Routing in Opportunistic Networks

Transcription

1 Routing in Opportunistic Networks Silvia Giordano work in collaboration with: Marco Conti, Jon Crowcroft, Pan Hui, Hoang Anh Nguyen, Andrea Passarella and the whole Haggle project thanks to Cecilia Mascolo and Martin May SUPSI-DTI 1 Outline Internet: where are we? ubiquitous computing/communication clean-slate approach New Networking Technologies Haggle DTN Opportunistic Networks Routing in OppNets context based routing C-BR classification SANETs SUPSI-DTI 2 1

2 Ubiquitous Computing: Business Users users must interrupt their work when they move users can work without interrupting their work even when they move SUPSI-DTI 3 Ubiquitous Computing: Other Users visiting people young people SUPSI-DTI 4 2

3 Example: realize the ubiquitous computing with MANET technology. First dream M. Conti and S. Giordano. Multihop Ad Hoc Networking: The Theory. IEEE Communications Magazine, SUPSI-DTI 5 Missing a pragmatic approach! Example: realize the ubiquitous computing with MANET technology. M. Conti and S. Giordano. Multihop Ad Hoc Networking: The Reality. IEEE Communications Magazine, SUPSI-DTI 6 3

4 But pervasiveness is a reality Chatpen Digital Information recognized and transmitted by pen CeivaDigital Picture Frame Plug & Play 640 x 480 VGA JPEG Image Format Storage: 10 photos/250 on - line SUPSI-DTI 7 Clean-slate approach Heterogeneous networks technologies Heterogeneous applications New architecture and communications paradigms: DTNs ANA Haggle: opportunistic networks, social networks SUPSI-DTI 8 4

5 The Internet Hourglass Voice, Video, P2P, , youtube,. Protocols TCP, UDP, SCTP, ICMP, Changing/updating the Internet core is difficult or impossible! (e.g. Multicast, Mobile IP, QoS, ) Everything on IP Application layer IPv6 Link IP on layer Everything Ethernet, WIFI (802.11), ATM, SONET/SDH, FrameRelay, modem, ADSL, Cable, Bluetooth Homogeneous networking abstraction Disruptive approaches need a disruptive architecture SUPSI-DTI 9 Application layer Link layer Approach 1: ANA We have to widen the waist and host more paradigms Federation instead of homogeneous abstraction May the best concept wins We need the framework for these multiple networks SUPSI-DTI 10 maym@tik.ee.ethz.ch 10 5

6 ANA "one-size-fits-all" A network compartment implements the operational rules and administrative policies for a given communication context Application ANA framework Link layer Publish Subscribe Compartment PodNet compartment SUPSI-DTI 11 maym@tik.ee.ethz.ch 11 Application layer Link layer Approach 2: DTN Support interoperability across radically heterogeneous networks Tolerate delay and disruption Acceptable performance in high loss/delay/error/disconnected environments Decent performance for low loss/delay/errors Still IP based SUPSI-DTI 12 maym@tik.ee.ethz.ch 12 6

7 Delay-Tolerant Networking Architecture* Components Flexible naming scheme Message abstraction and API Extensible Store-and-Forward Overlay Routing Per-(overlay)-hop reliability and authentication Still IP based *from DTN group slides SUPSI-DTI 13 bandwidth Example Routing Problem * time (days) bike satellite phone Connectivity: Village 1 City 2 City Internet bike *from DTN group slides 3 1 Village SUPSI-DTI 14 7

8 DTNs: What? Network with intermittent connectivity Data are disseminated exploiting host mobility long and variable delays high error rates heterogeneity SUPSI-DTI 15 DTNs: Why? To cover regions with no infrastructure Human social networks Intelligent highways Emergency rescue Sensor networks Wildlife monitoring SUPSI-DTI 16 8

9 Application layer Data Haggle Link layer Physical Link No more layers! Approach 3: Haggle new autonomic networking architecture enables communication for intermittent network connectivity, exploits autonomic opportunistic communications (i.e., when no end-to-end communication infrastructures). support interoperability across radically heterogeneous networks SUPSI-DTI Haggle ( ) I have 100M bytes of data, who can carry for me? Give it to me, I have 1G bytes phone flash. Thank you but you are in the opposite direction! We can also carry for you! Don t give to me! I am running out of storage. Reach an access point. Started January EU FP6 FET developed 3 prototypes (Infant, Child & Young Haggle) 100MB Finally, it code on sourceforge arrives Internet SUPSI-DTI 18 9

10 User a revolutionary paradigm for communicating without end-to-end connection relax the end-to-end paradigm HIS Forward Resource control Incentive Authentication Classifier Haggle control Bluetooth GPRS Connectivity Cradle Ethernet management SUPSI-DTI 19 a context based routing fwding manager with more strategies Uh! I love La Bonheme! (My wife should have it ) Sorry, I don t know any singer called Bonheme! There is one in my pocket Search La Bonheme.mp3 for me Search La Bonheme.mp3 for me Search La Bonheme.mp3 for me SUPSI-DTI 20 10

11 Opportunistic Networks Opportunistic networking: we are surrounded by communicating devices and most of applications are not real time constrained relax the end-to-end paradigm communication is on-the-fly heterogeneous devices mobile phones laptops, palm sensors, etc... exploit mobility real best-effort SUPSI-DTI 21 Opportunistic Networks The Missing Link? In oppnets, diverse systems not deployed originally as oppnet nodes join an oppnet dynamically in order to perform certain tasks they have been invited (or ordered) to participate in Not always connected, internet connectivity islands Huge amount of untapped resources in portable devices Local wireless bandwidth Storages CPUs A packet can reach destination using network connectivity or user mobility SUPSI-DTI 22 11

12 Objectives Oppnets are envisioned to provide, among others: Bridges between disjoint communication media Additional platforms for offloading tasks Additional sensing modalities by integrating existing independent sensory systems Oppnets can exploit context information There is one in my pocket Search La Bonheme.mp3 for me CONTEXT Search La Bonheme.mp3 for me Search La Bonheme.mp3 for me SUPSI-DTI 23 OppNets routing: context information the context in which the users communicate. Examples: users' working address and institution, personal information, habits SUPSI-DTI 24 12

13 OppNets routing: context information Thus: given context information about the destination suitable forwarders can be found based on: the probability of meeting with other users the probability of visiting particular places I ve a message for Jack Jack is at school school bus is just coming I ll wait the school bus and give the message to someone in it, who is going to school, who is likely to meet Jack in few minutes SUPSI-DTI 25 Context aware routing classification context-oblivious: basically exploit some form of flooding partially context-aware (aka mobility-based): exploit some particular piece of context information to optimise the forwarding task. fully context-aware (aka social contextaware): not only exploit context information to optimise routing, but also provide general mechanisms to handle and use context information SUPSI-DTI 26 13

14 Context-oblivious Protocols exploit some form of flooding. a message should be disseminated as widely as possible. might be the only solution when no context information is available. generate a high overhead may suffer high contention and potentially lead to network congestion SUPSI-DTI 27 Context-oblivious Protocols To limit this overhead, the common technique is to control flooding by limiting the number of copies allowed to exist in the network by limiting the maximum number of hops a message can travel. In the latter case, when no relaying is further allowed, a node can only send directly to the destination when (in case) it is met SUPSI-DTI 28 14

15 Context-oblivious Protocols Epidemic Routing A. Vahdat and D. Becker. Epidemic Routing for Partially Connected Ad Hoc Networks. Technical Report CS , CS. Dept. Duke Univ., Network Coding J. Widmer and J. Le Boudec. Network coding for efficient communication in extreme networks. Proc. of the ACM SIGCOMM WDTN, Spray&Wait T. Spyropoulos, K. Psounis, and C. Raghavendra. Efficient routing in intermittently connected mobile networks: The multiple-copy case, ACM/IEEE Transactions on Networking, 16, SUPSI-DTI 29 Context-oblivious Protocols Epidemic Routing When two nodes meet exchange summary vectors with compact unambiguous representation of the messages currently stored in their local buffers. Then, each node requests from the other the messages it is currently missing. hop count limits dissemination (hc = 1 -> only to dest.). A V[msg] a,b,d B SUPSI-DTI 30 15

16 Network Coding Context-oblivious Protocols an original approach to limit message flooding. ca a + c ac a + c A B C In general, network coding-based routing outperforms flooding, as it is able to deliver the same amount of information with fewer messages injected into the network SUPSI-DTI 31 Spray&Wait Context-oblivious Protocols two phases: the spray phase and the wait phase. spray phase: L copies of the same message are spread over the network wait phase: each relay node stores its copy and eventually delivers it to the destination when (in case) it comes within reach. L can is chosen based on a target average delay. spray can be performed in multiple ways (e.g. binary) SUPSI-DTI 32 16

17 Partially Context-aware exploit some particular piece of context information to optimise the forwarding task. e.g. mobility of nodes Main difference with fully context-aware: generally is customized for a specific type of context information does not provide a full-fledged set of algorithms to gather and manage any type of context information SUPSI-DTI 33 Partially Context-aware Probabilistic ROuting Protocol using History of Encounters and Transitivity (PROPHET) A. Lindgren, A. Doria, and O. Schelen. Probabilistic routing in intermittently connected networks. ACM Mobile Computing and Communications Review, MV and MaxProp B. Burns, O. Brock, and B. Levine. MV routing and capacity building in disruption tolerant networks. Proc. of the 24th INFOCOM, J. Burgess, B. Gallagher, D. Jensen, and B. Levine. MaxProp: Routing for Vehicle-Based Disruption-Tolerant Networks. Proc. of the 25th INFOCOM, SUPSI-DTI 34 17

18 Partially Context-aware Last Encounter Routing and Spray&Focus M. Grossglauser and M. Vetterli. Locating nodes with EASE: last encounter routing in ad hoc networks through mobility diffusion. Proc. of the 22 nd INFOCOM, T. Spyropoulos and K. Psounis. Spray and focus: Efficient mobilityassisted routing for heterogeneous and correlated mobility. Proc. of IEEE Percom International Workshop on Intermittently Connected Mobile Ad hoc Networks, March MobySpace Routing T. C. V. Leguay, J.; Friedman. Evaluating mobility pattern space routing for dtns. Proc. of the 25th INFOCOM, Bubble Rap P. Hui and J. Crowcroft. Bubble rap: Forwarding in small world dtns in every decreasing circles. Technical report, Technical Report UCAM-CL- TR684, Univ. of Cambridge, SUPSI-DTI 35 PROPHET Partially Context-aware evolution of Epidemic: during a contact, nodes also exchange their delivery predictability (DP) to destinations of messages they store in their buffers messages are requested only if DP is higher than that of the node currently storing the message. DP is the probability for a node to encounter a certain destination increases when the node meets the destination decreases (according to an ageing function) between meetings context information used by PROPHET is the frequency of meetings between nodes SUPSI-DTI 36 18

19 MV and MaxProp Partially Context-aware in addition, also exploit information about the frequency of visits to specific physical places. Last Encounter Routing and Spray&Focus exploit the fact that the decreasing gradient of the time lag identifies a suitable path towards the destination. MobySpace Routing the mobility pattern of nodes is the context information used for routing SUPSI-DTI 37 Bubble-Rap automatically inferring the parameters of the underlying social structure. exploiting the structure properties to select paths. dynamically identifies users communities, and ranks the nodes sociability (measured as the number of links) inside each community SUPSI-DTI 38 19

20 Fully Context-aware provide general mechanisms to handle and use context information more general than PCA: works with any context information can be customized for the specific environment it must work SUPSI-DTI 39 Fully Context-aware Context-Aware Routing (CAR) M. Musolesi, S. Hailes, and C. Mascolo. Adaptive routing for intermittently connected mobile ad hoc networks. Proc. of WoWMoM History Based Opportunistic Routing (HiBOp) C. Boldrini, M. Conti, I. Jacopini, A. Passarella, and P. IIT-CNR. HiBOp: a History Based Routing Protocol for Opportunistic Networks. Proc. of WoWMoM 2007 Probabilistic Routing Protocol for Intermittently Connected Mobile Ad hoc Network (PROPICMAN/Spatio-Tempo) H. Anh Nguyen, S. Giordano, A. Puiatti. Probabilistic Routing Protocol for Intermittently Connected Mobile Ad Hoc Networks (PROPICMAN). Proceedings of IEEE WoWMoM/AOC H. Anh Nguyen, S. Giordano, Spatiotemporal Routing Algorithm in Opportunistic Networks. Proceedings of IEEE WoWMoM/AOC SUPSI-DTI 40 20

21 Context-Aware Routing (CAR) Nodes are divided in groups Nodes inside the same group have social links between each other Are connected by an underlying MANET routing protocol Nodes of different groups can have social links as well Nodes reconfigure SUPSI-DTI 41 CAR A node selects the fwder node/group Probabilistically Based on the social attraction among nodes multi-attribute utility-based framework is defined to combine context information for computing the delivery probabilities CAR compute delivery probabilities proactively, and disseminate them in their ad hoc cloud. context information is exploited to evaluate probabilities just for known destinations SUPSI-DTI 42 21

22 CAR: conjunction beteewn PCA & FCA The framework is general enough to accommodate different types of context information. Context information in CAR consists of the logical connectivity of nodes, such as the rate of connectivity change, the delivery probability of the neighbor nodes to the destination, and device information, such as residual battery life. However, the social context information is not taken into consideration as in social context-based category CAR can be seen as a conjunction between mobility-based protocols and social context-based protocols SUPSI-DTI 43 HiBOp Mimic the establishment of social relationships among people Users describe themselves Users decide what information about themselves and their habits they want to make publicly available users store this information into their own devices HiBOp is not bound to any predefined set of information Identity Table Personal Information Name Donald Surname Duck d.duck@cnr.it Phone Work Street Via Moruzzi, 1 City Institute attributes Pisa IIT-CNR values SUPSI-DTI

23 HiBOp in a nutshell Sender does several copies (k) of the message To achieve a target delivery probability k = min K K i =0 (i ) (1 p succ max ) p l Messages are discarded after being handed over to the next hop i.e., message spread is controlled by the target delivery probability SUPSI-DTI Propicman: Probabilistic Routing Protocol for Intermittently Connected Mobile Ad hoc Network Exploit the context information of nodes to select the best next hop candidate. Each node has a common node profile with evidence/value pairs. Evidences have different weights. Compute the deliverity probability to the destination. Zero-conf SUPSI-DTI 46 23

24 Propicman social approach Common set of evidences for every node - T Evidence names(e) E1 E2 En Weights(W) W1 W2 Wn Node profile: Evidence names Values E1 E2 En V1 V2 Vn Hashed values H(E1,V1) H(E2,V2) H(En,Vn) An example: Evidence names Values Hashed values Evidence names(e) Weights(W) Work place SUPSI 26a48976c47c0d2 b2383 Work place City Profession City Profession Lugano 14a12de02cd9c0c 5fd94 c056817c8f c83 Nationality 2 Nationality Swiss 536a195c27b76cc 2b SUPSI-DTI 47 Propicman P A1 A 1 Sender h M P A A *max{p h M Ai } P A2 A 2 P A3 A 3 P An A n SUPSI-DTI 48 24

25 Propicman Neighbors do local matching between received h M with its node profile. Message header h M H(E D1,V D1) H(E D2,V D2) H(E D3,V D3) H(E D4,V D4) H(E Dn,V Dn) MAC S SN H(E 1,V 1) H(E 2,V 2) H(E 3,V 3) H(E 4,V 4) H(E n,v n) Then neighbor node (A) can compute its P A: i Hashed values in node profile 0 P A = matched ( Wi) WM : sum of the weights of all the evidences that S knows about D. W Mi SUPSI-DTI 49 Morning SUPSI-DTI 50 25

26 Afternoon SUPSI-DTI 51 Evening SUPSI-DTI 52 26

27 Similarity! a period One cycle 7am-8am 8am-9am 9am-12am 12am-1pm 1pm-5pm 5pm-8pm 8pm-7am SUPSI-DTI 53 SpatioTempo Design Each period can have different contact opportunities. In some periods source node encounters few devices, some encounter bunch of devices with high DP. T i : period ith of the cycle c-1 S i : set of encountered nodes in period ith of the cycle c-1 S i T i S i + 1 T i + 1 c Si c 1 Ti 1 One cycle SUPSI-DTI 54 27

28 Variation of the sets of encounter nodes The variation of the set of encounter nodes in a period is indicated by EV i EV i is computed with respect to the past cycle (same period) and to the previous period (same cycle) S i c 1 T i S i + 1 c 1 T i + 1 c Si 1 c Ti 1 EV i EV i S = i S S c i 1 i S S i SUPSI-DTI 55 c i 1 S c i 1 SpatioTempo Design Each person is supposed to have a set of frequent contact persons (FD). The source node S wants to send a message M to the destination D. There are 2 cases: D belongs to FD D is an occasional destination (D belongs to OD) SUPSI-DTI 56 28

29 Recording task T i c 1 T i + 1 c Ti 1 S i S i S i + 1 S i S i + 1 S i SUPSI-DTI 57 Sliding window cycle S i S i + 1 c Si 1 t i Compute the number of retransmissions D belongs to OD: Basically use Propicman to select the next hop with the number (F) of rebroadcasting M H, depends on the battery level (b), the priority level of µ message ( ) and EV i in each period. 0 if b < β (0< β <1) F= round( EV i µ * t* ) i : number of units of time in the period SUPSI-DTI 58 29

30 D belongs to FD Delivery probability of a period: Delivery probability to reach the destination if the sender sends the message in this period. The number of retransmissions: c c DP i DP i F= round( FM * DPi * EVi + ( FM Fm ) * DPi + Fm ) c DPi * EVi + 1 F M, F : maximum and minimum number of times that S is able to m send during the period c c * 1 DP i + η DP i = 1 + η c SUPSI-DTI 59 D belongs to FD Given t i is number of units of time in the period And if S can resend M H each unit of time in the period, then F equals ti M c t * DP * EV + ( t 1) * DP F = round ( i i i i ) DP c i * EV i + 1 c SUPSI-DTI 60 30

31 DP DP DP M H M H M H DP M H maxdp maxdp M H M H M H maxdp S S S S S S T i T i + 1 Cycle c Ti SUPSI-DTI 61 OD/FD management & Message lifetime After one cycle, nodes review its OD, FD lists Each member in FD, OD of S has a frequency level (FL), which indicates how often S has a message to it. FL D > δ : removed FL D > δ : promoted OD FD Message lifetime: M TTL = µ * C T FL D > δ : demoted C T : The duration of time of a cycle SUPSI-DTI 62 31

32 Predict next best interval Having information on the past cycle and past interval, we can predict next best interval S i c 1 T i S i + 1 c 1 T i + 1 c Si 1 c Ti 1 c DP i SUPSI-DTI 63 Opportunistic Sensor Networks Context oblivious: Data Mules, SWIM, ZebraNet, Message Ferrying, DFT-MSN Z. Zhang. Routing in intermittently connected mobile ad hoc networks and delay tolerant networks: overview and challenges. IEEE Communications Surveys & Tutorials, R. C. Shah, S. Roy, S. Jain, W. Brunette. Data MULEs: Modeling a three-tier architecture for sparse sensor networks. SNPA, Muhammad Mukarram Bin Tariq, Mostafa Ammar, Ellen Zegura. Message Ferry Route Design for Sparse Ad Hoc Networks with Mobile Nodes. In Proceedings of ACM Mobihoc06, SUPSI-DTI 64 32

33 Opportunistic Sensor Networks Context aware SCAR = Sensor based Context-Aware Routing Opportunistic Mobile Sensor Data Collection with SCAR. Bence Pasztor, Mirco Musolesi and Cecilia Mascolo. In Proceedings of MASS 2007 PROSAN: PRobabilistic Opportunistic Routing in SANets Probabilistic Opportunistic Routing in SANETs. A.H. Nguyen, S. Giordano. In Proceedings of ACM SANET MobiCOM SUPSI-DTI 65 Motivation Pervasiveness and repeating behavior are true also in SANET: people who repeatedly stay and pass in area A are the ones who most likely want, need and will support a reaction. By naturally acting, people can enable the communication between sensors and actuators, and trigger the reaction SUPSI-DTI 66 33

34 Motivating Scenario same for animals SUPSI-DTI 67 SCAR Mobile sensors collect information and try to deliver this opportunistically to one of the sinks Energy is an issue (batteries cannot be changed so easily) They exploit best carriers to deliver the information back to sinks Choice of the best carriers is key to the protocol: SCAR uses utility theory and Kalman Filter based prediction of various attributes to choose SUPSI-DTI 68 34

35 Assumptions Data is reasonably delay tolerant Clocks of the sensors are (reasonably) synchronized and the sensors wake up at regular intervals Sensors are all cooperating (possibly owned by same developer) Mobility is not source of energy consumption SUPSI-DTI 69 The key ideas of SCAR Each sensor collects and stores its data Each sensor will try to deliver its data in bundles to a number of neighbouring sensors which seem to be the best carriers to reach a sink Decision process based on the prediction of the evolution of the system: mobility of nodes colocation with the sinks battery level The forecasted values of the context attributes are then combined to define a probability P(s i ) of the sensor s i of delivering the bundles to one of the sinks SUPSI-DTI 70 35

36 Multiple-carrier Selection Each node maintains a list of the neighbouring nodes (including itself) ordered according to their delivery probabilities Each source node then replicates the bundle to the first R nodes (R-1 nodes if the node itself is in the first R positions of the list) Data bundles are only replicated on R nodes by the source, while they are forwarded (deleted from a node and copied on another) if the carriers, while roaming, find either a sink or a better carrier. The value of R is specified by the user and can be considered as a priority level associated to the data retrieved by the sensor SUPSI-DTI 71 Bundle Forwarding Each node keeps monitoring if there are neighbours with better probability of delivery than its own If this is the case the data bundles are shifted to the other nodes Key problem: intelligent replication SUPSI-DTI 72 36

37 Forecasting Techniques for Probabilistic Routing Each node calculates its delivery probability using multicriteria decision theory over the evolution of its context described by a set of attributes. In particular, we consider three attributes: Colocation with the sinks Change degree of connectivity (ie how many nodes have been seen in a fraction of time) Battery level Our aim is to maximize the combined utility function (i.e., selection of the nodes with the best trade-off between these attributes) SUPSI-DTI 73 Prosan design Based on user profiles Using users social behavior node profile - to predict the mobility of nodes. Routing with Zero Knowledge Sender does not need to know any information about other nodes no synchronizing, no local table. Only senders broadcast helloing messages. No sharing user information Nodes does not need to share, periodically update their information to others. Keep privacy amongst members SUPSI-DTI 74 37

38 Forwarding strategy Delivery probability (DP): The probability of a node to meet and forward a message to other nodes toward destination. The sender S wants to send a message M to the destination actuator D, S will choose the route that has the highest delivery probability (DP) to send M to D. Before sending the message content M to the destination D, S sends M s header to its neighbors. n h M = Concatenat ion (H(E Di,V Di ), MAC S, SN M ) (1) Example: i =1 26a48976c47c0d2b2383a34010eae937ebcf8d1f6c554b9f7214a12de02cd9c0c5fd9486e b60f6d2b008 ff1c110c09b62a536a195c27b76b17564e92f25260a3c6599f05fef87f3110b7f171871aef40a2b17fc1bbf88053aa cea724c0a08419a9f211b077bcc244c056817c8c4ca4238a0b923820dcc509a6f75849b SUPSI-DTI 75 Forwarding strategy (cont) Node profile Sender h M B A H(E D1, V D1) H(E D2, V D2) H(E D3, V D3) Evidence names E1 E2 En Message header h M H(E D4, V D4) Value V1 s V2 Vn H(E Dn, V Dn) Hashed H(E1,V1) values H(E2,V2) H(En,Vn) MAC S SN H(E 1,V 1) H(E 2,V 2) H(E 3,V 3) H(E 4,V 4) Hashed values in node profile H(E n,v n) SUPSI-DTI 76 38

39 Forwarding strategy (cont) Neighbors do local matching between received h M with its node profile. Message header h M H(E D1,V D1) H(E D2,V D2) H(E D3,V D3) H(E D4,V D4) H(E Dn,V Dn) MAC S SN H(E 1,V 1) H(E 2,V 2) H(E 3,V 3) H(E 4,V 4) H(E n,v n) With each matching, neighbor nodes profilelook up in T to get the corresponding weight. Then neighbor node (A) can compute its DP A as follows: matched ( Wi) 0 DP A = 1 WMi WMi Hashed values in node : sum of the weights of all the evidences that S knows about D SUPSI-DTI 77 Forwarding strategy (cont) DP DP DP DP Sender DP h M i A 0 DP N = 1 Mi DP B Delivery Probability DP Matched ( W ) W DP S will choose the route(s) with highest DP it receives from its neighbor nodes S only sends the message to these routes if DP route > DP S SUPSI-DTI 78 39

40 Simulation environment Hybrid P-MULEs nodes with higher capacity use Prosan move as MULEs Simulation area rectangle 870x520 meters divided in 6 areas 50 nodes (1 p-mule) use sleep mode Simulation approach static actuator mobile actuator trajectory of p-mule SUPSI-DTI 79 Overhead Simulation results for static actuator P-Mules 1000 PROSAN Hybrid P-Mules Bytes 1 30 Tests Delay Msec P-Mules PROSAN Hybrid P-Mules Tests SUPSI-DTI 80 40

41 Overhead Simulation results for mobile actuator 1400 Bytes P-Mules 800 PROSAN 600 Hybrid P-Mules Tests Delay Msec Tests P-Mules PROSAN Hybrid P-Mules SUPSI-DTI 81 Conclusion context information in opportunistic routing can be very advantageous exploit effectively the mobility, the social connections and the community relationships: reduce significantly the number of nodes involved in reduce the number of broadcasts efficiently control message lifetime assumption for effectively using context information people move in a predictable fashion based on repeating behavioral patterns at different timescales (day, week, and month) or these patterns are known -> mobility as expression of the users social behavior Haggle architecture for supporting context aware routing SUPSI-DTI 82 41

42 Thanks SUPSI-DTI 83 42