tier Data Collection Network
- Routing Restoration of the Backbone Network
- Based on optimization theory and linear
programming approach - Reserve network resource or not
- Proactive Restoration
- Reactive Restoration
- Restoration scale
- Global Restoration
- End-to-end Restoration
- Local Restoration