Recent analysis of elite-level half-pipe snowboard competition has revealed a number of sport specific key performance
variables (KPV's) that correlate well to score. Information on these variables is difficult to acquire and analyse, relying
on collection and labour intensive manual post processing of video data. This paper presents the use of inertial sensors
as a user-friendly alternative and subsequently implements signal processing routines to ultimately provide automated,
sport specific feedback to coaches and athletes. The author has recently shown that the key performance variables
(KPV's) of total air-time (TAT) and average degree of rotation (ADR) achieved during elite half-pipe snowboarding
competition show strong correlation with an athlete's subjectively judged score. Utilising Micro-Electrochemical
System (MEMS) sensors (tri-axial accelerometers) this paper demonstrates that air-time (AT) achieved during half-pipe
snowboarding can be detected and calculated accurately using basic signal processing techniques. Characterisation of the
variations in aerial acrobatic manoeuvres and the associated calculation of exact degree of rotation (DR) achieved is a
likely extension of this research. The technique developed used a two-pass method to detect locations of half-pipe
snowboard runs using power density in the frequency domain and subsequently utilises a threshold based search
algorithm in the time domain to calculate air-times associated with individual aerial acrobatic manoeuvres. This
technique correctly identified the air-times of 100 percent of aerial acrobatic manoeuvres within each half-pipe
snowboarding run (n = 92 aerial acrobatic manoeuvres from 4 subjects) and displayed a very strong correlation with a
video based reference standard for air-time calculation (r = 0.78 ± 0.08; p value < 0.0001; SEE = 0.08 ×/÷ 1.16; mean
bias = -0.03 ± 0.02s) (value ± or ×/÷ 95% CL).
KEYWORDS: Time metrology, Statistical analysis, Gait analysis, Signal attenuation, Data analysis, Kinematics, Gold, Failure analysis, Receivers, Data communications
A biomechanical variable of interest to sprint coaches is foot-ground contact time. Contact time can be easily measured in a laboratory environment using a force platform, but is difficult to measure in the field. The focus of this paper is on the development and validation of an accelerometer-based method for estimating contact time during sprinting that could be used in the field. Tri-axial accelerometers were mounted on the tibia of the right leg of 6 subjects who performed maximal running trials from a stationary start, and running trials at a range of steady state speeds (jog, run and sprint). Ground contact times were measured using a force platform, and estimated from 3D accelerometer data. The mean error
between the force plate and accelerometer-based measures of contact time were 0 ± 12 ms, 2 ± 3 ms, and 1 ± 1 ms for the jog, run and sprint. For steps 1, 3 and 5 of the acceleration phase of the maximal sprint the mean errors were 8 ± 9 ms, 2 ± 5 ms, and 0 ± 1 ms respectively. Overall it was concluded from our analysis that close estimates of contact time during running can be obtained using body mounted accelerometers, with the best estimates obtained in conditions associated with the highest accelerations.
Motion analysis systems measure and calculate the position of markers fixed to the body but generally restrict measurement to the laboratory environment. In contrast, inertial measurement devices are small, lightweight and self-contained and data collection is not restricted to a laboratory. Most research using inertial measurement in human locomotion studies has focused on walking. This paper describes a wireless accelerometer-based method for measuring shank angular velocity during sprint running. The system consists of body-mounted electronics with a wireless connection to a PC programmed with the necessary equations to interpret the acceleration data. The hardware incorporates two sets of accelerometers measuring acceleration in each of the three axes. The two 3D accelerometers are fixed to a frame so that their axes are aligned and they are separated by a prescribed distance. By calculating the difference in acceleration between the two 3D sensors, the gravitational component and linear acceleration components are cancelled leaving the rotational acceleration components. An onboard microcontroller digitises the accelerometer signals and the data is transmitted wirelessly to a PC to calculate the angular velocity with minimal latency. Tests were conducted on several subjects running at a constant velocity for several different speeds. The angular rate output from the accelerometer-based system was compared to data obtained from an optical motion analysis system. Validation test results indicate an accurate result was obtained. The design's suitability for acquiring data during elite athlete sprint training is examined and other applications considered. Error reduction strategies will also be discussed.
In this paper the operation of capacitive soil moisture sensors are modeled using an electrical circuit analogue. This model aims to predict the response of capacitive sensors for a variety of soil types, moistures, soil conductivity and sensor operating frequencies. The model is extensively validated under a variety of conditions for a variety of sensor circuits and measurement techniques. The deposition of a conducting film composed of clay-like soil material over the sensing surface of a soil moisture sensor is shown to be the cause of hysteresis when the sensor is operated at low frequencies (10KHz). As the frequency is increased (10MHz) the effect of the conducting film becomes insignificant. Surface chemistry analysis techniques were used to identify the soil deposits on the conducting film. This research is motivated by the design of a small disposable sensor printed on a flexible plastic substrate measuring soil moisture as a function of the number of point contacts terminating on the insulated sensor electrode. In controlled conditions the sensor exhibits a linear response across most of its range to water content changes, but in some soils the reading becomes "stuck" on a high reading and does not return to a lower reading until the soil has dried considerably.
Wireless networks for sensor applications are required to support an adequate data throughput, range, node density and must consume as little power as possible. The Bluetooth specification has been designed for low power, medium data rate, cable replacement solutions and is therefore useful for wireless sensor networks. However it has a limitation of a maximum number of eight active devices per Bluetooth network (piconet). To be useful in wireless sensor networks a Bluetooth piconet requires a means to communicate to more than the maximum of eight active devices. This paper demonstrates techniques for expanding the usefulness of Bluetooth for wireless sensor networks. This has been done by using multiple access points, sharing the active member addresses of the Bluetooth piconet and utilising multiple piconet and scatternet tree structures. A comparison of existing piconet handoff mechanisms has been conducted and these have been evaluated for feasibility with the available hardware's limitations. Scatternet and piconet sharing mechanisms have been developed that allow a Bluetooth structure to support more than eight devices. These structures have been implemented with existing Bluetooth hardware and are compared via theoretical simulation and experimental results. The developed network of multiple Bluetooth access points combined with the developed Bluetooth structures provides several wireless networks suitable for sensor applications.
The use of environmental sensors in agriculture and precision agriculture applications is becoming more common, although implementation strategies and capital costs prohibit widespread adoption by many in the industry. Typical costs for agricultural monitoring systems can be in the tens of thousands of dollars per site. This paper presents low cost, wireless sensor nodes and a corresponding low power network. The nodes use biodegradable plastic to house the sensor, support electronics, RF transceiver and a 433 MHz antenna. In this paper the antenna design and network topology is discussed together with the propagation problems associated with a field environment in which the vegetation changes weekly. It is envisaged that such a platform could be ploughed in to the field at the end of its working life. The total cost of construction of the prototype platform is approximately $US10 per sensor. A communication protocol was also developed to allow many of these devices to be installed simultaneously and for the transmission of collected data and dynamic configuration and reprogramming. A receiver system allows for the collation and presentation of collected data. Low cost soil moisture sensors were coupled to the platform and installed in a commercial nursery wholesaler. Field trials of the network were successfully conducted.
This paper reviews current piezo-resistive characteristics pertaining to conventional and novel piezo-resistive strain transducers. These characteristics govern the performance of the sensor node. In this application, low power consumption, high signal to noise ratio (SNR), sensitivity and resolution in the sensor node are optimized for a distributed sensor network. In this low frequency application at < 100 Hz, it is found that electrical noise can limit the nominal resistance of the strain gauge to be used. By reducing the nominal resistance to lower the SNR, power consumption is increased. Optimization of the nominal resistance for excess noise and other material parameters must take place. Typical values have been used to explore the SNR over a range of resistance values and against frequency. The trade-off is also optimized in the volume and sheet resistance of the piezo-resistive material. Irreversible phenomena such as ageing and material creep are responsible for very low frequency drift (approaching DC) with respect to time and temperature. It is found that this drift is material specific and can be numerically compensated in situ. Maximizing sensitivity of the transducer is desirable to reduce the overhead at the sensor front-end. This overhead is shown to be dependant on gauge factor and the configuration of the strain-sensing circuit. The configuration of the strain-sensing circuit impacts on cost, complexity and SNR.
The measurement of sport specific performance characteristics is an important part of an athletes training and preparation for competition. Thus automated measurement, extraction and analysis of performance measures is desired and addressed in this paper. A tri-axial accelerometer based system was located on the lower back or swimmers to record acceleration profiles. The accelerometer system contained two ADXL202 bi-axial accelerometers positioned perpendicular to one another, and can store over 6 hours of data at 150Hz per channel using internal flash memory. The simultaneous collection of video and electronics touch pad timing was used to validate the algorithm results. Using the tri-axial accelerometer data, algorithms have been developed to derive lap times and stroke count. Comparison against electronic touch pad timing against accelerometer lap times has produced results with a typical error of better than ±0.5 seconds. Video comparison of the stroke count algorithm for freestyle also produced results with an average error of ±1 stroke. The developed algorithms have a higher level of reliability compared to hand timed and counted date that is commonly used during training.
In a sensor employing changes in capacitance the introduction of two additional reference electrodes can assist in the minimization and correction of errors introduced in the manufacturing process and from changes in environmental conditions. The two extra electrodes that reflect the maximum and minimum values of the measuring electrode are constructed during the same lithographic process and in close proximity to the active electrode. In preliminary trials of 79 sensors the uncorrected error in reading was nearly 20% of full scale and dropped to 4% of full scale after applying the correction. The technique with support electronics printed on flexible substrates allow the sensor to be small, integrated and “smart”.
Although historically among the most popular of sports, today, combative sports are often viewed as an expression of our savage past. Of primary concern are the long term effects of participating in these sports on the health of participants. The scoring of such sports has also been the subject of much debate, with a panel of judges making decisions about very quick events involving large sums of prize money. This paper describes an electronic system for use primarily in the sport of boxing, though it is suitable for martial arts such as karate and taekwondo. The technology is based on a previously described sensor platform and integrates a network of sensors on the athlete’s head, body and hands. Using a Bluetooth network, physical contacts are monitored in near real-time or post event on a remote computer to determine legal hits and hence derivative measures like scoring and final outcomes. It is hoped that this system can be applied to reduce the need for full contact contests as well as provide a more reliable method of determining the outcome of a bout. Other benefits presented here include the ability to analyse an athlete's performance post match or training session, such as assessing the efficacy of training drills and effects of fatigue.
A wireless network of multiple sensor nodes for monitoring large numbers of mobile agents is described and investigated. Wireless monitoring provides time critical information from a number of data sources allowing near real-time analysis of the collected data. The developed wireless network provides a moderate data rate, is able to support many wireless nodes and is a low power solution. Novel network structures have been developed to satisfy all of these requirements.
This paper evaluates a number of currently available wireless communication protocols, concluding that a Bluetooth wireless network satisfies the above criteria. To support a large number of devices, topologies using inter-piconet and piconet sharing methods have been developed. These network structures are outlined in detail and have been developed with the current Bluetooth hardware limitations in mind. The proposed wireless networks have been developed to be implemented with current Bluetooth hardware. A summary of network performance is included for each developed network structure, and from these figures an appropriate network structure has been chosen that satisfies the requirements of a wireless sensor network for monitoring mobile agents.
Traditional ECG viewing techniques use a flat file structure and the relationship of the leads to physical structure is not clear. State space allows a 3D representation that is more representative of anatomical structure and electrical activity. This paper demonstrates how novel visualisation techniques allow easier identification of anomalies. The methods employed use Taken’s state-space theory to plot the amplitude of user selected leads on the relative axes in the state space domain. By plotting the combined values of separate leads, the direct relationship between the different viewing angles of the electrodes can be seen. A graphical user interface (GUI) was developed to view MIT-BIH database files, and files from a cardiology clinic, in various state-space formats. This software allows the user to rotate the 3D models and provides a cross-sectional view of the plots at user selected coordinates. The usefulness of these models were determined by combining the orthogonal views of leads I, aVF, and V2. This enabled the user to collaborate the vector values of the lead locations with the conventional ECG characteristics.
An automated classification technique is desirable to identify the different stages of sleep. In this paper a technique for differentiating the characteristics of each sleep phase has been developed. This is an ideal pre-processor stage for classifying systems such as neural networks. A wavelet based continuous Morlet transform was developed to analyse the EEG signal in both the time and frequency domain. Test results using two 100 epoch EEG test data sets from pre-recorded EEG data are presented. Key rhythms in the EEG signal were identified and classified using the continuous wavelet transform. The wavelet results indicated each sleep phase contained different rhythms and artefacts (noise from muscle movement in the EEG); providing proof that an EEG can be classified accordingly. The coefficients founded by the wavelet transform have been emphasised by statistical techniques. Hypothesis testing was used to highlight major differences between adjacent sleep stages. Various signal processing methods such as power spectrum density and the discrete wavelet transform have been used to emphasise particular characteristics in an EEG. By implementing signal processing methods on an EEG data set specific rules for each sleep stage have been developed suitable for a neural network classification solution.
KEYWORDS: Data modeling, Algorithm development, Video, Signal analysis, Motion models, Calibration, Modeling, Signal processing, Image segmentation, Error control coding
Triaxial accelerometers have been used to measure human movement parameters in swimming. Interpretation of data is difficult due to interference sources including interaction of external bodies. In this investigation the authors developed a model to simulate the physical movement of the lower back. Theoretical accelerometery outputs were derived thus giving an ideal, or noiseless dataset.
An experimental data collection apparatus was developed by adapting a system to the aquatic environment for investigation of swimming. Model data was compared against recorded data and showed strong correlation. Comparison of recorded and modeled data can be used to identify changes in body movement, this is especially useful when cyclic patterns are present in the activity. Strong correlations between data sets allowed development of signal processing algorithms for swimming stroke analysis using first the pure noiseless data set which were then applied to performance data. Video analysis was also used to validate study results and has shown potential to provide acceptable results.
KEYWORDS: Sensors, Data communications, RF communications, Data storage, Packaging, Power supplies, Data acquisition, Microsensors, Calibration, Voltage controlled current source
A modular self-contained modular platform is described, for easy integration with micro sensors and other sensor elements. The platform is designed to be physically robust and suitable for harsh environments. The platform features switch able power modes, signal processing capabilities and extensive I/O for sensor and external device communications, data download and transmission. The modular design allows flexible implementation of required functionality depending on the particular application and also provides flexibility for packaging solutions. Two practical applications of the platform are presented to demonstrate its use. Firstly a variety of human exercise activities are investigated using accelerometers. Secondly a weather station made up of environmental sensors using off the shelf and prototype sensors is described. Both of these applications differ greatly in their operational requirements. These implementations demonstrate the adaptability of the platform for different applications.
Integrated smart sensors are quickly becoming an industry norm and often require multi-stage, multi-skilled design. This paper describes the fabrication of a temperature compensated light sensor using only a basic fabrication laboratory. A complete description of how to build the sensing elements, support electronics and communications is described and test results are presented. The construction of a light sensor using a shottkey barrier diode between the nickel and n-type silicon has been previously described by the authors. In this two such sensors are used as active and passive sensor elements to compensate for temperature effects. The outputs from each are differentially amplified, conditioned and buffered using an LM324 die to provide a temperature compensated output. Further reduction in size is possible when the temperature sensors are mounted front to back on a single silicon substrate and coupled directly to the LM324 die. External communications are only power, ground and an analogue signal.
KEYWORDS: Calibration, Sensors, Magnesium, Time metrology, Packaging, Microelectronics, Error control coding, Microtechnology, Kinematics, Chemical elements
A triaxial accelerometer calibration technique that evades the problems of the conventional calibration method of aligning with gravity is proposed in this paper. It is based on the principle that the vector sum of acceleration from three sensing axes should be equal to the gravity vector. The method requires the accelerometer to be oriented and stationary in 6 different ways to solve for the 3 scale factors and 3 offsets. The Newton-Raphson method was employed to solve the non-linear equations in order to obtain the scale factors and offsets. The iterative process was fast, with an average of 5 iterations required to solve the system of equations. The accuracy of the derived scale factors and offsets were determined by using them to calculate the gravity vector magnitude using the triaxial accelerometer to measure gravity. The triaxial accelerometer was used to measure gravity 264 times to determine the accuracy of the 44 acceptable sets of scale factors and offsets derived from the calibrations (gravity was assumed to equal 9.8000 ms-2 during the calibration). It was found that the best calibration calculated the gravity vector magnitude to 9.8156 ± 0.4294 ms-2. This equates to a maximum of 4.5% error in terms of a constant acceleration measurement. Because of the principle behind this method, it has the disadvantage that noise/error in only one axis will cause an inaccurate determination of all the scale factors and offsets.
Measurement characterisation of error has been performed from synthetic acceleration data. A variety of kinematics computation techniques were investigated and then applied. Data acquisition specifications, including analogue to digital conversion (ADC) resolution (6, 8, 10, 12 and 14 bit) and sampling rate (100, 150, 200, 250, 300, 375, 500 and 600 Hz), have been varied to investigate the effect on the accuracy of kinematics data with respect to short (20m) displacements. The magnitude of the errors in acceleration, velocity and position are reported for the simulated data. Also, error reduction techniques, including over-sampling and oversampling/multiple point averaging/reduced data transmission, were implemented to examine their effectiveness. The results of this investigation show that MEMS accelerometers are subjected to significant amount of errors, and require accurate calibration and characterisation of errors. Error reduction techniques are also necessary to ensure accurate computation of kinematics information.
The use of silicon-based sensors requires the addition of external support electronics to allow for compatibility with external logging and display instruments. The development of a smart sensor technology, where the support electronics are incorporated into the sensor allows for a simpler interface. To achieve this integration techniques are required for the connection of substrate sensors with drive and support circuitry (operational amplifiers and CMOS circuitry), for effective encapsulation into a single packaged device. In this paper a literature review of basic peripheral and internal interconnect techniques is presented. Three techniques for interconnects were experimentally investigated (wraparound, thermomigration and etched micro via’s) using in-house fabrication equipment and the results presented and discussed. An integrated "smart" light sensor was constructed by forming a schotkey diode on n-type silicon. The sensor was integrated with a commercially available LM324 quad operational amplifier die and etched micro via`s were used to connect between the electronics on one side and the silicon sensor on the other side so forming a smart sensor. The light level sensor was calibrated and tested for suitability as a solar intensity monitor.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.