1. Motivation for Mobile Computing

2. Motivation Cellular phones & pagers Global positioning System Cordless Computer Peripherals Cordless Telephone Sets Satellite Television

3. Protocol stack issues in mobile computing Physical Layer Datalink layer Network Layer Transport Layer Middleware and Application Layer

4. Mobility Issues in Mobile Computing The near by station is responsible for handling node’s transmission. Then the address changes according to the base station The changes of node’s address is determined by routing protocols The data communication used by node depend on its location Load balancing is carried out by transferring to another node which is idle to improve the performance of the system.

5. Mobile Access to Resource In order to access resources of the mobile they need to have a name – Discovery services: Uses location of the address – Directory services: Maintains a list of available resources which are registered to directory. It allows browsing and searching of all resources which are registered.

6. Data Dissemination Data dissemination means distributing and pushing data generated by a set of computing systems or broadcasting data from audio, video and data services. The output is sent to mobile devices. Then the mobile device can select, tune and cache the required data items. Mobile devices can participate in one or more distributed computing systems

7. Cache Management It is the technique to reduce traffic on the wireless channel and to speed up the access. It places the data on the mobile unit (MU). This also causes the cache maintenance because the user uses the cached data repeatedly without sending messages to retrieve it. Cache Maintenance Semantic Data caching

8. Cache Management Cache Maintenance – This prone to disconnections and involves periodic broadcast of a cache invalidation reports by the database server. – These reports contains a list of recently updated data items. – Communication costs can be reduced by applying the caching techniques for “query processing, communication cost etc.,” – The techniques are “pre fetching, replacement strategies and consistency of the cache memory” which are used in combination with broadcasting techniques

9. Cache Management Semantic Data Caching: – The cache contents depends upon the results of earlier transactions or by the semantic data set. – The client maintains semantic description of data in its cache instead of maintaining a list of pages or tuples – The server processes simple predicates on the database and results are cached in the client

10. Data DeliveryPoint – to - PointBroadcast Push based Data Scheduling / Delivery (Publish Subscribe) - Float Based - Probabilistic based broadcast - Broadcast disks (B disk) - Optimal push scheduling Pull based data scheduling / Delivery (On Demand) - On demand scheduling for equal size items - On demand scheduling for variable sized Items - Energy efficient scheduling Push pull data Delivery (Hybrid) - Balancing push pull - Adaptive Hybrid Broadcast

11. Data Dissemination Strategies Point to point Broadcast system Push based data scheduling / Delivery Pull based data scheduling / Delivery Hybrid data delivery / scheduling

12. Data Dissemination Strategies Point to point – A logical path created between source server& the destination client. The server and client exchange the data upon the request of the client Mobile Client Mobile Server

13. Data Dissemination Strategies Broadcast system – Data delivered to all the clients resides in a particular range of transmission area – Ex: Radio or TV broadcast Mobile Server Mobile Client 1 Mobile Client 2 Mobile Client 3 Mobile Client 4

14. Data Dissemination Strategies Broadcast system (Advantages) – Several client requests can be served simultaneously by performing a single broadcast of data items – The download link channel with high capacity than uplink channel helps the wireless network for asymmetric communication. So this channel helps to broadcast data items to the clients – No additional cost is required for its implementation

15. Data Dissemination Strategies Push based Scheduling / Delivery – The data will be explicitly pushed or enters the clients, where as the clients may or may not accept the data. Ex: TV, Radio – Several data scheduling algorithms used Flat broadcast Probabilistic based broadcast Broadcast disks (Bdisk) Optimal scheduling

16. Data Dissemination Strategies Pull based data scheduling/ Delivery – The client makes an explicit request for the data i.e., the client pulls the data from the channel. – The client pulls the data from the channel. The client uses a low bandwidth uplink channel to demand data from broadcast server. Which is known as on demand data delivery mode. The algorithms used are On demand / Pull based scheduling for equal size items On demand / Pull based scheduling for variable size items Energy efficient scheduling

17. Data Dissemination Strategies Hybrid data delivery / Scheduling – It is a combination of both push and pull data delivery methods and also called as pushpull data delivery – The methods to implement are Balancing push / pull data Adaptive hybrid broadcast

18. Security Issues in Mobile Networks Battery power Limited Memory Less powerful CPU High latency due to frequent on /off states

19. Security attributes & Requirements Confidentiality Authentication Integrity Authorization Auditing Non repudiation Access Control

20. Mobile Adhoc Networks (MANETS) It is a self configuring network of mobile nodes each has a router which are connected through various links forming an arbitrary topology. It is autonomous and compromises of routing nodes which can move freely by network topology discovery and message delivery. The nodes in MANET generates the user application traffic and controls routing protocols.

21. Mobile Adhoc Networks (MANETS) Problems raised from MANETS – Wireless medium which posses the sharing nature – Limited wireless connectivity range – Mobility of nodes – Energy constraints

22. Mobile Adhoc Networks (MANETS) The applications are ranged from small static networks to large dynamic networks. MANETS need more efficient algorithms to schedule and network routing – Wireless link nature – Propagation path loss – Multiuser interference – Power consumed – Fading – Frequent changes in topology

23. Applications of Ad Hoc Networks(MANETS) Used to transfer data in local groups where the mobility of nodes is high Used in emergency operations – Ex: Earth quake, Sending information to the victims about environment etc., Used to allow business conferences anywhere any time without a fixed infrastructure Used in Military operations to communicate between fast moving objects – Ex: Military troops and fighter planes Continued…

24. Used in real time video streaming Used in hybrid wireless architecture(HWA) – Ex: Integrated Cellular Ad Hoc Relay (ICAR) – Multi hop Cellular Networks (MCN) Used in mobile or fixed nodes of users by providing an alternate communication infrastructure.

25. Due to non centralized control MANETS have following issues to handle during the implementation of MANETS applications -Dynamic Topologies -Energy constrained Operations - Bandwidth constrained, variable capacity links - Limited physical security

26. -Dynamic Topologies A BD A E C A C B D E BE D Source Sink Source Sink SourceSink

27. -Dynamic Topologies: The nodes must be dynamic, and they connect or disconnect from the network at any time. The connections may be unidirectional or bi-directional. The above diagram shows that node ‘C’ is disconnecting from the network and ‘E’ is ready to connect. -Energy Constrained Operations: The transmission range is depends upon the battery power of the nodes. By increasing the transmission power of the node it may get connected to or disconnected from the network.

28. -Energy Constrained Operations: A BD E C A C BD E LowHigh

29. -Bandwidth constrained, Variable Capacity Links: The wireless links have -Limited bandwidth -Low capacity -Low rate -Low throughput -Exposed to noise -Subject to interference

30. -Limited Physical Security: The following security threats may cause in MANETS -Eves dropping -Spoofing -Denial of service

31. Issues in MANETS Mobility management Location Management QOS Management Routing Issues Transport Layer Issues Security Issues MAC Issues Cache Management

32. Issues in MANETS Mobility management: The nodes may change their operations from one hop to the other in MANETS, because of this the address of the nodes may change dynamically in Ad Hoc Networks. Which are too much than compared in case of Mobile IP. Temporary alternatives are provided by Mobile IP. Location Management: By using location discovery a specific mobile node is identified

33. Issues in MANETS QOS Management: The quality of service helps to resolve the service issues raising in wireless Ad Hoc Networks. The extensive usage of the VOIP and rapid growth in multimedia applications over the internet the QOS management became a mandatory task.

34. Issues in MANETS Routing Issues: Routing in the MANETs is a crucial issue because of the following reasons – Presence of Gateways – Multi-hop relaying – Dynamic topological changes – Hybrid network characters These issues can be resolved by introducing routing protocols which are mentioned earlier.

35. Issues in MANETS Transport layer issues: The TCP extensions defined for Ad Hoc networks helps to provide suitable solution in this layer Security Issues: Because of the broadcast behavior the information in the MANETS may be accessible to the potential hackers. Especially in case of data transfer in online transactions. Due to this several security measures are introduced for MANETS

36. Issues in MANETS MAC Issues: Based on the nature of MANETs the nodes are identified according to – Presence of mobility – Hidden and exposed nodes difficulties The various nodes in the Ad Hoc networks share a common media. With the help of MAC protocol the common channel should be accessed in the distributed manner. No other node is depends upon the central coordinator because they are not static.

37. Issues in MANETS Cache Management: The consistent cache is a must in this distributed data management. The cache management helps the nodes in attaining the distributed execution of tasks, result aggregation and reporting.

38. MAC Layer Issues Link breakage when distance increases, due to fall in receive signal power. Large amount of bandwidth is consumed for rerouting, degrading throughput. Collisions occur due to interference between nodes. Long backoff or frequent backoffs consume resources.

39. Routing Protocols Flooding – route discovery message by source is forwarded to all nodes except the sending node. Requirements of routing algorithms: Every node is a host as well as router, algorithms with high processing and memory requirements cannot be used. Routing should not be completely centralized or completely distributed.

40. Destination Sequenced Distance Vector Routing(DSDV) Based on Bellman Fords routing, with some improvements. Guarentees loop free pathsusing sequence numbers. Each node maintains routing table containing entries - Next hop toward each destination - Number of hops required to reach the destination - A destination sequence number that is created by destination, Each node appends sequence number and sends local routing table to neighbours. Neighbours listen and update information for node along with hop count. If there is previous entry for X, entry having larger sequence number is stored. This means, new information has arrived from a node. Drawback is periodicity of updation, settling time for destination etc., If routes are updated too fast, it is communication overhead. If they are updated too slow, routing information may be stale.

41. Multihop capability, loops must be prevented. Due to frequent path breaks, algorithms with faster route maintenance must be used. In MANETs some routes get congested while some are underutilized. Hence the need for balancing load distribution among different routes.

43. Proactive or table driven: Every table needs to maintain complete routing table of the network. Disadvantages: Routing table continuously updated, over head on network Advantage lower latency o route discovery since routes are available in tables. Eg., DSDV, FSR

44. FSR FSR is functionally similar to LSR in that it maintains a topology map at each node. The key difference is the way in which routing information is disseminated. In LSR, link state packets are generated and flooded into the network whenever a node detects a topology change. In FSR, link state packets are not flooded. Instead, nodes maintain a link state table based on the up-to-date information received from neighboring nodes, and periodically exchange it with their local neighbors only (no flooding). Accuracy of route increases as the packet comes closer to destination., hence the name.

46. Each node selects a set of its neighbor nodes as “multipoint relays” (MPR). In OLSR, only nodes, selected as such MPRs, are responsible for forwarding control traffic, intended for diffusion into the entire network. MPRs an efficient mechanism for flooding control traffic by reducing the number of transmissions required.

47. HELLO messages are broadcast to nodes, allowing them to lean topology upto 3 hops. Node sends topology control messages to all neighbours in the MPR set. A node processes all received TC messages, but forwards them only to those in the MPR set. Each node maintains a topology table based on link information from the TC messages. OLSR has very good performance.

48. Reactive routing Schemes Dynamic Source Routing: Outline ❒ Introduction ❒ Route Discovery ❒ Route Cache ❒ Route Maintenance ❍ Preventing Route Reply Storms ❍ Route Request hop limits ❍ Packet Salvaging ❍ Automatic Route Shortening ❍ Increased spreading of Route Error messages ❒ Summary

49. Route discovery ❒ When node S wants to send a packet to node D, but does not know a route to D, node S initiates a route discovery ❒ Source node S floods Route Request (RREQ) ❒ Each RREQ, has sender’s address, destination’s address, and a unique Request ID determined by the sender ❒ Each node appends own identifier when forwarding RREQ

50. DSR – Route Discovery 1. Node S needs a route to D 2. Broadcasts RREQ packet 3. Node A receives packet, has no route to D ❒ Rebroadcasts packet after adding its address to source route

51. ❒ Upon receiving a RREQ, the node takes the following actions: 1. The node is the Target (Destination) ❍ Returns a Route Reply (RREP) message to the sender ❍ Copies the accumulated route record from RREQ into RREP ❍ Sender upon receiving RREP, caches the route in its route cache for subsequent routing

52. 2. The node is the intermediate node ❍ The node discards this message, if 1. The message has the same ID i.e. has seen it before OR 2. Finds its own address in the route record ❍ If Not, The node appends its own address to the route record in the ROUTE REQUEST message 3. Propagates the message to the next hop neighbors

53. 4. Node C receives RREQ, has no route to D ❍ Rebroadcasts packet after adding its address to source route

54. 4. Node C receives RREQ, has no route to D ❍ Rebroadcasts packet after adding its address to source route 5. Node D receives RREQ, unicasts RREP to C ❍ Puts D in RREP source route

55. 4. Node C receives RREQ, has no route to D ❍ Rebroadcasts packet after adding its address to source route 5. Node D receives RREQ, unicasts RREP to C ❍ Puts D in RREP source route

56. 6. Node C receives RREP ❍ Adds its address to source route ❍ Unicasts to A

57. 7. Node S receives RREP ❍ Uses route for data packet transmissions

58. When node S sends a data packet to D, the entire route is included in the packet header ❍ hence the name source routing DSR Optimization: Route Caching ❒ Each node caches a new route it learns by any means ❒ When node S finds route [S,E,F,J,D] to node D, node S also learns route [S,E,F] to node F ❒ When node K receives Route Request [S,C,G] destined for node D, node K learns route [K,G,C,S] to node S ❒ A node may also learn a route when it overhears Data packets

59. Use of Route Caching ❒ When node S learns that a route to node D is broken, ❍ Can use another route from its local cache, if such a route to D exists in its cache. ❍ Otherwise, node S initiates route discovery by sending a route request ❒ Node X on receiving a Route Request for some node D can send a Route Reply if node X knows a route to node D ❒ Use of route cache ❍ can speed up route discovery ❍ can reduce propagation of route requests

60. An example of route caching

61. Route maintenance Monitors the route and informs the sender of any routing errors due to link failures. When an RERR packet is received, the hop in error is removed from the hosts cache, all routes containing the hop are also deleted. RREQ packets are forwarded using flooding. One drawback may be cache pollution due to stale routes which may propagate to other caches.

62. AODV Discover routes only when requested by source node. Supports both unicast and multicast routing. Nodes on receiving RREQ packets set up backward pointers to the source in the routing tables.

63. Path discovery The RREQ contains the following fields: When an intermediate node receives an RREQ, if it has already received an RREQ with the same broadcast_id and source address, it drops the redundant RREQ and does not rebroadcast it.

64. Reverse-Path Setup There are two sequence numbers (in addition to the broadcast_id) included in an RREQ: the source sequence number and the last destination sequence number known to the source. The source sequence number is used to maintain freshness information about the reverse route to the source, and the destination sequence number specifies how fresh a route to the destination must be before it can be accepted by the source.

65. If the RREQ’s sequence number for the destination is greater than that recorded by the intermediate node, the intermediate node must not use its recorded route to respond to the RREQ. Instead, the intermediate node rebroadcasts the RREQ. The intermediate node can reply only when it has a route with a sequence number that is greater than or equal to that contained in the RREQ. An RREP contains the following information:

66. The forward path setup as the RREP travels from the destination D to the source node S. Nodes that are not along the path determined by the RREP will time out after ACTIVE_ROUTE_TIMEOUT (3000 milliseconds) and will delete the reverse pointers.

67. Associativity based routing Route is selected based on associativity of nodes. Nodes periodically broadcast L am alive beacons to neighbours. Threshold associativity is used. Eg., in a LAN of size 10m, mobile node moving at 10 m/s, associativity threshold is 5 ticks. Disadantage is, if a route breaks due to link failure, discovering partial route results in high latency and wastage of bandwidth

68. Zone Routing Protocol Network is divided into zones. node knows its physical location using GPS. When a node has packets to send to a particular destination, it checks it routing table. All nodes within at most d hops from the node X are said to be in routing zone of node X. All nodes exactly d hops are said to be peripheral nodes of X’s routing zone.

69. Intrazone routing protocols (IARP) and Interzone routing protocol (IERP) are used. IARP used for routes with short distances from any node – using proactive distance vector or link state routing. IERP is on demand used for nodes which are far away. Zone based hierarchichal link state protocol, each node has node link state packets and zone link state packets. Each node asynchronously broadcasts link request and nodes reply with responses. Disadvantage is it requires more memory.

70. Location aided routing It is a reactive routing protoco tha uses geographical coordinates to direct RREQ messages to a previously known location of the destination. The expected zone is the area in which the destination is most likely to be discovered. It is calculated based on previous locaion of destination at time t0 as well as the estimate of the velocity v with which the destination wa travelling at t0. if current time is t1, the expected zone can be calculated as a circle with radius v(t1-t0) centered at an earlier location of destination.

71. RREQ messages by sender are forwarded by nodes within the zone and are discarded by nodes outside the rectangular zone(request zone).

72. Transport layer issues in MANETs When packets are lost due to time out or due to duplicate acknowledgements, TCP slows down the sending rate by adjusting its congestion window. However, TCP cannot provide reliable service in MANETs because, it misinterprets packet loss as loss due to congestion. Causes of packet los - Signal attenuation, multipath fading, Doppler shift etc., Moreover, node mobility causes disconnections

73. In exponential backoff algorithm, if a network remains disconnected for more than Retransmission time out(RTO) of a particular node, an exponential back off algorithm is triggered, which doubles RTO whenever the timeout expires. Normal TCP is expected to measure RTT and hence RTO accurately. However, in a multihop wireless environment, RTT is not very accurate.

74. Due to variation in bandwidth, whenever TCP increases congestion window and sends more packets, chances of collision increases. Effect of longer paths also has oscillatory effect, due to random link failures which also effects TCP throughput. TCP mechanisms for Adhoc networks must be designed in collaboration with the MAC and the network layer. Eg., when network layer detects a route failure, it should inform transport layer about it. TCP may in response freeze all its values including its congestion window and transmission timer.

75. Threats in Adhoc networks Wireless signals are easier to eavesdrop compared to wired signals. Nodes move randomly and independently of each other, making static security solution inadequate. Processing capability, energy in mobile nodes are limited and require energy conserving security mechanisms. Lack of central authority and cooperation of all nodes makes chances for attacks by malicious nodes. Battery power of mobile nodes is exhaustable and attacker can replay false packets, leading to exhaustion of energy at the nodes. False routing information also reduces network lifetime.

76. Security attacks in MANETs Passive attacks: Attacker does not disrupt the operation of the network, but eavesdrops to discover valuable information by listening to routing traffic – both data and control packets. He uses the captured information to find which protocol is being used, location of nodes and so on. These attacks are very difficult to detect and to prevent.

77. Security attacks in MANETs Active attacks: The attacker aims to modify routing tables, propagate false routing information, gain access to routing information and use to disrupt the network. Further classified into external, internal attacks depending on whether the attacking node belongs to the network or does not.

78. Active attacks types Packet dropping: Attacker disrupts network by selectively or completely dropping the packets. In blackhole attacks, both data and control packets are dropped eg., by declaring a better route through the malicious node. Spoofing attack – impersonation creating routing loops, route disruption.Grey hole attack – selectively drops data packets, allows control packets.

79. Active attacks types Modification of Protocol message fields: Intercept, filter routing protocol packets, modify and inject packets. Eg., AODV malicious nodes may divert traffic towards itself. In DSR, routing loops are created by modifying source route.

80. Active attacks types Rushing attacks: Attacker rushes RREQ packets towards destination, preventing broadcast to neighbouring nodes. Fabrication attacks: Generation of false packets: Route deletion Route poisoning Buffer overflow

81. Wormhole attacks: attacker nodes create tunnels between nodes. Attacker receives packets from one end of tunnel and forwards them to another network. Packets are there by relayed to another path Eg., in AODV “HELLO” packets may be tunneled.

82. Security Principles Uses key management services for binding between keys belonging to nodes. Establishing trust: distributed trust models are used using distributed public key cryptographic models to distribute private keys of certification authority over number of servers using trusted Certification Authority(CA). Information from any (t+1) servers can be combined out of n in nCt+1 ways to create a complete secret key.

83. Watchdog and pathrater Watchdog detects malicious nodes by overhearing transmissions of one hop neighbours. Also maintains buffers of recently transmitted packets. Pathrater gives rating for every node 0 to 1 in a network. Rating of a path is the average of rating of all nodes in the path. Path with highest rating is chosen for routing.

84. Intrusion Detection Misuse detection – Uses patterns of well known attacks or weakspots. However only works for well known attacks Anamoly detection systems- Observe significant deviations from normal usage profiles. However some attacks cannot be detected by localized monitoring and may need digital signatures, but this method is inefficient and is a challenge to be addressed.

85. Wireless Sensor Networks Consist of tens or hundreds of small power constrained nodes deployed in remote locations which they are expected to monitor for months or years in time. Applications – traffic control, biomedicine, hazardous environment exploration, environmental monitoring, military tracking Earthquake response systems, learning environments, intelligent battle fields Recent advances – low cost low power multifunctional sensor devices.

86. Wireless Sensor Networks functioning Each node is equipped with one or more sensors, a radio transceiver, a small microcontroller, operating system and a battery. Size and cost constraints on energy, memory, computational speed and bandwidth. Devices are commonly called motes. Motes cooperate to form a sensor network. WSNs are capable of collecting, processing and storing data from sensors.

87. MEMS micro electromechanical systems is a technology to miniaturize components like motes. In a WSN setup, nodes automatically setup topologies with a multihop communication. Data collected by sensor nodes are aggregated and forwarded to the sink. New nodes may be added to nodes removed seamlessly without effecting reliability of service.

88. OS support in sensor devices OS is less complex due to constraints and special requirements since sensor network applications are not interactive as applications for PCs, they do not need support for UIs. Virtual memory is unnecessary or impossible to implement due to constraints. Examples of OS for WSN – TinyOS, Contiki and Nano-RK.

89. WSN characteristics Nodes must be able to accept queries from remote sites, interact with physical environment, actuate response in sensor readings and relay sensed information through sensor networks. WSNs must be densely deployed from thousands to millions of sensing devices. Energy conservation is the center of focus due to limited capacity and impossibility of recharging

90. WSN characteristics Used in real time applications such as in military and nuclear power plant management or fire fighting where deadline violations due to processing or transmission of collected data can cause catastrophic events. Collected, delivered data must be valid at the time of action. Protocols proposed for MANETs are not suitable for WSNs.

91. WSN characteristics Data integrity and reliability are also in addition to real time communication are also an issue. To summarize, characteristics of WSN are: Wireless transmission - Heterogenity Limited power of nodes Self-configuration - Large scale of Mobility of nodes(limited) deployment Dynamic topologies - Unattended operation

92. Sensor Network Operation Sensor network operation: Five phases in sensor network deployment: Planning Deployment Post-deployment Operation phase Post-operation phase

93. Sensor Network Operation Design issues in sensor networks: MAC protocol issues. Routing protocol issues Data dissemination Location and management issues Query processing Key distribution and security measures

94. Sensor architecture – cluster management Clustering organizes sensor nodes into groups, so that sensor nodes communicate information only to cluster heads(CHs) and CHs communicate the aggregated information to the processing center. Clustering saves energy of sensor nodes. Examples of clustering algorithms are LEACH and time controlled clustering algorithm(TCCA).

95. Sensor architecture – cluster management A sensor node is chosen as cluster head by running leader election algorithm. Each round consists of setup and data communication phase. Setup-clusters are formed using a distributed algorithm In data communication phase, data gathered in CHs from nodes are routed to the base station for forward journey.

96. Wireless Mesh Networks IEEE 802.11(WLANs) require several access points, each of them to be connected to a wired network. Very costly. WiMAX 802.16e-2005 provides broadband wireless access to users in metropolitan areas. A WMN consists of mesh routers and mesh clients. Mesh routers that connect to Internet are called gateways.

97. WMNs are dynamically self-organized and self configured with nodes automatically forming adhoc network and maintaining mesh connectivity. Datarate of Mesh Networks limited by implementing technology – 54 Mbps for IEEE 802.11b/g, 600Mbps using IEEE 802.11 b/g.

98. \ WMN makes use of mesh routers for multihop communication to extend coverage area. Mesh routers are also equipped with multiple wireless interfaces. Mesh routers provide broadband wirless ervices like VOIP, multimedia at high data rates to home and office users.

99. Network size in WMNs is very large and power consumption at nodes is very less as they are stationary.