vault backup: 2024-10-02 23:25:27

.obsidian/workspace.json

@@ -8,31 +8,53 @@
         "type": "tabs",
         "children": [
           {
-            "id": "1391874bd127fece",
+            "id": "fce83e74db3de5b7",
             "type": "leaf",
             "state": {
-              "type": "markdown",
+              "type": "pdf",
               "state": {
-                "file": "Foundation of data science/slides/Untitled.md",
+                "file": "Biometric Systems/slides/LEZIONE2_Indici_di_prestazione.pdf",
-                "mode": "source",
+                "page": 13,
-                "source": false
+                "left": -26,
+                "top": 2,
+                "zoom": 0.5666666666666667
               }
             }
           },
           {
-            "id": "160ca1e7ba4d7af7",
+            "id": "24553d742ee49625",
             "type": "leaf",
             "state": {
-              "type": "markdown",
+              "type": "pdf",
               "state": {
-                "file": "conflict-files-obsidian-git.md",
+                "file": "Autonomous Networking/slides/3 WSN.pdf",
-                "mode": "source",
+                "page": 46,
-                "source": false
+                "left": -26,
+                "top": 382,
+                "zoom": 0.5666666666666667
               }
             }
           }
         ],
         "currentTab": 1
+      },
+      {
+        "id": "730530e270323c7d",
+        "type": "tabs",
+        "children": [
+          {
+            "id": "06d416b60db86f57",
+            "type": "leaf",
+            "state": {
+              "type": "markdown",
+              "state": {
+                "file": "Autonomous Networking/notes/3 WSN.md",
+                "mode": "source",
+                "source": false
+              }
+            }
+          }
+        ]
       }
     ],
     "direction": "vertical"
@@ -82,7 +104,8 @@
       }
     ],
     "direction": "horizontal",
-    "width": 300
+    "width": 300,
+    "collapsed": true
   },
   "right": {
     "id": "11560c155f3d8f6e",
@@ -98,7 +121,7 @@
             "state": {
               "type": "backlink",
               "state": {
-                "file": "conflict-files-obsidian-git.md",
+                "file": "Autonomous Networking/slides/3 WSN.pdf",
                 "collapseAll": false,
                 "extraContext": false,
                 "sortOrder": "alphabetical",
@@ -115,7 +138,7 @@
             "state": {
               "type": "outgoing-link",
               "state": {
-                "file": "conflict-files-obsidian-git.md",
+                "file": "Autonomous Networking/slides/3 WSN.pdf",
                 "linksCollapsed": false,
                 "unlinkedCollapsed": true
               }
@@ -138,7 +161,7 @@
             "state": {
               "type": "outline",
               "state": {
-                "file": "conflict-files-obsidian-git.md"
+                "file": "Autonomous Networking/slides/3 WSN.pdf"
               }
             }
           },
@@ -172,8 +195,22 @@
       "obsidian-git:Open Git source control": false
     }
   },
-  "active": "160ca1e7ba4d7af7",
+  "active": "24553d742ee49625",
   "lastOpenFiles": [
+    "Autonomous Networking/notes/3 WSN.md",
+    "Autonomous Networking/slides/3 WSN.pdf",
+    "Biometric Systems/final notes/Untitled.md",
+    "Biometric Systems/slides/LEZIONE2_Indici_di_prestazione.pdf",
+    "Biometric Systems/final notes/1. Introduction.md",
+    "[LEZIONE2_Indici_di_prestazione.pdf.md",
+    "[[[LEZIONE2_Indici_di_prestazione.pdf.md",
+    "().md",
+    "a.md",
+    "Pasted image 20241002181936.png",
+    "Pasted image 20241002181932.png",
+    "Pasted image 20241002135922.png",
+    "Pasted image 20241002114133.png",
+    "Pasted image 20241002114048.png",
     "Foundation of data science/slides/Untitled.md",
     "conflict-files-obsidian-git.md",
     "Autonomous Networking/slides/2 RFID.pdf",
@@ -183,23 +220,15 @@
     "Autonomous Networking/images/Pasted image 20240928193208.png",
     "Autonomous Networking/images/Pasted image 20240928191316.png",
     "Autonomous Networking/images/Pasted image 20240928183123.png",
-    "Autonomous Networking/images/Pasted image 20240928181424.png",
-    "Autonomous Networking/images/Pasted image 20240928181304.png",
-    "Autonomous Networking/images/Pasted image 20240928175441.png",
-    "Autonomous Networking/images/Pasted image 20240928165943.png",
-    "Autonomous Networking/images/Pasted image 20240928165439.png",
     "Autonomous Networking/slides",
     "Autonomous Networking/notes",
     "Autonomous Networking/images",
     "Autonomous Networking",
     "Foundation of data science/slides/notes 2.md",
     "Foundation of data science/slides/FDS_intro_new.pdf",
-    "Biometric Systems/final notes/1. Introduction.md",
     "Foundation of data science/slides",
     "Foundation of data science",
-    "LICENSE",
     "Biometric Systems/slides/lezione1 notes.md",
-    "Biometric Systems/slides/LEZIONE2_Indici_di_prestazione.pdf",
     "Untitled.canvas",
     "Untitled.md"
   ]

Autonomous Networking/notes/3 WSN.md

@@ -0,0 +1,113 @@
+The main difference between an RFID network and a WSN is that nodes:
+- are battery powered
+- can sense the environment
+- can listen to the channel (carrier sense) and transmit spontaneously
+- can make more complex computation
+- can send packets to other nodes (e.g. for multi-hop communication)
+
+#### Roles of partecipants in WSN
+- Sources of data: measure data, report them somewhere
+- Sinks of data: interested in receiving data from WSN
+- Actors/actuators: control some devices based on data
+
+#### Deployiment options
+- Random deployiment
+	- dropped from an aircraft
+	- usually uniform random distribution for nodes over finite area is assumed
+- Regular deployment
+	- wel planned, fixed
+	- not necessarily geometric structure, but that is often a convenient assumption
+- Mobile sensor nodes
+	- Can move to compensate for deployment shortcomings
+	- Can be passively moved by some external force (wind, water)
+	- Can actively seek out "interesting" areas
+#### Characteristics of WSN
+- Scalability
+	- they need to support **large number of nodes**
+	- performance should not degrade with increasing number of nodes
+- Wide range of densities (very application dependent)
+- Limited resources for each device
+	- low amount of energy
+	- low cost, size and weight
+	- nodes may not have a global ID (e.g. an IP)
+- Mostly static topology
+
+- Service in WSN (not simply moving bits like traditional networks)
+	- in-network processing
+	- provide answers
+	- comunication is triggered by events
+	- asymmetric flow of information (from sensors to sink)
+- QoS
+	- traditional metrics do not apply
+- Fault tollerance
+	- be robust against node failure
+	- running out of energy, physical destruct
+- Lifetime
+	- the network should fulfill as long as possible
+	- lifetime of individual nodes relatively unimportant
+	- but if a critical node dies, the network dies
+- Programmability
+	- being able to re-program nodes on-field, to improve flexibility
+- Maintainability
+	- WSN has to adapt to changes
+
+#### Typical Adopted Mechanisms
+- Multi-hop wireless communication
+- Energy-efficient operation (both for computation, sensing, actuation)
+- Self-configuration
+- Collaboration & in-network processing
+	- the nodes in the network collaborate towards a joint goal
+	- pre-processing the data before sending it to the sink, to improve efficiency
+
+#### Mechanism to meet requirements
+- Data centric networking
+	- focussing network design on data, not on node identifiers
+- Locality
+	- do things locally as far as possible
+- Exploit tradeoffs
+	- e.g between invested energy and accuracy
+
+> [!PDF|yellow] [[3 WSN.pdf#page=29&color=yellow|3 WSN, p.29]]
+> > WSN: reasoning of existence
+> 
+> collect, couple, provide, establish
+#### Main sensor node components
+- antenna and RF transceiver
+- memory unit
+- CPU
+- sensor unit (i.e. thermostat)
+- power source (typ. battery)
+- operating system
+	- TinyOS
+
+sensing, processing and networking is all done by the sensor node.
+
+#### WSN vs conventional networks
+
+| **Conventional networks**                                           | **WSN**                                                   |
+| ------------------------------------------------------------------- | --------------------------------------------------------- |
+| general purpose design                                              | serving a single application or a bouquet of applications |
+| network performance and latency                                     | energy is the primary challenge                           |
+| devices and networks operate in controlled / mild environments      | unattended, harsh conditions & hostile environments       |
+| global knowledge is feasible and centralized management is possible | localized decisions - no support by central entity        |
+#### Wireless signal issues
+- **Attenuation**: the strength of the signal decreases rapidly over distance
+- **Multi-path propagation**:
+	- when a radio wave encounter an obstacle, all or part of the wave is reflected, with a loss of power
+	- a source signal can arrive, to successive reflections, to reach a station through multiple paths
+- **Interference:**
+	- from the same source (multi-path propagation): signal arrives multiple time
+	- from multiple sources: more stations transmit simultaneously
+
+We use **SNR** to measure the ratio of good to bad signal (signal to noise). Higher is better.
+
+> [!PDF|yellow] [[3 WSN.pdf#page=49&selection=77,0,77,15&color=yellow|3 WSN, p.49]]
+> > Synchronization
+> 
+> nodes have clocks but they may not be synchronized!
+
+To address these issues, we use MAC protocols. We need a protocol suitable for wireless networks, which emphasize energy-efficient operation.
+### CSMA/CA
+![[Pasted image 20241002114133.png]]
+
+IFS is random, so hopefully only a node starts transmitting at the same time.

Biometric Systems/final notes/Untitled.md

@@ -0,0 +1,99 @@
+#### Problems of biometric systems:
+- wide intra-class variations
+	- maybe different facial expression, different light, different view point...
+- very small inter-class variations
+	- two different person very similar (i.e. twins)
+
+- possible spoofing attacks, in different moments
+	![[Pasted image 20241002181936.png]]
+
+- [non universality](LEZIONE2_Indici_di_prestazione.pdf#page=6&selection=0,10,0,26&color=yellow|LEZIONE2_Indici_di_prestazione, p.6)
+	- e.g. people with no voice, people with cataract, people with poor fingerprints...
+
+Most difficult traits to exploit:
+- retina fundus
+- behavioral traits (i.e. way of walking)
+- handwriting
+
+### [What to compare?](LEZIONE2_Indici_di_prestazione.pdf#page=8&selection=0,10,0,26&color=yellow|LEZIONE2_Indici_di_prestazione, p.8)
+- **Sample
+	- raw captured data
+- **Hand-crafted features**
+	- manually engineered by the data scientist and extracted from samples
+	- can also be substituted with **embeddings**: features automatically extracted by deep architectures
+- **Template**
+	- collection of features extracted from the row data, examples:
+		- a histogram representing the frequencies of relevant values in the image (e.g. greylevel values)
+		- a vector of values each representing a relevant measure (e.g. Bertillon measures)
+		- time series of acceleration values (one per axis)
+		- a set of triplets as for relevant fingerprint points representing the coordinates of the points and the direction of the tangent to the ridge in that point.
+
+
+> [!PDF|red] [[LEZIONE2_Indici_di_prestazione.pdf#page=8&selection=11,1,14,16&color=red|LEZIONE2_Indici_di_prestazione, p.8]]
+> > Hand-crafted features
+> 
+> not the template of the entire biometric system.
+
+
+### Comparing templates
+- Euclidian distance
+- Cosine similarity
+	- cosine of the angle between two vectors
+- Pearson correlation
+- Bhattacharyya distance
+
+
+> [!PDF|yellow] [[LEZIONE2_Indici_di_prestazione.pdf#page=9&selection=8,0,10,31&color=yellow|LEZIONE2_Indici_di_prestazione, p.9]]
+> > or cosine similarity may provide either a distance measure or a similarity measure
+> 
+> shows "more stuff" than Euclidian distance, such as orientation ecc.. Shows how templates are similar to eachother. While distance shows how templates are... distant!
+
+> [!PDF|yellow] [[LEZIONE2_Indici_di_prestazione.pdf#page=10&selection=3,1,4,21&color=yellow|LEZIONE2_Indici_di_prestazione, p.10]]
+> > (Pearson) Correlation
+> 
+> how signals are similar to eachother. Often used to compare fingerprints, by computing the correlation between two fingerprints.
+
+Histograms needs other ways to be compared.
+
+The same happens with time series: speed for example may speed the final outcome of the time series, even if the patterns are the same.
+
+So what do we do?
+sometimes we use correlation, but Dynamic time Warping is the most used.
+
+![[Pasted image 20241002135922.png]]
+
+each point is paired with the most convenient one. It's not necessaty that points corresponds to the same instant in time.
+
+if using deep learning we should use the architecture to extract the embeddings (for both gallery and probe templates).
+
+//
+after normalization in range [0, 1] we will have that distance = 1 - similarity.
+
+#### Possible errors: verification
+- Genuine Match (GM, GA): the claimed identity is true and subject is accepted
+- False Rejection (FR, FNM, type I error): claimed identity is true but the subjet is rejected
+- Genuine Reject (GR, GNM): an impostor is rejected
+- False Acceptance (FA, FM, type II error): an impostor is accepted :/
+
+It's important to define a good threshold.
+If too high we will get a lot of false acceptance. If too low we will get a lot of false rejection!
+
+When computing rates:
+- False Rejection Rate (FRR) is the number of FR divided by ONLY the number of GM+FR.
+	- in fact, GM + FR have the same denominator and sum up to 1.
+- False Acceptance Rate is the number of FA divided by FA + GR
+
+Equal Error Rate is the value at a specific threshold, where FAR and FRR are the same value.
+
+two synthetic metrics could be ERR and area below ROC curve.
+
+(we might have more templates for the same person to address inter-class variation.
+Of course templates should be different, not computed i.e. by frames of the same video, as some of them could be blurred and close frames are exactly the same!)
+
+> [!PDF|yellow] [[LEZIONE2_Indici_di_prestazione.pdf#page=20&selection=119,0,119,4&color=yellow|LEZIONE2_Indici_di_prestazione, p.20]]
+> > When
+> 
+> in false acceptance we can have two possible scenarios
+> - pj does not belong to the gallery (most trivial)
+> - pj belongs to an enrolled subject but the probe claimed another identity, not the real one.
+