ROUTING TECHNIQUES IN WIRELESS SENSOR NETWORKS: A SURVEY

Outline  Background  Classification of Routing Protocols  Data Centric Protocols  Flooding and Gossiping  SPIN  Directed Diffusion  Rumor Routing

Background  Sensor nodes  Small, wireless, battery powered  Energy, bandwidth constrained  Data sensing, relaying, aggregating  No global addressing scheme  Sink nodes  More powerful nodes  Usually gateway to wired networks  Data collecting and processing

Goal  So at network layer, it is highly desirable to find methods for energy efficient route discovery and relaying of data from the sensor nodes to the BS so that the life time of the network is maximized.

Routing Protocols

Data-centric Protocols  The ability to query a set of sensor nodes  Attribute-based naming  Data aggregation during relaying  For example:  Flooding & Gossiping  SPIN  Directed Diffusion  Rumor Routing

Flooding & Gossiping  In flooding, sensor broadcasts packets to all its neighbors till dst reached or packets' ttl == 0  In gossiping, sensor sends packets to a randomly selected neighbor which does the same

Flooding & Gossiping(cont.)  Pros  Simple  No routing, no state maintenance  Cons  Implosion  Overlap  Resource blindness  Delay in Gossiping

SPIN – Sensor Protocols for Information via Negotiation  Metadata negotiation done before transmitting the actual data.  3-way handshake: ADV, REQ, DATA  Event-driven

SPIN(cont.)  Pros  Solve the classic problems  Topological changes are localized  Cons  No guarantee on the delivery of data

Directed Diffusion  Sink node floods named “interest” with larger update interval  Sensor node sends back data via “gradients”  Sink node then sends the same “interest” with smaller update interval  Query-driven

Directed Diffusion (cont)  Pros  On demand route setup  Each node does aggregation and caching, thus good energy efficiency and low delay  Cons  Query-driven, not a good choice for continuous data delivery  Extra overhead for data matching and queries

Rumor Routing  A trade-off between Query & Event flooding  An agent, a long-lived packet, is generated when events happen  The agent propagate the event to distant nodes

Rumor Routing (cont)  Pros  Avoid query flooding  Cons  Performs well only when # of events is small  Overhead to maintain agents and event-tables

Hierarchical Routing Protocols  When sensor density increases single tier networks cause  Gateway overloading  Increased latency  Large energy consumption  Clustered Network allow coverage of large area of interest and additional load without degrading the performance

Hierarchical Routing Protocols  Idea  Partition the entire network into regions or clusters.  Select one or more nodes as the cluster head.  The routing is nodes -> cluster head A -> cluster head B -> cluster head C -> nodes  The routing between cluster heads and the routing within a cluster may follow different protocols (EGP-BGP and IGP- OSPF).  Issues resolved: Energy wasted in collision, collision avoidance, idle-listening  Achieves  Better scalability  Removes the load to less powerful nodes

The LEACH Protocol  Low-Energy Adaptive Clustering Hierarchy.  Distributed cluster formation technique that enables self-organization of large numbers of nodes.

LEACH - Setup  Set up phase  Cluster Head (CH) selection (random + rotating)  ADV  Join REQ  TDMA SCH prepared by CH (no collisions and reduced energy consumption)

LEACH - Steady State  Broken into frames, where nodes send their data to the cluster head at most once per frame during their allocated transmission slot.  Once the cluster head receives all the data, it performs data aggregation.

PEGASIS  Power-Efficient GAthering in Sensor Information Systems.  The key idea in PEGASIS is to form a chain among the sensor nodes so that each node will receive from and transmit to a close neighbor.  Well, what was so bad in LEACH except for a bad name...

PEGASIS - Concept  Be Greedy!  Align with the one that has the max signal strength, form a near-optimal chain.  Communicate with neighbors only.  But who takes care of communicating to BS?

PEGASIS - Leader  The main idea in PEGASIS is for each node to receive from and transmit to close neighbors and take turns being the leader for transmission to the BS.  Nodes take turns transmitting to the BS (i mod N node in round i out of N nodes shall transmit to BS).

Passing the buck...  Token passing approach

The TEEN Protocol  Threshold sensitive Energy Efficient sensor Network protocol.  Proactive Protocols (LEACH)  The nodes in this network periodically switch on their sensors and transmitters, sense the environment and transmit the data of interest.  Reactive Protocols (TEEN)  The nodes react immediately to sudden and drastic changes in the value of a sensed attribute.

TEEN - Functioning  At every cluster change time, the cluster-head broadcasts to its members  Hard Threshold (HT) This is a threshold value for the sensed attribute. It is the absolute value of the attribute beyond which, the node sensing this value must switch on its transmitter and report to its cluster head.  Soft Threshold (ST) This is a small change in the value of the sensed attribute which triggers the node to switch on its transmitter and transmit.

TEEN - Hard Threshold  The first time a parameter from the attribute set reaches its hard threshold value, the node switches on its transmitter and sends the sensed data.  The sensed value is stored in an internal variable in the node, called the sensed value (SV).

TEEN - Soft Threshold  The nodes will next transmit data in the current cluster period, only when both the following conditions are true:  The current value of the sensed attribute is greater than the hard threshold.  The current value of the sensed attribute differs from SV by an amount equal to or greater than the soft threshold.

TEEN - Drawback  If the thresholds are not reached, the user will not get any data from the network at all and will not come to know even if all the nodes die-Adaptive Periodic TEEN  This scheme practical implementation would have to ensure that there are no collisions in the cluster.

GPSR: GREEDY PERIMETER STATELESS ROUTING FOR WIRELESS NETWORKS

Presentation Overview  Introduction  Algorithm Key Ideas & Concepts  Examples  Evaluation Matrices & Results  Summary

GPSR Overview  Routing Protocol that uses positions of routers and a packet’s destination to make packet forwarding decisions  GPSR keeps state only about the local topology (for only a single hop)  The word “Stateless” is not meant literally but refers to small, purely local state  GPSR scales better in per-router state than shortest- path and ad-hoc routing protocols  GPSR finds correct new route quickly under frequent topology changes

Algorithm’s Key Ideas  “The position of a packet’s destination and positions of the candidate next hops are sufficient to make correct forwarding decisions, without any other topological information.”  “Routing protocol that rely on end-to-end state concerning the path, face scaling challenge with increasing number of routers & rate of change of topology (mobility).”  “GPSR generates routing protocol traffic independent of the length of the routes through the network, therefore generates a constant, low volume protocol messages as mobility increases,”

GPSR Modes  Modes  Nominal: Greedy Forwarding  Special : Perimeter Forwarding  Greedy Forwarding  Uses only information about router’s immediate neighbors  Forwarding node makes locally optimal greedy choice of the packet’s next hop

Greedy Forwarding Example Follows successive closer geographic hops, until the destination is reached

Perimeter Forwarding  Used in the region where Greedy forwarding fails.  There are topologies in which the only route to a destination requires a packet move temporarily farther in geometric distance from the destination  Special mechanism (right hand rule) is used in such special situation

Perimeter Forwarding Example Right Hand Rule: x receives a packet from y, and forwards it to its first neighbor counterclockwise about itself, z When Greedy Forwarding Fails. x is a local maximum in geographic proximity to D w and y are farther from D

Perimeter Forwarding Example D is the destination; x is the node where the packet enters perimeter mode; forwarding hops are solid arrows; the line xD is dashed. GPSR forwards packets along the face intersected by the line xD x forwards the packet to the first edge counterclockwise about x form the line xD, follows right hand rule thereafter

Network Graphs Full Graph Greedy Forwarding Planar Graph Perimeter Forwarding

Planarized Graphs The RNG graph. For edge(u; v) to be included, the shaded lune must contain no witness w. The GG graph. For edge(u; v) to be included, the shaded circle must contain no witness w.

GPSR Operation  GPSR combines greedy forwarding on full network graph and perimeter forwarding in planarized network graphs  All nodes maintain neighbor table, which stores the address and location of their single-hop radio neighbors  The table provides all states required for forwarding decisions

GPSR Operation  All packets are marked initially as greedy mode  Upon receiving a greedy-mode packet, a node searches its neighbor table geographically closest to the packet’s destination.  If the neighbor is closer to the destination, it forwards the packet to the neighbor.  If no neighbor is closer, the packet is marked into perimeter mode.

GPSR Operation  In perimeter mode, the packet is forwarded using a simple planner graph  The packet is forwarded to progressively closer faces of the planner graph using the right hand rule  In perimeter node, the GPSR records the location where greedy failed, and the first edge a packet crosses on a new face  When the destination is not reachable, the packet will tour unsuccessfully around the entire face as no intersection with xD is detected  Upon traversing on the face second time, the repetition is detected to correctly drop the packet

Algorithm Evaluation Matrices  Packet delivery success rate  Fraction of applications’ packets delivered successfully by the routing algorithm  Routing protocol overhead  Per-node state: storage required at each node  Protocol message cost: number of protocol packets sent by the routing algorithm  Optimality of path lengths taken by data packets

Packet Delivery Success Rate GPSR with varying beacon intervals, B, compared with Dynamic Source Routing (DSR). 50 nodes GPSR delivers a slightly greater fraction of packets successfully than DSR Pause time is the time duration for which all nodes hold the same positions at waypoints. The mobility model used is the random way point model which generates waypoints at random.

Routing Protocol Overhead Total routing protocol packets sent network-wide during the simulation for GP R with varying beacon intervals, B, compared with DSR. 50 nodes. GPSR offers greater savings in routing protocol overhead

Packet Delivery Success Rate Packet Delivery Success Ra t e. F o r G P S R with B = 1.5 compared with DSR. 50, 112, a n d 200 node s. GPSR delivers 97% of its packets along optimal-length paths vs. 84.9% for DSR

Routing Protocol Overhead Total routing protocol packets sent network-wide during the simulation for GPSR with B = 1.5 compared with DSR. y axis log-scaled. 50, 112, and 200 nodes. GPSR generates drastically less routing protocol traffic w.r.t. to DSR

Summary  GPSR routing algorithm uses geography to achieve small per-node route state, small routing protocol message complexity, and extremely robust packet delivery  GPSR generates routing protocol traffic independent of the length of the routes through the network, therefore generates a constant, low volume protocol messages as mobility increases  GPSR performs better than DSR. GPSR keeps states proportional to no. of neighbors; DSR keeps sates proportional to product of no of routes learned and route length in hops.  Besides Hierarchy and Caching, Geography is leverages for scaling routing

 Thank you