Edhem Custovic

Sleep patterns and match performance in elite Australian basketball athletes

OBJECTIVES To assess sleep patterns and associations between sleep and match performance in elite Australian female basketball players. DESIGN Prospective cohort study. METHODS Seventeen elite female basketball players were monitored across two consecutive in-season competitions (30 weeks). Total sleep time and sleep efficiency were determined using triaxial accelerometers for Baseline, Pre-match, Match-day and Post-match timings. Match performance was determined using the basketball efficiency statistic (EFF). The effects of match schedule (Regular versus Double-Header; Home versus Away) and sleep on EFF were assessed. RESULTS The Double-Header condition changed the pattern of sleep when compared with the Regular condition (F(3,48)=3.763, P=0.017), where total sleep time Post-match was 11% less for Double-Header (mean±SD; 7.2±1.4h) compared with Regular (8.0±1.3h; P=0.007). Total sleep time for Double-Header was greater Pre-match (8.2±1.7h) compared with Baseline (7.1±1.6h; P=0.022) and Match-day (7.3±1.5h; P=0.007). Small correlations existed between sleep metrics at Pre-match and EFF for pooled data (r=-0.39 to -0.22; P≥0.238). Relationships between total sleep time and EFF ranged from moderate negative to large positive correlations for individual players (r=-0.37 to 0.62) and reached significance for one player (r=0.60; P=0.025). CONCLUSIONS Match schedule can affect the sleep patterns of elite female basketball players. A large degree of inter-individual variability existed in the relationship between sleep and match performance; nevertheless, sleep monitoring might assist in the optimisation of performance for some athletes.

Clock synchronisation in multi-transceiver HF radar system

The TIGER-3 radar is being developed as an “all digital” radar with 20 integrated digital transceivers, each connected to a separate antenna. Accurate coordination of all 20 transceivers is essential for both generation of transmit signals and collection and merging of receive data to form a standard Su-perDARN data set. This paper proposes a clock synchronisation method to coordinate the operation of the entire system using Field Programmable Gate Array (FPGA) technology. The method is a co-operation between hardware and software to achieve the necessary clock quality and synchronisation requirements. It is extremely important that the clock signals are kept aligned in time within specified bounds. To achieve this a 125Mhz common master clock is sent from a clock controller to a clock buffer, which then distributes the signals to the transceivers. In turn, each transceiver sends back a clock signal which is a buffered version of the common clock in the same bundle. In order to synchronise clocks on the transceivers, phase delays of round-trip clock paths are measured on the clock controller board with the accuracy of 31.25ps. The measurement is performed by shifting the common clock phase at a resolution of 1/256 of the clock period until the return clock and the common clock are in phase. Once the measurement cycle is complete, each transceiver adjusts the phase of its clock as directed by the clock controller. Experimental results show that the phase noise of the transmit signal generated from the synchronised clocks at transceivers is less than −100 dBc/Hz, while the SNR of the transmit signal is ≈ 90 dB for the entire 8–20 MHz range.

The Effect of Match Schedule on Accelerometry-Derived Exercise Dose during Training Sessions throughout a Competitive Basketball Season

Accelerometry-derived exercise dose (intensity × duration) was assessed throughout a competitive basketball season. Nine elite basketballers wore accelerometers during a Yo-Yo intermittent recovery test (Yo-Yo-IR1) and during three two-week blocks of training that represented phases of the season defined as easy, medium, and hard based on difficulty of match schedule. Exercise dose was determined using accumulated impulse (accelerometry-derived average net force × duration). Relative exercise intensity was quantified using linear relationships between average net force and oxygen consumption during the Yo-Yo-IR1. Time spent in different intensity zones was computed. Influences of match schedule difficulty and playing position were evaluated. Exercise dose reduced for recovery and pre-match tapering sessions during the medium match schedule. Exercise dose did not vary during the hard match schedule. Exercise dose was not different between playing positions. The majority of activity during training was spent performing sedentary behaviour or very light intensity activity (64.3 ± 6.1%). Front-court players performed a greater proportion of very light intensity activity (mean difference: 6.8 ± 2.8%), whereas back-court players performed more supramaximal intensity activity (mean difference: 4.5 ± 1.0%). No positional differences existed in the proportion of time in all other intensity zones. Objective evaluation of exercise dose might allow coaches to better prescribe and monitor the demands of basketball training.

Multi-biophysical event detection using blind source separated audio signals

This paper aims to use signal processing techniques to identify biophysical events using audio signals. The processing technique proposed is a combination of the mel frequency cepstral coefficients (MFCC) used as features, independent component analysis (ICA) and principle component analysis (PCA) algorithms to separate sources and noise. It was found that compressing the data into the energies of 26 filter banks mapped to the mel frequencies, sufficient descriptive information was conserved as validated by visually identifiable source signal patterns. Subsequently performing PCA isolated global background noise to an individual component. Further, by performing ICA, components contained independent and visually identifiable patterns that correlated to events associated with heart rate, squatting motion and involuntary abdominal movements. This componentized feature space provides an optimized source for building a discriminant function for the classifier used in the machine learning algorithm to provide simultaneous and automatic classification of these biophysical events.

Fetal Movement Measurement and Technology: A Narrative Review

Fetal movement counts have long been used as a measure of fetal well-being but with advancing technology, such counts have been supplanted as the primary measure. Despite the new technologies used in standard clinical practice, the stillbirth rate has not reduced significantly worldwide. Each method of assessing fetal movement has limitations with different methods performing better in different situations. No one method is universally superior. This paper aims to introduce the reader to the broad range of assessment methods, both potential and actual, used to determine fetal movement. These assessment methods are assembled into a taxonomy: maternal involvement, clinician involvement, technology-assisted, and automated technology. A brief historical and technological overview and the expected measurements of each assessment method are described. All reviewed methods have value, but actography appears to offer the most potential by complementing existing approaches. Further research is required to evaluate the suitability of fetal movement assessment and the response to it.

Improved impedance matching for SuperDARN TTFD antennas

This paper presents a novel antenna impedance matching system used in the latest Australian SuperDARN class HF radar, at Buckland Park, South Australia. Earlier radar designs used an off-the-shelf log-periodic wideband antenna that is significantly easier to match over the SuperDARN frequency band, but expensive to buy and mount, and had limited capability for azimuthal beamforming. The newer TTFD antenna, used in many recent SuperDARN radars, offers improvement in these areas, but is in essence a narrow band antenna. It is capable of wideband operation at the cost of being difficult to match, frequency dependant, high-impedance and complex load. Previous TTFD matching transformers utilising toroids have been measured and evaluated for their suitability for the Buckland Park radar. A new system based on an LC matching network circuit has been devised to replace them. The design approach and results of the new matching circuit are detailed.