SlideShare a Scribd company logo

Infrastructureless Wireless networks

Gwendal Simon, profile picture
Gwendal Simon

The document discusses infrastructure-less wireless networks. It provides an overview of relevant literature, motivations for infrastructure-less networks when infrastructure is unavailable or limited, and examples of multi-hop wireless network applications including ad-hoc and sensor networks. It also introduces popular models for representing wireless networks as graphs and discusses issues like broadcasting, mobility models, data gathering, time synchronization, and routing protocols.

TechnologyBusiness
Related topics:
Wireless Sensor Networks
1 of 93
Downloaded 352 times
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
Infrastructure-less Wireless Networks Gwendal Simon Department of Computer Science Institut Telecom 2009
Literature Books include: “Algorithms for sensor and ad hoc networks”, D. Wagner and R. Wattenhofer “Wireless sensor networks: an information processing approach”, F. Zhao and L. Guibas and journal/conferences include: ACM SigMobile (MobiHoc, SenSys, etc.) IEEE MASS and WCNC Elsevier Ad-Hoc Network, Wireless Networks 2 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Motivations Current wireless net. require an infrastructure: cellular network: interconnected base stations wifi Internet: an access point and Internet 3 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Motivations Current wireless net. require an infrastructure: cellular network: interconnected base stations wifi Internet: an access point and Internet Same flaws than centralized architectures: cost scalability privacy dependability 3 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Motivations Sometimes, there is no infrastructure transient meeting disaster areas military interventions alter-communication 4 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Motivations Sometimes, there is no infrastructure transient meeting disaster areas military interventions alter-communication Sometimes not every station hear every other station limited wireless transmission range large-scale area 4 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Multi-hop Wireless Networks Nodes: portable wireless devices transmission ranges do not cover the area density ensures network connectivity Links: wireless characteristics transmission model: local broadcasting energy consumption: transmission is costly Behavior: devices emit, receive and forward data 5 / 41 Gwendal Simon Infrastructure-less Wireless Networks
A Taxonomy of Applications 6 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Ad-Hoc vs. Sensor Networks Ad-Hoc Networks Sensor Networks nodes powerful wifi devices tiny zigbee nodes algorithms all-to-all routing echo to sink mobility human or car motions failures performance criteria quality of service energy consumption 7 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Ad-Hoc Applications Delay-Tolerant Network (social media application) assumption: no connectivity, but high mobility objective: ensuring eventual message delivery 8 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Ad-Hoc Applications Delay-Tolerant Network (social media application) assumption: no connectivity, but high mobility objective: ensuring eventual message delivery Mesh Networks (rural wireless coverage) assumption: some nodes have Internet access objective: maintaining path to these nodes 8 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Ad-Hoc Applications Delay-Tolerant Network (social media application) assumption: no connectivity, but high mobility objective: ensuring eventual message delivery Mesh Networks (rural wireless coverage) assumption: some nodes have Internet access objective: maintaining path to these nodes Vehicular Ad-Hoc Networks assumption: a particular mobility model objective: mostly services related to car safety 8 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Sensor Network Applications Sink-Based Networks (monitoring of natural areas) assumption: one sink retrieves all sensed data objective: increasing life-time 9 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Sensor Network Applications Sink-Based Networks (monitoring of natural areas) assumption: one sink retrieves all sensed data objective: increasing life-time Mobile Object Tracking (area surveillance) assumption: sensors know their location objective: determining hostile position 9 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Sensor Network Applications Sink-Based Networks (monitoring of natural areas) assumption: one sink retrieves all sensed data objective: increasing life-time Mobile Object Tracking (area surveillance) assumption: sensors know their location objective: determining hostile position Multi-Sink Networks (intervention teams) assumptions: mobile sinks and fixed sensor objectives: increasing sink coverage 9 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Short Introduction to Popular Models 10 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Network as a Graph Unit-Disk Graph: 05 03 → node position 07 09 → circular transmission 12 11 06 10 01 → boolean connections 02 00 08 04 11 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Network as a Graph Unit-Disk Graph: 05 03 → node position 07 09 → circular transmission 12 11 06 10 01 → boolean connections 02 00 08 04 11 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Network as a Graph Unit-Disk Graph: 05 03 → node position 07 09 → circular transmission 12 11 06 10 01 → boolean connections 02 00 08 04 11 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Interferences Signal-to-noise-plus-interference (SINR) ratio Pu d(u,v )α Pw ≥β N+ w ∈V {u} d(w ,v )α Pu : power level of sender u d(u, v ): distance between u and v α: path-loss exponent N: noise β: minimum ratio 12 / 41 Gwendal Simon Infrastructure-less Wireless Networks
A Tour of the Most Studied Issues 13 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Broadcasting I: Stormy Effect Broadcast: a simple basic problem : a source emits a message all nodes within the network eventually receive the message a simple and efficient solution: upon first reception of message, forward it. Limits of flooding in wireless networks: redundant messages interferences 14 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Broadcasting II: Proposals Probabilistic flooding: idea: forward the message with some probability p drawbacks: no guarantee of delivering refinements: adjust p to node density 15 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Broadcasting II: Proposals Probabilistic flooding: idea: forward the message with some probability p drawbacks: no guarantee of delivering refinements: adjust p to node density Constrained flooding: idea: only some nodes forward the message implementation: build the Minimum Connected Dominating Set drawbacks: maintaining cost in dynamic systems 15 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Mobility Models I Few theoretical proof, few real implementations ⇒ generate realistic node motions for simulations 16 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Mobility Models I Few theoretical proof, few real implementations ⇒ generate realistic node motions for simulations The simplest model: Random Waypoint 1. each node picks a random position uniformly 2. it travels toward this destination with a speed v 3. once it reaches it, it stops during few seconds 4. back to 1 16 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Mobility Models II: Improvements Basic Structural Flaws: non-uniform distribution of node location: higher node distribution in the center average speed decay: low speed nodes spend more time to travel Realistic Mobility Models: group movement area popularity urban models community-based 17 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Mobility Models II: Improvements Basic Structural Flaws: non-uniform distribution of node location: higher node distribution in the center average speed decay: low speed nodes spend more time to travel Realistic Mobility Models: group movement area popularity urban models community-based the most realistic one : using real traces! 17 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Localized Data Gathering Basic idea: query data from sensors within an area two rounds: query diffusion retrieve data from sensors main objectives: minimize energy consumption minimize the delay A problem related with broadcasting except: only sensors from the queried area are reached: complex queries are possible (average, max, etc.) 18 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Time Synchronization I Different time on nodes: different oscillator frequency ⇒ frequency error absolute difference between clocks ⇒ phase error The need of a common clock localization protocols some MAC protocols data fusion in sensor network 19 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Time Synchronization II Broadcasting standard time via GPS system: √ precision, simple implementation × expensive devices × limited usage (outdoor environment) Achieve a common time distributively: √ (almost) no special devices required √ more tolerant to the environment × special protocols × message overhead, multi-hop delays 20 / 41 Gwendal Simon Infrastructure-less Wireless Networks
A Focus on Routing Protocols 21 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Routing Protocols Objective: select a path between a source and a destination Main design challenges: unstable network topology low-cost devices (energy, computing. . . ) Main routing mechanisms: neighbor discovering route setup route maintenance 22 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Proactive routing vs. On demand routing Proactive Reactive Setup all-to-all on demand Maintenance regularly during utilization Advantages no setup delay no unused routes Disadvantages fixed overhead long setup delay Main examples OLSR AODV 23 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery G F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery Broadcasting RREQ Mes- G sage. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery Replying RREP to G source. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Discovery Forward path G setup. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Maintenance Link breaks between B G and D. F H S B E D A 25 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Maintenance Sending RERR mes- G sage. F H S B E D A 25 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Maintenance Restarting route discov- G ery. F H S B E D A 25 / 41 Gwendal Simon Infrastructure-less Wireless Networks
AODV Route Maintenance New route G discovered. F H S B E D A 25 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Some Tricks Intelligent flooding (detect close destination) idea: init TTL at 1, then 2, then 3. . . idea: flood slowly and send message to stop it 26 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Some Tricks Intelligent flooding (detect close destination) idea: init TTL at 1, then 2, then 3. . . idea: flood slowly and send message to stop it Route caching (use past flooding) idea: during flood, answer for a distant node drawback: contradict reactive routing philosophy 26 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Some Tricks Intelligent flooding (detect close destination) idea: init TTL at 1, then 2, then 3. . . idea: flood slowly and send message to stop it Route caching (use past flooding) idea: during flood, answer for a distant node drawback: contradict reactive routing philosophy Local maintenance (almost unchanged route) idea: instead of NAK s, look for d by yourself drawback: sometimes it does not work 26 / 41 Gwendal Simon Infrastructure-less Wireless Networks
A Proactive Routing Protocol: OLSR Objective: make use of Multi-Point Relay (MPR) acting as super-peers easing topology discovery handling most of the traffic OLSR message types: HELLO: discover 1-hop and 2-hop neighbors topology discovery through MPR 27 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Neighbor sensing F G H S B D E A 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Neighbor sensing F G Nb:{S}, H 2hopNb:{} Broadcasting S B HELLO Message. Nb:{S}, 2hopNb:{} D E A Nb:{S}, 2hopNb:{} 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Neighbor sensing F G H S B Nb:{E}, Nb:{S,E}, 2hopNb:{} 2hopNb:{} D E A Nb:{E}, 2hopNb:{S} 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Neighbor sensing F G Nb:{F}, H 2hop Nb:{S} S B Nb:{E,F}, Nb:{S,E,F}, 2hop Nb:{} 2hop Nb:{} D E A 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Neighbor sensing F G Nb:{S,B,G}, Nb:{F,B,H}, H 2hopNb:{E,A,H} 2hopNb:{S,E,A,D} Nb:{G,D}, 2hopNb:{B,F,A} S B Nb:{E,F,B}, Nb:{S,E,F,A,G}, 2hopNb:{G,A} 2hopNb:{D,H} D E A Nb:{A,H}, Nb:{S,A,B}, Nb:{E,B,D}, 2hopNb:{B,E,G} 2hopNb:{F,G,D} 2hopNb:{S,F,G,H} 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
MPR selection F G Nb:{S,B,G}, H 2hopNb:{E,A,H} S B Nb:{E,F,B}, Nb:{S,E,F,A,G}, 2hopNb:{G,A} 2hopNb:{D,H} D E A Nb:{S,A,B}, 2hopNb:{F,G,D} 29 / 41 Gwendal Simon Infrastructure-less Wireless Networks
MPR selection F G HELLO message H indicating B as S B MPR of S and B note S as its MPR selector. D E A 29 / 41 Gwendal Simon Infrastructure-less Wireless Networks
MPR selection F G MPR Selector: MPR Selector: H {} {B,F,H} MPR Selector: {G,D} S B MPR Selector: MPR Selector: {} {S,G,E,F,A} D E A MPR Selector: MPR Selector: MPR Selector: {A,H} {} {B,E,D} 29 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Topology Table Each node maintains a Topology Table containing all possible destinations notifying a MPR to reach them Structure of Topology Table (on S for example): Dest Addr Last Hop Seq Holding Time G B 1 10 A B 4 20 D A 6 10 H G 5 15 ... ... ... ... 30 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Building the Topology Table F G MPR Selector: H {B,F,H} S B D E A 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Building the Topology Table F G H S B Topology Table Des Lhop Seq Htime D F E G 2 30 A H G 2 30 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Building the Topology Table F G H MPR Selector: {G,D} S B MPR Selector: {S,G,E,F,A} D E A 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Building the Topology Table F G H Topology Table Des Lhop Seq Htime S F G B2 30 H G 2 30 B G 2 30 D E A 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Building the Topology Table F G H Broadcasting contin- S B ues. . . D E A MPR Selector: MPR Selector: {A,H} {B,E,D} 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Building the Topology Table F G MPR Selector: MPR Selector: H {} {B,F,H} MPR Selector: {G,D} S B MPR Selector: MPR Selector: {} {S,G,E,F,A} D E A MPR Selector: MPR Selector: MPR Selector: {A,H} {} {B,E,D} 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 E A 6 30 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 E A 6 30 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 E A 6 30 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 A B 2 E A 6 30 G B 2 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 A B 2 E A 6 30 G B 2 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 A B 2 E A 6 30 G B 2 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 A B 2 E A 6 30 G B 2 D A 6 30 H B 3 A D 7 30 D B 3 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
Any Hybrid Approach ? Merging advantages from both approaches: build a routing table at 4 ∼ 5 hops launch a reactive process if d is not Applicative concerns: OLSR is attractive because networks often small AODV scales well but no all-to-all routing 33 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Research Activity: Multi-Sinks Query Range 34 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable 50 0 50 100 150 200 250 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable 50 0 50 100 150 200 250 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable 50 0 50 100 150 200 250 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable 50 0 50 100 150 200 250 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable Networking assumptions: 50 IEEE 802.15.4 MAC layer, ZigBee tree routing No in-network data aggregation, compression 0 Static sensors100 50 and sinks 150 (may extend to mobile 250 200 sinks) 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 u1 = 5 S2 50 u2 = 1 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 (4, 2) no u1 = 4 S2 50 u2 = 2 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 (4, 2) no (3, 2) yes u1 = 3 S2 50 u2 = 2 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 (4, 2) no (3, 2) yes u1 = 2 (2, 3) yes S2 50 u2 = 3 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 (4, 2) no (3, 2) yes u1 = 2 (2, 3) yes S2 50 u2 = 3 62 configurations 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Which Configuration is Better? 200 Basic considerations: (4, 2): not feasible, (1, 1): inefficient 150 100 50 0 50 100 150 200 250 300 37 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
Which Configuration is Better? 200 Basic considerations: (4, 2): not feasible, (1, 1): inefficient 150 Optimality criteria: 1 Maximum Impact Range 100 (5, 1): Sum up to 6 Max-Min Fairness 4 (2, 3) = (3, 2) (5, 1) 3 2 50 (?, ?, ?, ?) 0 50 100 150 200 250 300 37 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
Problem Formulation Multi-Dimensional Multiple Choice Knapsack Problem a NP-complete problem 38 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Problem Formulation Multi-Dimensional Multiple Choice Knapsack Problem a NP-complete problem Toward a distributed heuristic algorithm only local views of the network only local optimal solutions 38 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Protocol 200 At each sink: enlarge requirement periodically receive notification from sensors 150 adjust requirement if it is smaller At each sensor: 3 100 measure the traffic detect congestion A 50 4 2 1 solve the local problem notify related sinks 0 50 100 150 200 250 300 39 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
Conclusion 40 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Personal Thoughts Great theoretical importance: a lot of new and scientifically exciting problems a multi-disciplinary field (network, algorithms, computational geometry, probabilities) 41 / 41 Gwendal Simon Infrastructure-less Wireless Networks
Personal Thoughts Great theoretical importance: a lot of new and scientifically exciting problems a multi-disciplinary field (network, algorithms, computational geometry, probabilities) Unsure applicative importance: no killer application yet cellular networks just do what we want 41 / 41 Gwendal Simon Infrastructure-less Wireless Networks

More Related Content

PPTX
Iot architecture
Niranjan Kumar
 
PDF
IoT Networking
Hitesh Mohapatra
 
PPTX
Home automation using IoT
Athira_1993
 
PDF
IoT Communication Protocols
Pradeep Kumar TS
 
PDF
Internet of Things- Applications
Ravindra Dastikop
 
PPTX
6-IoT protocol.pptx
Pratik Gohel
 
PPTX
IoT Based Home Automation System Presantation
Farhan Ahmed Rahee
 
PPTX
5G applications
Udara Sandaruwan
 
Iot architecture
Niranjan Kumar
 
IoT Networking
Hitesh Mohapatra
 
Home automation using IoT
Athira_1993
 
IoT Communication Protocols
Pradeep Kumar TS
 
Internet of Things- Applications
Ravindra Dastikop
 
6-IoT protocol.pptx
Pratik Gohel
 
IoT Based Home Automation System Presantation
Farhan Ahmed Rahee
 
5G applications
Udara Sandaruwan
 

What's hot (20)

PPTX
Internet of Things (IoT)
Prakhyath Rai
 
PPTX
Iot ppt
piyuu7
 
PPTX
Internet of Things (IoT) and its applications
Sarwan Singh
 
PDF
The essential role of AI in the 5G future
Qualcomm Research
 
PDF
Advanced: Private Networks & 5G Non-Public Networks
3G4G
 
PPTX
10 min IoT ppt
Vaishnavu Pathayathodi
 
PDF
IoT and m2m
pavan penugonda
 
PPTX
An iot based secured smart e-campus
shivaraj kumar gowdas
 
PDF
IoT material revised edition
pavan penugonda
 
PPTX
Web servers for the Internet of Things
Alexandru Radovici
 
PDF
5G Multi-Access Edge Compute
Michelle Holley
 
PDF
Understanding IoT
Md. Shamsul Haque
 
PPTX
Internet of Things (IoT) - Introduction ppt
sutrishnakar1995
 
PPTX
Internet of things
Sandisiwe Sixanda
 
PPTX
Home automation- An Iot application
Archana Golhar
 
PDF
WiGig PPT
Seminar Links
 
PDF
Internet Of Things (Question Paper) [October – 2018 | Choice Based Syllabus]
Mumbai B.Sc.IT Study
 
PDF
IoT architecture
Sumit Sharma
 
PDF
IoT Solutions for Smart Energy Smart Grid and Smart Utility Applications
Eurotech
 
PPTX
Challenges and application of Internet of Things
Ashutosh Bhardwaj
 
Internet of Things (IoT)
Prakhyath Rai
 
Iot ppt
piyuu7
 
Internet of Things (IoT) and its applications
Sarwan Singh
 
The essential role of AI in the 5G future
Qualcomm Research
 
Advanced: Private Networks & 5G Non-Public Networks
3G4G
 
10 min IoT ppt
Vaishnavu Pathayathodi
 
IoT and m2m
pavan penugonda
 
An iot based secured smart e-campus
shivaraj kumar gowdas
 
IoT material revised edition
pavan penugonda
 
Web servers for the Internet of Things
Alexandru Radovici
 
5G Multi-Access Edge Compute
Michelle Holley
 
Understanding IoT
Md. Shamsul Haque
 
Internet of Things (IoT) - Introduction ppt
sutrishnakar1995
 
Internet of things
Sandisiwe Sixanda
 
Home automation- An Iot application
Archana Golhar
 
WiGig PPT
Seminar Links
 
Internet Of Things (Question Paper) [October – 2018 | Choice Based Syllabus]
Mumbai B.Sc.IT Study
 
IoT architecture
Sumit Sharma
 
IoT Solutions for Smart Energy Smart Grid and Smart Utility Applications
Eurotech
 
Challenges and application of Internet of Things
Ashutosh Bhardwaj
 
Ad

Viewers also liked (20)

PPTX
Ad-Hoc Networks
Mshari Alabdulkarim
 
PPTX
Waste water treatment processes
Ashish Agarwal
 
ODP
Intro to Algebra II
teamxxlp
 
PPT
Orbital Notation
ZB Chemistry
 
PPTX
Top 10 senior administrative officer interview questions and answers
annababy1245
 
PDF
Packet capture and network traffic analysis
CARMEN ALCIVAR
 
PPT
Virtualization In Software Testing
Colloquium
 
PPT
Video Quality Measurements
Yoss Cohen
 
PPTX
Vendor quality management
G2Link
 
PDF
Digital Platform Selection Best Practices
edynamic
 
PPT
Analysis of water pollution presentaion by m.nadeem ashraf
University of Agriculture Faisalabad
 
PDF
Hands-On Lab: Let's Build an ITSM Dashboard
CA Technologies
 
PPT
Defining Workplace Safety
Bruce Lambert
 
DOCX
Str581 final exam part 1
Rogue Phoenix
 
PPTX
Which test cases to automate
sachxn1
 
RTF
Chem Lab Report (1)
Abigail Delgado
 
PPTX
E leave management-system
Arti Sehgal
 
PDF
PL/SQL Unit Testing Can Be Fun!
Raimonds Simanovskis
 
PDF
Catalogo de Productos Nutrilite (Amway)
Odair y Martha Multinivel
 
PDF
Telecom Roaming Overview
Shilpin Pvt. Ltd.
 
Ad-Hoc Networks
Mshari Alabdulkarim
 
Waste water treatment processes
Ashish Agarwal
 
Intro to Algebra II
teamxxlp
 
Orbital Notation
ZB Chemistry
 
Top 10 senior administrative officer interview questions and answers
annababy1245
 
Packet capture and network traffic analysis
CARMEN ALCIVAR
 
Virtualization In Software Testing
Colloquium
 
Video Quality Measurements
Yoss Cohen
 
Vendor quality management
G2Link
 
Digital Platform Selection Best Practices
edynamic
 
Analysis of water pollution presentaion by m.nadeem ashraf
University of Agriculture Faisalabad
 
Hands-On Lab: Let's Build an ITSM Dashboard
CA Technologies
 
Defining Workplace Safety
Bruce Lambert
 
Str581 final exam part 1
Rogue Phoenix
 
Which test cases to automate
sachxn1
 
Chem Lab Report (1)
Abigail Delgado
 
E leave management-system
Arti Sehgal
 
PL/SQL Unit Testing Can Be Fun!
Raimonds Simanovskis
 
Catalogo de Productos Nutrilite (Amway)
Odair y Martha Multinivel
 
Telecom Roaming Overview
Shilpin Pvt. Ltd.
 
Ad

Similar to Infrastructureless Wireless networks (20)

PDF
clustering Algorithms for Mobile Ad Hoc Networking (Slides for my opening def...
niminming
 
PDF
Mobile Networking Solutions for First Responders
Förderverein Technische Fakultät
 
DOC
Ambitlick old projects
ambitlicksolutions
 
PDF
White spaces above 3 g hz and an application
Brough Turner
 
PPT
November 9, Planning and Control of Unmanned Aircraft Systems in Realistic C...
University of Colorado at Boulder
 
PDF
Overlapped carrier sense multiple access
ambitlick
 
PDF
Wireless Charging for Multi-Node Using Wireless Charging Vehicle
paperpublications3
 
PPTX
Routing protocols for ad hoc wireless networks
Divya Tiwari
 
PPT
Adhoc wireless
Ipsita Sharma
 
PPSX
Qo s provisioning for scalable video streaming over ad hoc networks using cro...
Mshari Alabdulkarim
 
PPTX
Ad hoc routing
its
 
PDF
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
ijceronline
 
PPTX
Implementation of Wireless Channel Model in MATLAB: Simplified
Rosdiadee Nordin
 
PPTX
Routing Protocols for Adhoc wireless Network Dated 07-05-2023.pptx
VanshikaAgarwal499417
 
PDF
An adaptive power controlled mac protocol for
ambitlick
 
PDF
Wireless Control Systems - from theory to a tool chain, Mikael Björkbom, Aalt...
The Research Council of Norway, IKTPLUSS
 
PPT
Energy conservation in wireless communication systems with relays
Aniruddha Chandra
 
PPTX
Energy Aware performance evaluation of WSNs.
ikrrish
 
PPTX
Presentation l`aquila new
ikrrish
 
PDF
Advanced Networking on GloMoSim
Kathirvel Ayyaswamy
 
clustering Algorithms for Mobile Ad Hoc Networking (Slides for my opening def...
niminming
 
Mobile Networking Solutions for First Responders
Förderverein Technische Fakultät
 
Ambitlick old projects
ambitlicksolutions
 
White spaces above 3 g hz and an application
Brough Turner
 
November 9, Planning and Control of Unmanned Aircraft Systems in Realistic C...
University of Colorado at Boulder
 
Overlapped carrier sense multiple access
ambitlick
 
Wireless Charging for Multi-Node Using Wireless Charging Vehicle
paperpublications3
 
Routing protocols for ad hoc wireless networks
Divya Tiwari
 
Adhoc wireless
Ipsita Sharma
 
Qo s provisioning for scalable video streaming over ad hoc networks using cro...
Mshari Alabdulkarim
 
Ad hoc routing
its
 
IJCER (www.ijceronline.com) International Journal of computational Engineerin...
ijceronline
 
Implementation of Wireless Channel Model in MATLAB: Simplified
Rosdiadee Nordin
 
Routing Protocols for Adhoc wireless Network Dated 07-05-2023.pptx
VanshikaAgarwal499417
 
An adaptive power controlled mac protocol for
ambitlick
 
Wireless Control Systems - from theory to a tool chain, Mikael Björkbom, Aalt...
The Research Council of Norway, IKTPLUSS
 
Energy conservation in wireless communication systems with relays
Aniruddha Chandra
 
Energy Aware performance evaluation of WSNs.
ikrrish
 
Presentation l`aquila new
ikrrish
 
Advanced Networking on GloMoSim
Kathirvel Ayyaswamy
 

More from Gwendal Simon (14)

PDF
Reproducible research at ACM MMSys
Gwendal Simon
 
PDF
Netgames: history and preparing 2018 edition
Gwendal Simon
 
PDF
Virtual Reality in 5G Networks
Gwendal Simon
 
PDF
Adaptive Delivery of Live Video Stream: Infrastructure cost vs. QoE
Gwendal Simon
 
PDF
Research on cloud gaming: status and perspectives
Gwendal Simon
 
PDF
DASH in Twitch: Adaptive Bitrate Streaming in Live Game Streaming Platforms
Gwendal Simon
 
PDF
Fast Near-Optimal Delivery of Live Streams in CDN
Gwendal Simon
 
PDF
Scadoosh: Scaling Down the Footprint of Rate-Adaptive Live Streaming on CDN I...
Gwendal Simon
 
PDF
Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...
Gwendal Simon
 
PDF
Time-Shifted TV in Content Centric Networks: the Case for Cooperative In-Netw...
Gwendal Simon
 
PDF
Internet : pourquoi ça marche
Gwendal Simon
 
PDF
Optimal Network Locality in Distributed Services
Gwendal Simon
 
PDF
Cloud Engineering
Gwendal Simon
 
PDF
peer-to-peer oppotunities
Gwendal Simon
 
Reproducible research at ACM MMSys
Gwendal Simon
 
Netgames: history and preparing 2018 edition
Gwendal Simon
 
Virtual Reality in 5G Networks
Gwendal Simon
 
Adaptive Delivery of Live Video Stream: Infrastructure cost vs. QoE
Gwendal Simon
 
Research on cloud gaming: status and perspectives
Gwendal Simon
 
DASH in Twitch: Adaptive Bitrate Streaming in Live Game Streaming Platforms
Gwendal Simon
 
Fast Near-Optimal Delivery of Live Streams in CDN
Gwendal Simon
 
Scadoosh: Scaling Down the Footprint of Rate-Adaptive Live Streaming on CDN I...
Gwendal Simon
 
Minimizing Server Throughput for Low-Delay Live Streaming in Content Delivery...
Gwendal Simon
 
Time-Shifted TV in Content Centric Networks: the Case for Cooperative In-Netw...
Gwendal Simon
 
Internet : pourquoi ça marche
Gwendal Simon
 
Optimal Network Locality in Distributed Services
Gwendal Simon
 
Cloud Engineering
Gwendal Simon
 
peer-to-peer oppotunities
Gwendal Simon
 

Recently uploaded (20)

PDF
Agricultural_Statistics_at_a_Glance_2022_0.pdf
SandeepSingh286037
 
PDF
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
SOFTTECHHUB
 
PPTX
Digital-Transformation-Roadmap-for-Companies.pptx
David Malick Dieng
 
PDF
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
Julien SIMON
 
PDF
Architecting across the Boundaries of two Complex Domains - Healthcare & Tech...
Peter Tan
 
PDF
Mobile App Security Testing_ A Comprehensive Guide.pdf
flufftailshop
 
PDF
A comparative analysis of optical character recognition models for extracting...
IAESIJAI
 
PPTX
KOM of Painting work and Equipment Insulation REV00 update 25-dec.pptx
muhammadmuneebnaeem7
 
PDF
Build a system with the filesystem maintained by OSTree @ COSCUP 2025
Jian-Hong Pan
 
PDF
Dropbox Q2 2025 Financial Results & Investor Presentation
Dropbox
 
PDF
Reach Out and Touch Someone: Haptics and Empathic Computing
Mark Billinghurst
 
PDF
Electronic commerce courselecture one. Pdf
OmerMohamed64
 
PDF
Unlocking AI with Model Context Protocol (MCP)
Brian McKeiver
 
PPTX
sap open course for s4hana steps from ECC to s4
sreeni2106invites
 
PDF
Advanced methodologies resolving dimensionality complications for autism neur...
IAESIJAI
 
PDF
Encapsulation_ Review paper, used for researhc scholars
gurumoop
 
DOCX
The AUB Centre for AI in Media Proposal.docx
GAONE9
 
PDF
Machine learning based COVID-19 study performance prediction
IAESIJAI
 
PDF
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
SOFTTECHHUB
 
PDF
Network Security Unit 5.pdf for BCA BBA.
Serpent6
 
Agricultural_Statistics_at_a_Glance_2022_0.pdf
SandeepSingh286037
 
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
SOFTTECHHUB
 
Digital-Transformation-Roadmap-for-Companies.pptx
David Malick Dieng
 
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
Julien SIMON
 
Architecting across the Boundaries of two Complex Domains - Healthcare & Tech...
Peter Tan
 
Mobile App Security Testing_ A Comprehensive Guide.pdf
flufftailshop
 
A comparative analysis of optical character recognition models for extracting...
IAESIJAI
 
KOM of Painting work and Equipment Insulation REV00 update 25-dec.pptx
muhammadmuneebnaeem7
 
Build a system with the filesystem maintained by OSTree @ COSCUP 2025
Jian-Hong Pan
 
Dropbox Q2 2025 Financial Results & Investor Presentation
Dropbox
 
Reach Out and Touch Someone: Haptics and Empathic Computing
Mark Billinghurst
 
Electronic commerce courselecture one. Pdf
OmerMohamed64
 
Unlocking AI with Model Context Protocol (MCP)
Brian McKeiver
 
sap open course for s4hana steps from ECC to s4
sreeni2106invites
 
Advanced methodologies resolving dimensionality complications for autism neur...
IAESIJAI
 
Encapsulation_ Review paper, used for researhc scholars
gurumoop
 
The AUB Centre for AI in Media Proposal.docx
GAONE9
 
Machine learning based COVID-19 study performance prediction
IAESIJAI
 
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
SOFTTECHHUB
 
Network Security Unit 5.pdf for BCA BBA.
Serpent6
 

Infrastructureless Wireless networks

  • 1. Infrastructure-less Wireless Networks Gwendal Simon Department of Computer Science Institut Telecom 2009
  • 2. Literature Books include: “Algorithms for sensor and ad hoc networks”, D. Wagner and R. Wattenhofer “Wireless sensor networks: an information processing approach”, F. Zhao and L. Guibas and journal/conferences include: ACM SigMobile (MobiHoc, SenSys, etc.) IEEE MASS and WCNC Elsevier Ad-Hoc Network, Wireless Networks 2 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 3. Motivations Current wireless net. require an infrastructure: cellular network: interconnected base stations wifi Internet: an access point and Internet 3 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 4. Motivations Current wireless net. require an infrastructure: cellular network: interconnected base stations wifi Internet: an access point and Internet Same flaws than centralized architectures: cost scalability privacy dependability 3 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 5. Motivations Sometimes, there is no infrastructure transient meeting disaster areas military interventions alter-communication 4 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 6. Motivations Sometimes, there is no infrastructure transient meeting disaster areas military interventions alter-communication Sometimes not every station hear every other station limited wireless transmission range large-scale area 4 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 7. Multi-hop Wireless Networks Nodes: portable wireless devices transmission ranges do not cover the area density ensures network connectivity Links: wireless characteristics transmission model: local broadcasting energy consumption: transmission is costly Behavior: devices emit, receive and forward data 5 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 8. A Taxonomy of Applications 6 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 9. Ad-Hoc vs. Sensor Networks Ad-Hoc Networks Sensor Networks nodes powerful wifi devices tiny zigbee nodes algorithms all-to-all routing echo to sink mobility human or car motions failures performance criteria quality of service energy consumption 7 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 10. Ad-Hoc Applications Delay-Tolerant Network (social media application) assumption: no connectivity, but high mobility objective: ensuring eventual message delivery 8 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 11. Ad-Hoc Applications Delay-Tolerant Network (social media application) assumption: no connectivity, but high mobility objective: ensuring eventual message delivery Mesh Networks (rural wireless coverage) assumption: some nodes have Internet access objective: maintaining path to these nodes 8 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 12. Ad-Hoc Applications Delay-Tolerant Network (social media application) assumption: no connectivity, but high mobility objective: ensuring eventual message delivery Mesh Networks (rural wireless coverage) assumption: some nodes have Internet access objective: maintaining path to these nodes Vehicular Ad-Hoc Networks assumption: a particular mobility model objective: mostly services related to car safety 8 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 13. Sensor Network Applications Sink-Based Networks (monitoring of natural areas) assumption: one sink retrieves all sensed data objective: increasing life-time 9 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 14. Sensor Network Applications Sink-Based Networks (monitoring of natural areas) assumption: one sink retrieves all sensed data objective: increasing life-time Mobile Object Tracking (area surveillance) assumption: sensors know their location objective: determining hostile position 9 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 15. Sensor Network Applications Sink-Based Networks (monitoring of natural areas) assumption: one sink retrieves all sensed data objective: increasing life-time Mobile Object Tracking (area surveillance) assumption: sensors know their location objective: determining hostile position Multi-Sink Networks (intervention teams) assumptions: mobile sinks and fixed sensor objectives: increasing sink coverage 9 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 16. Short Introduction to Popular Models 10 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 17. Network as a Graph Unit-Disk Graph: 05 03 → node position 07 09 → circular transmission 12 11 06 10 01 → boolean connections 02 00 08 04 11 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 18. Network as a Graph Unit-Disk Graph: 05 03 → node position 07 09 → circular transmission 12 11 06 10 01 → boolean connections 02 00 08 04 11 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 19. Network as a Graph Unit-Disk Graph: 05 03 → node position 07 09 → circular transmission 12 11 06 10 01 → boolean connections 02 00 08 04 11 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 20. Interferences Signal-to-noise-plus-interference (SINR) ratio Pu d(u,v )α Pw ≥β N+ w ∈V {u} d(w ,v )α Pu : power level of sender u d(u, v ): distance between u and v α: path-loss exponent N: noise β: minimum ratio 12 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 21. A Tour of the Most Studied Issues 13 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 22. Broadcasting I: Stormy Effect Broadcast: a simple basic problem : a source emits a message all nodes within the network eventually receive the message a simple and efficient solution: upon first reception of message, forward it. Limits of flooding in wireless networks: redundant messages interferences 14 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 23. Broadcasting II: Proposals Probabilistic flooding: idea: forward the message with some probability p drawbacks: no guarantee of delivering refinements: adjust p to node density 15 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 24. Broadcasting II: Proposals Probabilistic flooding: idea: forward the message with some probability p drawbacks: no guarantee of delivering refinements: adjust p to node density Constrained flooding: idea: only some nodes forward the message implementation: build the Minimum Connected Dominating Set drawbacks: maintaining cost in dynamic systems 15 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 25. Mobility Models I Few theoretical proof, few real implementations ⇒ generate realistic node motions for simulations 16 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 26. Mobility Models I Few theoretical proof, few real implementations ⇒ generate realistic node motions for simulations The simplest model: Random Waypoint 1. each node picks a random position uniformly 2. it travels toward this destination with a speed v 3. once it reaches it, it stops during few seconds 4. back to 1 16 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 27. Mobility Models II: Improvements Basic Structural Flaws: non-uniform distribution of node location: higher node distribution in the center average speed decay: low speed nodes spend more time to travel Realistic Mobility Models: group movement area popularity urban models community-based 17 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 28. Mobility Models II: Improvements Basic Structural Flaws: non-uniform distribution of node location: higher node distribution in the center average speed decay: low speed nodes spend more time to travel Realistic Mobility Models: group movement area popularity urban models community-based the most realistic one : using real traces! 17 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 29. Localized Data Gathering Basic idea: query data from sensors within an area two rounds: query diffusion retrieve data from sensors main objectives: minimize energy consumption minimize the delay A problem related with broadcasting except: only sensors from the queried area are reached: complex queries are possible (average, max, etc.) 18 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 30. Time Synchronization I Different time on nodes: different oscillator frequency ⇒ frequency error absolute difference between clocks ⇒ phase error The need of a common clock localization protocols some MAC protocols data fusion in sensor network 19 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 31. Time Synchronization II Broadcasting standard time via GPS system: √ precision, simple implementation × expensive devices × limited usage (outdoor environment) Achieve a common time distributively: √ (almost) no special devices required √ more tolerant to the environment × special protocols × message overhead, multi-hop delays 20 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 32. A Focus on Routing Protocols 21 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 33. Routing Protocols Objective: select a path between a source and a destination Main design challenges: unstable network topology low-cost devices (energy, computing. . . ) Main routing mechanisms: neighbor discovering route setup route maintenance 22 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 34. Proactive routing vs. On demand routing Proactive Reactive Setup all-to-all on demand Maintenance regularly during utilization Advantages no setup delay no unused routes Disadvantages fixed overhead long setup delay Main examples OLSR AODV 23 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 35. AODV Route Discovery G F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 36. AODV Route Discovery Broadcasting RREQ Mes- G sage. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 37. AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 38. AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 39. AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 40. AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 41. AODV Route Discovery Setting up G reverse path. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 42. AODV Route Discovery Replying RREP to G source. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 43. AODV Route Discovery Forward path G setup. F H S B E D A 24 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 44. AODV Route Maintenance Link breaks between B G and D. F H S B E D A 25 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 45. AODV Route Maintenance Sending RERR mes- G sage. F H S B E D A 25 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 46. AODV Route Maintenance Restarting route discov- G ery. F H S B E D A 25 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 47. AODV Route Maintenance New route G discovered. F H S B E D A 25 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 48. Some Tricks Intelligent flooding (detect close destination) idea: init TTL at 1, then 2, then 3. . . idea: flood slowly and send message to stop it 26 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 49. Some Tricks Intelligent flooding (detect close destination) idea: init TTL at 1, then 2, then 3. . . idea: flood slowly and send message to stop it Route caching (use past flooding) idea: during flood, answer for a distant node drawback: contradict reactive routing philosophy 26 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 50. Some Tricks Intelligent flooding (detect close destination) idea: init TTL at 1, then 2, then 3. . . idea: flood slowly and send message to stop it Route caching (use past flooding) idea: during flood, answer for a distant node drawback: contradict reactive routing philosophy Local maintenance (almost unchanged route) idea: instead of NAK s, look for d by yourself drawback: sometimes it does not work 26 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 51. A Proactive Routing Protocol: OLSR Objective: make use of Multi-Point Relay (MPR) acting as super-peers easing topology discovery handling most of the traffic OLSR message types: HELLO: discover 1-hop and 2-hop neighbors topology discovery through MPR 27 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 52. Neighbor sensing F G H S B D E A 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 53. Neighbor sensing F G Nb:{S}, H 2hopNb:{} Broadcasting S B HELLO Message. Nb:{S}, 2hopNb:{} D E A Nb:{S}, 2hopNb:{} 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 54. Neighbor sensing F G H S B Nb:{E}, Nb:{S,E}, 2hopNb:{} 2hopNb:{} D E A Nb:{E}, 2hopNb:{S} 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 55. Neighbor sensing F G Nb:{F}, H 2hop Nb:{S} S B Nb:{E,F}, Nb:{S,E,F}, 2hop Nb:{} 2hop Nb:{} D E A 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 56. Neighbor sensing F G Nb:{S,B,G}, Nb:{F,B,H}, H 2hopNb:{E,A,H} 2hopNb:{S,E,A,D} Nb:{G,D}, 2hopNb:{B,F,A} S B Nb:{E,F,B}, Nb:{S,E,F,A,G}, 2hopNb:{G,A} 2hopNb:{D,H} D E A Nb:{A,H}, Nb:{S,A,B}, Nb:{E,B,D}, 2hopNb:{B,E,G} 2hopNb:{F,G,D} 2hopNb:{S,F,G,H} 28 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 57. MPR selection F G Nb:{S,B,G}, H 2hopNb:{E,A,H} S B Nb:{E,F,B}, Nb:{S,E,F,A,G}, 2hopNb:{G,A} 2hopNb:{D,H} D E A Nb:{S,A,B}, 2hopNb:{F,G,D} 29 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 58. MPR selection F G HELLO message H indicating B as S B MPR of S and B note S as its MPR selector. D E A 29 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 59. MPR selection F G MPR Selector: MPR Selector: H {} {B,F,H} MPR Selector: {G,D} S B MPR Selector: MPR Selector: {} {S,G,E,F,A} D E A MPR Selector: MPR Selector: MPR Selector: {A,H} {} {B,E,D} 29 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 60. Topology Table Each node maintains a Topology Table containing all possible destinations notifying a MPR to reach them Structure of Topology Table (on S for example): Dest Addr Last Hop Seq Holding Time G B 1 10 A B 4 20 D A 6 10 H G 5 15 ... ... ... ... 30 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 61. Building the Topology Table F G MPR Selector: H {B,F,H} S B D E A 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 62. Building the Topology Table F G H S B Topology Table Des Lhop Seq Htime D F E G 2 30 A H G 2 30 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 63. Building the Topology Table F G H MPR Selector: {G,D} S B MPR Selector: {S,G,E,F,A} D E A 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 64. Building the Topology Table F G H Topology Table Des Lhop Seq Htime S F G B2 30 H G 2 30 B G 2 30 D E A 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 65. Building the Topology Table F G H Broadcasting contin- S B ues. . . D E A MPR Selector: MPR Selector: {A,H} {B,E,D} 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 66. Building the Topology Table F G MPR Selector: MPR Selector: H {} {B,F,H} MPR Selector: {G,D} S B MPR Selector: MPR Selector: {} {S,G,E,F,A} D E A MPR Selector: MPR Selector: MPR Selector: {A,H} {} {B,E,D} 31 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 67. Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 E A 6 30 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
  • 68. Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 E A 6 30 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
  • 69. Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 E A 6 30 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
  • 70. Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 A B 2 E A 6 30 G B 2 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
  • 71. Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 A B 2 E A 6 30 G B 2 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
  • 72. Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 A B 2 E A 6 30 G B 2 D A 6 30 A D 7 30 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
  • 73. Building the Routing Table Topology Table on S Neighbor Table on S Des Lhop Seq Htime Nb:{E,F,B}, F G 2 30 2hopNb:{G,A} H G 2 30 B G 2 30 Routing Table on S F B 3 30 Des Nhop Hops A B 3 30 E E 1 E B 3 30 F F 1 G B 3 30 B B 1 B A 6 30 A B 2 E A 6 30 G B 2 D A 6 30 H B 3 A D 7 30 D B 3 H D 7 30 D H 8 30 32 / 41 G Gwendal Simon8 H 30 Infrastructure-less Wireless Networks
  • 74. Any Hybrid Approach ? Merging advantages from both approaches: build a routing table at 4 ∼ 5 hops launch a reactive process if d is not Applicative concerns: OLSR is attractive because networks often small AODV scales well but no all-to-all routing 33 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 75. Research Activity: Multi-Sinks Query Range 34 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 76. Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable 50 0 50 100 150 200 250 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 77. Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable 50 0 50 100 150 200 250 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 78. Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable 50 0 50 100 150 200 250 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 79. Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable 50 0 50 100 150 200 250 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 80. Multi-sink Multi-hop WSN 200 Target application: “fireman application” Many sensors (small, blue) and some firemen (large, green) 150 Firemen talk directly with the sensors Gather only local information On demand, fixed rate data gathering 100 Hop based query, constrained flooding Simple to deploy and scalable Networking assumptions: 50 IEEE 802.15.4 MAC layer, ZigBee tree routing No in-network data aggregation, compression 0 Static sensors100 50 and sinks 150 (may extend to mobile 250 200 sinks) 300 35 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 81. Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 u1 = 5 S2 50 u2 = 1 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 82. Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 (4, 2) no u1 = 4 S2 50 u2 = 2 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 83. Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 (4, 2) no (3, 2) yes u1 = 3 S2 50 u2 = 2 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 84. Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 (4, 2) no (3, 2) yes u1 = 2 (2, 3) yes S2 50 u2 = 3 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 85. Network Sharing Without Congestions 200 Capacity of sensors c = 5 Each flow consumes r = 1 Nodes within u hops generate traffic 150 Configurations Feasible? (5, 1) yes S1 100 (4, 2) no (3, 2) yes u1 = 2 (2, 3) yes S2 50 u2 = 3 62 configurations 0 50 100 150 200 250 300 0 36 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 86. Which Configuration is Better? 200 Basic considerations: (4, 2): not feasible, (1, 1): inefficient 150 100 50 0 50 100 150 200 250 300 37 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 87. Which Configuration is Better? 200 Basic considerations: (4, 2): not feasible, (1, 1): inefficient 150 Optimality criteria: 1 Maximum Impact Range 100 (5, 1): Sum up to 6 Max-Min Fairness 4 (2, 3) = (3, 2) (5, 1) 3 2 50 (?, ?, ?, ?) 0 50 100 150 200 250 300 37 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 88. Problem Formulation Multi-Dimensional Multiple Choice Knapsack Problem a NP-complete problem 38 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 89. Problem Formulation Multi-Dimensional Multiple Choice Knapsack Problem a NP-complete problem Toward a distributed heuristic algorithm only local views of the network only local optimal solutions 38 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 90. Protocol 200 At each sink: enlarge requirement periodically receive notification from sensors 150 adjust requirement if it is smaller At each sensor: 3 100 measure the traffic detect congestion A 50 4 2 1 solve the local problem notify related sinks 0 50 100 150 200 250 300 39 / 0 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 91. Conclusion 40 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 92. Personal Thoughts Great theoretical importance: a lot of new and scientifically exciting problems a multi-disciplinary field (network, algorithms, computational geometry, probabilities) 41 / 41 Gwendal Simon Infrastructure-less Wireless Networks
  • 93. Personal Thoughts Great theoretical importance: a lot of new and scientifically exciting problems a multi-disciplinary field (network, algorithms, computational geometry, probabilities) Unsure applicative importance: no killer application yet cellular networks just do what we want 41 / 41 Gwendal Simon Infrastructure-less Wireless Networks