diff --git a/manuscript/20210908_Adriaens_UWB_.log b/manuscript/20210908_Adriaens_UWB_.log index b6748e8e1f0dee792e5dc76b05cfe1230446709c..a72a5a010d8ef2c4a490db31aafbcd511a925178 100644 --- a/manuscript/20210908_Adriaens_UWB_.log +++ b/manuscript/20210908_Adriaens_UWB_.log @@ -1,4 +1,4 @@ -This is pdfTeX, Version 3.141592653-2.6-1.40.22 (MiKTeX 21.2) (preloaded format=pdflatex 2021.5.7) 15 SEP 2021 15:10 +This is pdfTeX, Version 3.141592653-2.6-1.40.22 (MiKTeX 21.2) (preloaded format=pdflatex 2021.5.7) 15 SEP 2021 15:37 entering extended mode **./20210908_Adriaens_UWB_.tex (20210908_Adriaens_UWB_.tex @@ -277,7 +277,7 @@ ers/adria036/AppData/Local/Programs/MiKTeX/fonts/type1/public/amsfonts/cm/cmr7. pfb><C:/Users/adria036/AppData/Local/Programs/MiKTeX/fonts/type1/public/amsfont s/cm/cmsy10.pfb><C:/Users/adria036/AppData/Local/Programs/MiKTeX/fonts/type1/pu blic/amsfonts/cm/cmsy7.pfb> -Output written on 20210908_Adriaens_UWB_.pdf (8 pages, 153273 bytes). +Output written on 20210908_Adriaens_UWB_.pdf (8 pages, 153974 bytes). PDF statistics: 310 PDF objects out of 1000 (max. 8388607) 0 named destinations out of 1000 (max. 500000) diff --git a/manuscript/20210908_Adriaens_UWB_.pdf b/manuscript/20210908_Adriaens_UWB_.pdf index 61c50b47d0870019cd64fbbd5fddf997636a203a..961098c5eb9be60a873af28880a01aa0561a1b8d 100644 Binary files a/manuscript/20210908_Adriaens_UWB_.pdf and b/manuscript/20210908_Adriaens_UWB_.pdf differ diff --git a/manuscript/20210908_Adriaens_UWB_.synctex.gz b/manuscript/20210908_Adriaens_UWB_.synctex.gz index 38bdf3f7e75f2ec9feb48bb9b83c809f64dd4001..34e7e33e5c3e4f5d5fd3c0dedc31b069b6a3e0fb 100644 Binary files a/manuscript/20210908_Adriaens_UWB_.synctex.gz and b/manuscript/20210908_Adriaens_UWB_.synctex.gz differ diff --git a/manuscript/20210908_Adriaens_UWB_.tex b/manuscript/20210908_Adriaens_UWB_.tex index 5c5a904e96473b8b725ec451aa9a38f1e14cdc5c..cb548655f96fd64bf272cd8fea1bede82810e709 100644 --- a/manuscript/20210908_Adriaens_UWB_.tex +++ b/manuscript/20210908_Adriaens_UWB_.tex @@ -57,12 +57,13 @@ \item Implications +Precision livestock farming solutions typically aim Cow behavior, varying from feeding and drinking, over where and when animals rest, lie down and ruminate, and social interactions potentially offers new paths both for research and commercial decision support systems that can help the farmer manage their herd, optimize production and quickly act upon health and welfare problems. A continuous and essential step to better unlock the potential of cow behavioral analyses is the development of new sensor technologies, and more importantly, the corresponding data-processing algorithms that allow just and timely interpretation of their data. -Ultra wide band is technology allows the transmission of high amounts of data over small distances with very low energy. In an indoor positioning system, Radio-Frequency identification (RFID) signals are transmitted across a wide bandwidth and captured by an antenna. The tags worn by the individual cows allow precise localization of the animals with low power usage, even in cluttered indoor environments {\color{b}{1}}. Upon development of appropriate data interpretation algorithms, indoor positioning systems allow studying and monitoring cow behavior, including general activity, resting, feeding, drinking and social interactions with a single sensor system, giving it a relative advantage over e.g. commercially available pedometer systems. Despite its continuous development and high potential for animal monitoring, UWB-based positioning is yet sparingly adopted for livestock applications. A main reason for this is, similar to many new sensor technologies, the lack of proper data-processing algorithms that can translate raw data time series into information valuable to the farmer on which decision support can be based. In the case of indoor positioning systems, data interpretation is complicated by inaccuracy and noise in the time series, missing data, and the heteroscedasticity in the time series. This heteroscedasticity partly results from differences in behavior, but also depends on e.g. the position of the animal in the barn with regard to the antenna and interactions of the signal with metal (e.g. the feeding rack) and water bodies (e.g. other cows). This heteroscedasticity and the unpredictability of noise impair straightforward interpretation of the positioning data and its derivatives (e.g. distance travelled). Nonetheless, +Ultra wide band is technology allows the transmission of high amounts of data over small distances with very low energy. In an indoor positioning system, Radio-Frequency identification (RFID) signals are transmitted across a wide bandwidth and captured by an antenna. The tags worn by the individual cows allow precise localization of the animals with low power usage, even in cluttered indoor environments {\color{b}{1}}. Upon development of appropriate data interpretation algorithms, indoor positioning systems allow studying and monitoring cow behavior, including general activity, resting, feeding, drinking and social interactions with a single sensor system, giving it a relative advantage over e.g. commercially available pedometer systems. Despite its continuous development and high potential for animal monitoring, UWB-based positioning is yet sparingly adopted for livestock applications. A main reason for this is, similar to many new sensor technologies, the lack of proper data-processing algorithms that can translate raw data time series into information valuable to the farmer on which decision support can be based. In the case of indoor positioning systems, data interpretation is complicated by inaccuracy and noise in the time series, missing data, and the heteroscedasticity in the time series. This heteroscedasticity partly results from differences in behavior, but also depends on e.g. the position of the animal in the barn with regard to the antenna and interactions of the signal with metal (e.g. the feeding rack) and water bodies (e.g. other cows). This heteroscedasticity and the unpredictability of noise impair straightforward interpretation of the positioning data and its derivatives (e.g. distance traveled). Nonetheless, dedicated data processing enables classification of behavior from multiple animals with high accuracy. -In this study, a methodology to identify lying behavior of dairy cows using a UWB-based indoor positioning system was developed and validated against the lying bouts returned by a commercial accelerometer-based system. +In this study, a methodology to identify lying behavior of dairy cows using a UWB-based indoor positioning system was developed and validated against the lying bouts returned by a commercial accelerometer-based system. The methodology relies on the detection of changepoints in the position time series, {\color{r}elegantly avoiding fixed thresholds or severe assumptions on the statistical properties of the data.} \end{itemize} @@ -112,6 +113,10 @@ ON THE PENALTY \item Huge heteroscedasticity in the data, not only between behaviors but also between cows and within cows between days \end{itemize} +ON THE OPTIMIZATION +PELT- computational cost is linear in the number of observations -- motivation to detect at day level +PRUNED EXCAT LINEAR TIME METHOD OPTIMIZATION + {\color{r}TODO: rules to interpret the direction of the change in terms of level -- if higher --- higher : no change}