Body Sensor Networks Reliability Enhancement Algorithms in Sport and Biomedical Applications
Author(s)
Primary Supervisor
Thiel, David
Other Supervisors
O'Keefe, Steven
Year published
2017-05
Metadata
Show full item recordAbstract
Wireless Networking is the fastest growing segment of the networking industry. In medical and sport industries, Body Sensor Networks (BSNs) are used to monitor the human body. BSNs have numerous design challenges, including network topology, reliable data delivery in a timely manner, and minimal power consumption. Other performance parameters, such as throughput and coverage, are equally important.
The object of this project was to achieve an efficient and optimal design of a BSN that acquire movement and physiological data, using sensors at various places on the body and sending data to a receiving unit (gateway) located ...
View more >Wireless Networking is the fastest growing segment of the networking industry. In medical and sport industries, Body Sensor Networks (BSNs) are used to monitor the human body. BSNs have numerous design challenges, including network topology, reliable data delivery in a timely manner, and minimal power consumption. Other performance parameters, such as throughput and coverage, are equally important. The object of this project was to achieve an efficient and optimal design of a BSN that acquire movement and physiological data, using sensors at various places on the body and sending data to a receiving unit (gateway) located on the chest for re-transmission of information to a remote computer (locally or in the cloud) for further processing and presentation. This is important in sport for the athletes, coaches and the viewing public, it provides real-time monitoring for rapid improvement in player performance during training or rehabilitation sessions. An investigation into the best node locations on the human body was made to achieve maximum connectivity to a receiving unit (gateway) on the chest, and the best angular window for the nodes to transmit data during typical human movements, such as running and walking. Preliminary results showed that, while the distance between the transmitting and receiving nodes changes significantly, the presence of scattering from limbs causes the most significant effect on the received signal strength. These measurements demonstrated that a received signal strength greater than -70dBm (the radio communications threshold) can vary from 20% to 74% of the recorded time for different node locations. The use of an accelerometer sensor at each node allows these positions to be identified in real-time for burst transmission to occur reliably. Wireless accelerometer sensor modules were used to determine the link performance by recording the data and traffic lost on different runners and for different transmitter locations on the human body (foot, leg and arm), to identify these time windows from the diverse angles of rotation of the human limbs during running. The results showed that the sensor on the wrist gives the best connectivity. An approximate swing time calculation algorithm was employed to find the swing time effect on these losses. Different data rates were tested against traffic loss and showed 98% and 62% of reliability at 250kbps and 2Mbps respectively. With a central node on the chest, a novel energy-efficient time multiplexing transmission method for on-body wireless communication was implemented based on the human rhythmic movement of running. The running style of each individual allows the network to self-calibrate the communication scheme so that transmissions occur only when high link reliability is predicted. This technique takes advantage of the periodic running actions to implement a dynamic time division multiple access (TDMA) strategy for a five-node body network with very little communication overhead, long sleep times for the sensor transceivers and robustness to communication errors. The results showed that all wireless communications were successful, except when two nodes attempted to use the transmission medium simultaneously. An aggregated network reliability of 90% was achieved, compared to 63% when employing traditional time multiplexing algorithms. The results also showed a trade-off between the channel occupancy and traffic generated to provide high channel reliability for the body network. An advanced gesture transmission technique was adopted to collect acceleration data and to predict the best limb position for communications while the athlete is moving. This reduced the overall transmission power and increased the reliability. As a result, data losses were reduced from 30% to 1%, compared to continuous communications. Experimental measurements were reported on five human subjects moving in formations on a grassed field with a smart algorithm that takes advantage of the Carrier Sensing Multiple Access (CSMA) technique. The algorithm forces the sensor nodes to repeatedly sense the carrier frequency to recognize when other nodes are in or out of coverage. The test reports different sink node locations on the body and the effect of transmitter power on the network reliability. A sink placed on the head position provides a normally distributed coverage of approximately 5 meters with link reliability of at least 80%. The smart algorithm showed 100% successful wireless communications between sink nodes within coverage when the nodes were programmed not to use the transmission medium simultaneously. The neighbours list and time of existence of all neighbours for each node were recorded and modified accordingly at each stage of the test. The availability of this information in real-time can be used to determine athlete proximity in the playing field and their performance with respect to time spent on one or multiple locations on the field.
View less >
View more >Wireless Networking is the fastest growing segment of the networking industry. In medical and sport industries, Body Sensor Networks (BSNs) are used to monitor the human body. BSNs have numerous design challenges, including network topology, reliable data delivery in a timely manner, and minimal power consumption. Other performance parameters, such as throughput and coverage, are equally important. The object of this project was to achieve an efficient and optimal design of a BSN that acquire movement and physiological data, using sensors at various places on the body and sending data to a receiving unit (gateway) located on the chest for re-transmission of information to a remote computer (locally or in the cloud) for further processing and presentation. This is important in sport for the athletes, coaches and the viewing public, it provides real-time monitoring for rapid improvement in player performance during training or rehabilitation sessions. An investigation into the best node locations on the human body was made to achieve maximum connectivity to a receiving unit (gateway) on the chest, and the best angular window for the nodes to transmit data during typical human movements, such as running and walking. Preliminary results showed that, while the distance between the transmitting and receiving nodes changes significantly, the presence of scattering from limbs causes the most significant effect on the received signal strength. These measurements demonstrated that a received signal strength greater than -70dBm (the radio communications threshold) can vary from 20% to 74% of the recorded time for different node locations. The use of an accelerometer sensor at each node allows these positions to be identified in real-time for burst transmission to occur reliably. Wireless accelerometer sensor modules were used to determine the link performance by recording the data and traffic lost on different runners and for different transmitter locations on the human body (foot, leg and arm), to identify these time windows from the diverse angles of rotation of the human limbs during running. The results showed that the sensor on the wrist gives the best connectivity. An approximate swing time calculation algorithm was employed to find the swing time effect on these losses. Different data rates were tested against traffic loss and showed 98% and 62% of reliability at 250kbps and 2Mbps respectively. With a central node on the chest, a novel energy-efficient time multiplexing transmission method for on-body wireless communication was implemented based on the human rhythmic movement of running. The running style of each individual allows the network to self-calibrate the communication scheme so that transmissions occur only when high link reliability is predicted. This technique takes advantage of the periodic running actions to implement a dynamic time division multiple access (TDMA) strategy for a five-node body network with very little communication overhead, long sleep times for the sensor transceivers and robustness to communication errors. The results showed that all wireless communications were successful, except when two nodes attempted to use the transmission medium simultaneously. An aggregated network reliability of 90% was achieved, compared to 63% when employing traditional time multiplexing algorithms. The results also showed a trade-off between the channel occupancy and traffic generated to provide high channel reliability for the body network. An advanced gesture transmission technique was adopted to collect acceleration data and to predict the best limb position for communications while the athlete is moving. This reduced the overall transmission power and increased the reliability. As a result, data losses were reduced from 30% to 1%, compared to continuous communications. Experimental measurements were reported on five human subjects moving in formations on a grassed field with a smart algorithm that takes advantage of the Carrier Sensing Multiple Access (CSMA) technique. The algorithm forces the sensor nodes to repeatedly sense the carrier frequency to recognize when other nodes are in or out of coverage. The test reports different sink node locations on the body and the effect of transmitter power on the network reliability. A sink placed on the head position provides a normally distributed coverage of approximately 5 meters with link reliability of at least 80%. The smart algorithm showed 100% successful wireless communications between sink nodes within coverage when the nodes were programmed not to use the transmission medium simultaneously. The neighbours list and time of existence of all neighbours for each node were recorded and modified accordingly at each stage of the test. The availability of this information in real-time can be used to determine athlete proximity in the playing field and their performance with respect to time spent on one or multiple locations on the field.
View less >
Thesis Type
Thesis (PhD Doctorate)
Degree Program
Doctor of Philosophy (PhD)
School
Griffith School of Engineering
Copyright Statement
The author owns the copyright in this thesis, unless stated otherwise.
Subject
Algorithms
Biomedical Applications
Body Sensor