added activity to related work

tex_review/chapters/relatedwork.tex

@@ -52,25 +52,29 @@ while still increasing the quality (triangle-edges directly adhere to architectu
 for truly continuous transitions along the surface spanned by all triangles. 
 
 %activity recognition
-\addy{To go into the third dimension, cf. walking continuously along stairs, using barometer. The most basic approach is using absolute pressure, however this value highly differs between single buildings. Thus relative approaches initializing with a zero pressure or using a sliding window are more often integrated into the movement model, as it provides continuous updates with every incoming barometer reading. Thanks to the underlying mesh, the pedestrians current activity can also be used to provide continues floor changes. This is done by assigning different types to the triangles of the mesh, e.g. stair, floor or elevator. Depending on the recognized activity, the system is now able to allow or restrict the movement in certain areas of the building. }
+\addy{To be able to go into the third dimension, cf. walking continuously along stairs, many localization system consider using a barometer \cite{xia2015using, li2013using, Nurminen13-PSI}. 
+As absolute pressure readings highly differ between season, time of day and sometimes even individual areas of the building, approaches considering the relative change of pressure are mostly preferred. 
+For example, this can be done by initializing with a zero pressure at the beginning of every walk or using some sliding window \cite{zheng20163d, tian2015smartphone}. 
+Such approaches can then be easily integrated into the movement model, as they provide continuous updates with every incoming barometer reading.
+However, the accuracy then often depends on the quality of the sensor's (raw) measurements. 
+Using smoothing statistics like Kalman filtering or a moving average, might thus be valid options to further improve the results.
+Thanks to the underlying mesh another possibility to provide continues floor changes can be considered: the pedestrian's current activity. 
+This is done by assigning different labels to the triangles of the mesh, e.g. stair, floor or elevator. Depending on the recognized activity, the system is now able to allow or restrict the movement in certain areas of the building. A probabilistic model for incorporating this into the particle filter, will be presented within this work.}
 
-\addy{In recent years, many different activity recognition approaches could be presented for wearable sensors \cite{}. 
+\addy{In recent years, many different activity recognition approaches could be presented for wearable sensors \cite{shoaib2015survey}. 
 They occur in a wide variety of scenarios, such as in sports or in the health sector. 
-As modern smartphones become more and more powerful, classical approaches to pattern recognition can now be adapted directly. 
-Nevertheless, in context of this work 
-%
-%
-Many different activity recognition approaches
-%
-a aufwendige trainingsphase, as they are ... based on classical vorgehensweise von pattern recognition
-%
-In contrast to raw barometer data... more stable ...}
-
-
+As modern smartphones become more and more powerful, classical approaches of pattern recognition can now be adapted directly. 
+Nevertheless, in context of detecting activities in indoor environments, such approaches might be to much of a good thing. 
+For example, \cite{moder15} provide very promising results, but their approach requires an extensive training-phase using a set of previously recorded training data.
+Acquiring such data can not only be time-consuming, but often opens the need for a high diversity, to model multiple different movement patterns and thus prevent an overadaption of the classification to a small set of testers.  
+Because of this, we present a threshold-based activity recognition, that can cover a very general setting without much prior knowledge. 
+This is a very straightforward approach and can be found especially in early literature on activity detection using wearable sensors \cite{lee2002activity, sekine2000classification, veltink1996detection}. 
+In contrast to recent state-of-the-art methods, which try to incorporate many different activities, we are only interested in finding four activities, namely standing, walking, walking up and down. 
+This limitation allows to consider the present scenario in the best possible way. 
+It should be noted, that our approach cannot necessarily keep up with the more advanced methods mentioned above, but it shows suitable results in the context of indoor localization.}
 
 %eval - wifi, fingerprinting
-The outcomes of the state evaluation process depend highly on the used sensors. 
-Most smartphone-based systems are using received signal strength indications (RSSI) given by \docWIFI{} or Bluetooth as a source for absolute positioning information. 
+Most smartphone-based systems for indoor localization are using received signal strength indications (RSSI) given by \docWIFI{} or Bluetooth as a source for absolute positioning information. 
 At this, one can mainly distinguish between fingerprinting and signal strength prediction model based solutions \cite{Ebner-17}.
 Indoor localization using \docWIFI{} fingerprints was first addressed by \cite{radar}. 
 During a one-time offline-phase, a multitude of reference measurements are conducted. 

tex_review/chapters/transition.tex

@@ -104,11 +104,16 @@
 		It thus is always possible to walk from one polygon into another,
 		if they are adjacent.
 		Similar to the graph-based approach, adjacent polygons 
-		denote some sort of walkable surface.
-		Just as before, the navigation mesh can be \emph{automatically}
+		denote some sort of walkable surface.}
+		\addy{However, while a graph restricts the movement to edges and nodes, the mesh allows for a 
+		true continues movement. 
+		This is achieved by having the freedom to walk to any position, under the condition that it 
+		resides within a polygon.}
+		\add{Just as before, the navigation mesh can be \emph{automatically}
 		generated from the building's floor plan, based on
-		various algorithms \cite{navMeshAlg1}.
-	}
+		various algorithms \cite{navMeshAlg1, kallmann2010navigation}.
+		}
+
 	Using variably shaped/sized elements instead of rigid grid-cells
 	provides both, higher accuracy for reaching every corner, and a reduced
 	memory footprint as a single polygon is able to cover arbitrarily

tex_review/egbib.bib

@@ -3527,4 +3527,140 @@ number={},
 pages={1-8},
 IGNOREmonth={Oct},}
 
+@article{xia2015using,
+  title={{Using Multiple Barometers to Detect the Floor Location of Smart Phones with built-in Barometric Sensors for Indoor Positioning}},
+  author={Xia, Hao and Wang, Xiaogang and Qiao, Yanyou and Jian, Jun and Chang, Yuanfei},
+  journal={Sensors},
+  volume={15},
+  number={4},
+  pages={7857--7877},
+  year={2015},
+  publisher={Multidisciplinary Digital Publishing Institute}
+}
+
+@inproceedings{li2013using,
+  title={{Using Barometers to Determine the Height for Indoor Positioning}},
+  author={Li, Binghao and Harvey, Bruce and Gallagher, Thomas},
+  booktitle={Indoor Positioning and Indoor Navigation (IPIN), 2013 International Conference on},
+  pages={1--7},
+  year={2013},
+  organization={IEEE}
+}
+
+@article{zheng20163d,
+  title={{A 3D Indoor Positioning System based on Low-Cost MEMS Sensors}},
+  author={Zheng, Lingxiang and Zhou, Wencheng and Tang, Weiwei and Zheng, Xianchao and Peng, Ao and Zheng, Huiru},
+  journal={Simulation Modelling Practice and Theory},
+  volume={65},
+  pages={45--56},
+  year={2016},
+  publisher={Elsevier}
+}
+
+
+@inproceedings{moder20143d,
+  title={{3D Indoor Positioning with Pedestrian Dead Reckoning and Activity Recognition based on Bayes Filtering}},
+  author={Moder, Thomas and Hafner, Petra and Wisiol, Karin and Wieser, Manfred},
+  booktitle={Indoor positioning and indoor navigation (IPIN), 2014 international conference on},
+  pages={717--720},
+  year={2014},
+  organization={IEEE}
+}
+
+@article{tian2015smartphone,
+  title={{Smartphone-Based Indoor Integrated WiFi/MEMS Positioning Algorithm in a Multi-Floor Environment}},
+  author={Tian, Zengshan and Fang, Xin and Zhou, Mu and Li, Lingxia},
+  journal={Micromachines},
+  volume={6},
+  number={3},
+  pages={347--363},
+  year={2015},
+  publisher={Multidisciplinary Digital Publishing Institute}
+}
+
+@INPROCEEDINGS{moder15,
+author={T. Moder and K. Wisiol and P. Hafner and M. Wieser},
+booktitle={2015 International Conference on Indoor Positioning and Indoor Navigation (IPIN)},
+title={Smartphone-Based Indoor Positioning Utilizing Motion Recognition},
+year={2015},
+volume={},
+number={},
+pages={1-8},
+ISSN={},
+IGNOREmonth={Oct},}
+
+@article{shoaib2015survey,
+  title={{A Survey of Online Activity Recognition using Mobile Phones}},
+  author={Shoaib, Muhammad and Bosch, Stephan and Incel, Ozlem and Scholten, Hans and Havinga, Paul},
+  journal={Sensors},
+  volume={15},
+  number={1},
+  pages={2059--2085},
+  year={2015},
+  publisher={Multidisciplinary Digital Publishing Institute}
+}
+
+@article{lee2002activity,
+  title={{Activity and Location Recognition using Wearable Sensors}},
+  author={Lee, Seon-Woo and Mase, Kenji},
+  journal={IEEE pervasive computing},
+  volume={1},
+  number={3},
+  pages={24--32},
+  year={2002},
+  publisher={IEEE}
+}
+
+@inproceedings{lee2001recognition,
+  title={{Recognition of Walking Behaviors for Pedestrian Navigation}},
+  author={Lee, Seon-Woo and Mase, Kenji},
+  booktitle={Control Applications, 2001.(CCA'01). Proceedings of the 2001 IEEE International Conference on},
+  pages={1152--1155},
+  year={2001},
+  organization={IEEE}
+}
+
+@article{sekine2000classification,
+  title={{Classification of Waist-Acceleration Signals in a Continuous Walking Record}},
+  author={Sekine, Masaki and Tamura, Toshiyo and Togawa, Tatsuo and Fukui, Yasuhiro},
+  journal={Medical engineering \& physics},
+  volume={22},
+  number={4},
+  pages={285--291},
+  year={2000},
+  publisher={Elsevier}
+}
+
+@article{veltink1996detection,
+  title={{Detection of Static and Dynamic Activities using Uniaxial Accelerometers}},
+  author={Veltink, Peter H and Bussmann, HansB J and De Vries, Wiebe and Martens, WimL J and Van Lummel, Rob C},
+  journal={IEEE Transactions on Rehabilitation Engineering},
+  volume={4},
+  number={4},
+  pages={375--385},
+  year={1996},
+  publisher={IEEE}
+}
+
+@inproceedings{kallmann2010navigation,
+  title={{Navigation Queries from Triangular Meshes}},
+  author={Kallmann, Marcelo},
+  booktitle={International Conference on Motion in Games},
+  pages={230--241},
+  year={2010},
+  organization={Springer}
+}
+
+@article{vatti1992generic,
+  title={{A Generic Solution to Polygon Clipping}},
+  author={Vatti, Bala R},
+  journal={Communications of the ACM},
+  volume={35},
+  number={7},
+  pages={56--63},
+  year={1992},
+  publisher={ACM}
+}
+
+
 

tex_review/misc/functions.tex

@@ -155,7 +155,7 @@
 %comments for sensors journal
 \newcommand{\del}[1]{\textcolor{red}{\hcancel{#1}}}
 \newcommand{\add}[1]{\textcolor{blue}{#1}}
-\newcommand{\addy}[1]{\textcolor{purple}{#1}}
+\newcommand{\addy}[1]{\textcolor{teal}{#1}}