Overview Of Wireless Sensor Networks Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Sensor network is an infrastructure comprised of sensing (measuring), computing and communication elements that gives an administrator the ability to instrument, observe and react to events and phenomena in a specified environment. [1]

The four basic components of sensor network are explained in Figure 1.1

Distributed and localized sensors

An interconnecting network

A Central point of information clustering

A set of computing resources at that central point to handle data correlation, event trending, status queuing and data mining

FIGURE 1.1. Components of sensor nodes [1]

Sensor networking is a multi displinery area that involves among others, Radio and networking, signal processing, artificial intelligence, data base management, system architecture for operator friendly administration, resource optimization, power management algorithms and platform technology.[1]


Limited functional capabilities, including problems of size

Power factors

Node costs

Environmental factors

Transmission channel factors

Topology management complexity and node distribution

Standards versus proprietary solutions

Scalability concerns.


Military applications [1][2]

Monitoring enemy forces

Monitoring friendly forces and equipment

Military -theatre and battlefield surveillance


Battle damage assessment

Nuclear, biological and chemical attack detection and more

Environmental applications


Forest fire detection

Flood detection

Precision agriculture

Health applications

Remote monitoring of psychological data

Tracking and monitoring doctors and patients inside a hospital

Drug administration

Elderly assistance and more

Home applications

Home automation

Instrument environment

Automated meter reading

Commercial applications

Environmental control in industrial and office buildings

Inventory control

Vehicle tracking and detection

Traffic flow surveillance


Compared to adhoc networks the number of nodes in a Wireless Sensor Network is several orders of magnitude higher. And these nodes are densely deployed. They are prone to failure. The sensor network topology changes frequently. Adhoc networks are based on point to point communication where as sensor nodes mainly use a broadcast communication.

The main limitation of sensor nodes is their power, computational capabilities and memory. They cannot have global identification because of large number of overhead and large number of sensors.

FIGURE 1.2. Typical Sensor Network arrangement [1]


The scattered sensor nodes have the capacity of collecting data and route data back to the sink. The sink can communicate with task manager node via internet or satellite as explained in the figure 1.2. [1] [2]


Fault tolerance


Production costs

Operating environment

Sensor network topology

Hardware constraints

Transmission media

Power consumption

Fault tolerance [2] : It is the ability to sustain sensor network functionalities without any interruption due to sensor node failures.

Scalability: The number of sensor nodes deployed in studying a phenomenon can be calculated as

u(R) = (NπR2)/A

N = The number of scattered nodes in region A

R = Radio transmission range.

Product costs: Since large number of sensor nodes constitutes the sensor network the cost of each node decides the overall cost of the network. It is suggested that it should be less than US dollar.


The basic components of sensor node

Sensing node: The ADC and sensors constitute this part. ADC is used to convert the Analog signals produced into Digital signals. These signals are again fed to processing unit.

Processing Unit: The sensor nodes have to collaborate with other nodes in order to carry out the assigned sensing tasks. Processing unit manages these procedures.

Transreceiver: It is used to connect the node to the network.

Power unit: This is very important unit. It is supported by solar cells.

Application dependent units:

Location finding system: It is for the knowledge of location for sensor network routing techniques and sensing tasks. It has to be done with high accuracy.

Mobilizer: These are for moving the sensor nodes.

Size: They have to be so small that they can be fit in a matchbox and should be light enough to suspend in the air.


They must consume extremely low power

Operate in high volumetric densities

The production cost must be low

Should be dispensable and autonomous

Should be capable of operating unattended

Should be adaptive to environment.


Millions of sensor nodes are deployed through the sensor field. The different phases are:

Pre deployment and deployment: The sensor nodes can be placed one by one in the field or they can be thrown in as mass. A plane can be used for dropping the nodes or an artillery shell, rocket or missile also can be used. Humans or robots can be used for this purpose.

Post deployment: The change in sensor node position, reachability, available energy, malfunctioning and task details can change the topology.

Re deployment of additional nodes: The malfunctioning nodes are be replaced by additional nodes in this phase.


Sensor nodes are densely deployed inside the phenomenon or very close to it. Usually they work unattended. The environment may be [2]

Interior of large machinery

At the bottom of a ocean

In a biologically or chemically contaminated field

In a battle field

In a home or large building


Wireless medium is used to link the communicating nodes in a multihop sensor network.

Radio, infrared or optical media is used for these links. The medium should be chosen such that it is available worldwide to enable global operation. [2]

Types of Radio used are

Bluetooth compatible 2.4GHz transreceiver with integrated frequency synthesizer.

Single channel RF transreciever operating at 916 MHz

The wireless Integrated Network sensors also uses radio links for communication.

Infrared medium: It is license free and robust to get from electrical devices. They are cheaper and easier to build.

Optical medium: SMART DUST MOTO which is an autonomous sensing, computing and communication system that uses optical medium for transmission. Both infra red and optical medium needs line of sight between the sender and receiver.


The wireless sensor node being a microelectronic device can only contain limited power. Their life time shows strong dependence on battery life time. Each node has the dual role of data origination and data routing in multihop adhoc network. Little malfunctioning of a few nodes can cause significant topological changes and might require rerouting of packets and also reorganization of the network. Hence power consumption and power management are very important in wireless sensor networks.

The main task of sensor node is to detect events, perform quick local processing and transmit the data.

Power consumption can be divided into



Data processing.



The network layer of sensor networks contributes for routing in it. It is designed according to the following principles.

Power efficiency is always important.

Sensor networks are mostly data centric.

Data aggregation is useful only when it does not hinder the collaborative effort of the sensor nodes.

An ideal sensor network has attributed based addressing and location awareness.

Energy efficient routes can be found based on the available power (PA) in the nodes or the energy required (α) for transmission in the links along the routes. [2]

An energy efficient route is selected by one of the following approaches.

Maximum PA route

Minimum energy route

Minimum hope route

Maximum minimum PA route.


Interest dissemination is done to allot the sensing tasks to the sensor nodes. Sinks broadcast the interest. Sensor nodes broadcast an advertisement for the available data and wait for a request from the interested nodes. Data centric routing requires attribute based naming. [2]


Data aggregation can be perceived as a set of automated methods of combing the data that comes from many sensor nodes into a set of meaningful information. With this respect data aggregation is also called data fusion.

Network layer performs one important function of internetworking with external network like sensor networks, command and control systems and the internet. It can be used as gateway to other networks. A backbone can be created by connecting all sink nodes together and this backbone can access other networks via a gate way.


Small Minimum Energy Communication Networks (SMECN)


Gossiping -A derivative of flooding in which nodes do not broadcast but send the incoming packets to a randomly selected neighbor.

Sensing Protocols for Information via Negotiation (SPIN) -Spin has three types of messages. i) Adv ii) Req iii) data. SPIN is based on data centric routing where the sensor nodes broadcast an advertisement for the available data and wait for the request from interested sink

Sequential assignment routing: A set of algorithms that perform organization, management and mobility management operations in sensor networks are proposed. SMACS is a distributed protocol that enables a collection of sensor nodes to discover their neighbors and establish transmission/ reception schedules without the need of central management system.

SAR algorithm: In this algorithm multiple trees are created where the root of each tree is a one hop neighbor from the sink. Each tree grows outward from the sink. It avoids nodes with very low QOS. (Low throughput/high delay) and energy reserves. At the end of this procedure most nodes belong to multiple trees. This permits the sensor node to choose a tree to relay its information back to the sink. The two parameters associated with each path are i) Energy resources ii) Additive QoS metric. SAR selects the path based on the energy resources and additive QoS metric of each path, and packets priority level. During setup phase, a sensor node chooses a random number between 0 and1. If this random number is less than T(n), the sensor node is cluster head.

T(n)= P/1-p*[rmod1/p] if n Ѐ G

= 0 otherwise.

P = desired percentage to become

G = Set of nodes that have not been selected as cluster head in the

Last 1/p rounds

After cluster heads are selected, the cluster heads advertise to all sensor nodes in the

network that they are the new cluster heads. Once the sensor node receives the advertisement, they determine the cluster to which they want to belong that the signal strength of the advertisement from the cluster heads that they will be member of the cluster. Afterwards, the cluster heads assign the time on which sensor nodes can send data to the cluster heads based on TDMA approach.

During the steady phase, the sensor nodes can begin sensing and transmitting data to the cluster heads. The cluster heads also aggregate data from the nodes in the cluster before sending these data to a base station. After a period of time spent on the steady phase, the network goes into setup phase again and enters another round of selecting cluster heads.

Directed Diffusion: Sinks send out interest, which is task description, to all sensors. The task descriptors are named assigning attribute value pairs that describe the task. Each sensor node then stores the interest entry in its cache. The interest entry contains a timestamp field and several gradient fields. As interest is propagated throughout the sensor network the gradients from the source back to the sink are set up. When the source has data for the interest, the source sends the data along the interest's gradient path. Also the sink must refresh and reinforce the interest when it starts to receive data from the source.



Geographical routing [2] [3] uses location information to formulate an efficient route search toward the destination. Geographical routing is very suitable to sensor networks, where data aggregation is a useful technique to minimize the number of transmissions toward the base station by eliminating redundancy among packets from the different sources. [1] Geographical routing only requires the propagation of single hop topology information, like the best neighbor, to make correct forwarding decisions. Its localized approach reduces the need of maintaining routing tables, and hence reduces the control overhead. It does not require flooding. Only nodes that lie within the designated forwarding zone are allowed to forward the data packet. The forwarding region can be defined by the source node or by the intermediate nodes to exclude nodes that may cause a detour while forwarding the data packet. The second property of geographical routing is its position based routing. Here a node requires knowing only the location information of its direct neighbor. The mechanism used is greedy mechanism where each node forwards a packet to the neighboring node that is closest to the destination. The Euclidean distance to the destination is generally used as metric. Position based routing protocols have the potential to reduce control overhead and reduce energy, as flooding for node discovery and state propagation are localized to within a single hop.[1] The network density, the accurate localization of nodes and the forwarding rule decides the efficiency of the scheme.

Forwarding approaches: In position based routing, each node decides the next hop based on its own position, the position of its neighbors, and the destination node. But this local knowledge based forwarding causes suboptimal paths. [as shown in the fig1.4]. In greedy routing the neighbor is selected one that is closest to destination. The selection scheme used in this process considers only the set of nodes that are closer to the destination than the current message holder. If that set is empty the scheme fails. In the most forward within R strategy (MFR), where R represents the transmission range, a node transmits its packet to the most forward among its neighbors toward the destination. The greedy approach is myopic and does not minimize the remaining distance to the destination. In nearest -forward-progress scheme selects the nearest node with forward progress, in which the node currently holding the message selects the nearest node among all its neighbors which are closer to the destination.

FIGURE1.4. Localized and globalized forwarding decision

The compass routing scheme selects the node with the minimum angle between the straight line joining the current node and destination and the straight line joining a neighbor and the destination. The low energy forward scheme selects the node that locally minimizes the energy required, expressed in terms of joules per meter, to progress forward toward the target.

The accurate information about the geographical location of nodes is available from GPS device. But because of the resource and energy limitation of sensor nodes the GPS devices are prohibited. Nodes without GPS devices can use triangularization algorithms for their location determination and also their neighbors. Virtual strategies are used where virtual coordinate systems are defined. In this system each node is given a label that encodes its position in the original network topology in terms of a radius and an angle from a center location. [1]These virtual coordinates do not depend on physical coordinates and can be used efficiently in geographical routing by using node labels.

Even though it is simple, the greedy approach to geographical routing may ether fail to find a path, or produce inefficient routes. In WSN environments, where sensors are typically embedded in the environment or deployed in inaccessible areas, voids are likely to occur. The right hand rule has been used. The rule states that when a packet arrives at a given node Ni from node Nj ,the next hop to be traversed by the packet is the node sequentially counterclockwise from the node Ni with respect to the (Ni, Nj) edge. Perimeter traversal, in which packets are routed along the perimeter of the void, is used.

Combining greedy traversal and perimeter traversal, Greedy perimeter routing algorithm comes into existence. The routing algorithm begins with greedy mode, and when it fails the node records its location in the packet and marks the packet to be in perimeter mode. The perimeter mode packet then follows a simple planar graph traversal.

Geographical routing is characterized by its coordinates. When a node needs to send a packet, it will forward it to the neighboring node that is closer to the destination than all other nodes. All geographical information is kept in its routing table. The information is gathered by sending beacon messages, announcing its location. A Binary Location Index is formulated based on the binary encoded spatial frames for all participating nodes. This is to impart locational aspects in an algorithm in much simpler way and avoids the situation of 'hot spot'.

In the proposed method the entire service area is divided in to four zones and indexed as ( I,II.III,IV).These zones are subdivided into subzones and into regions, sub regions and lastly into grids. The grids are further decomposed into infinitesimal area called cells. After a long duration these Locational Areas and simple nodes reach their specified lowest energy. And it leads to a phenomenon called hot spot. This hot spot effect should be prolonged till the occurrence of maximum expected time (Horizon time). The selection of LAs and sensor nodes should be done likewise, for lifetime maximization of entire network.


The mobility support can be facilitated. Since each node sends its coordinates periodically, all its neighbors update their routing tables accordingly. Thus all nodes aware of its alive neighbor nodes.

It is scalable. The size of routing table depends on network density not on network population. Hence wider networks consisting of thousands of nodes can be realized without cluster formation.

Minimum overheads are introduced. The only information needed is the location of neighbors. Only localized interactions take place. Hence bandwidth is economized. The processing and transmission energy is saved and the dimensions of routing table are decreased.


Geographic routing represents the algorithmic process of determining the paths on which to send traffic in a network, using position information/geographic location only about source, neighbors and destination. It is considered substantially better from an energetic point of view due to the use of solely local information in the routing process. As a result of very little routing information being needed, no energy is spent on route discovery, queries or replies, node memory requirements are decreased and traffic overhead and computation time are considerably reduced. Also, in this sense it is different from source routing in which the sender makes some or all the routing decisions by having mapped the network and specifying in the packet header the hops that the message has to go through. In geographic routing, the process is localized and distributed so that all nodes involved in the routing process contribute to making routing decisions by using localization methods and computing the best forwarding options.


Greedy Perimeter Stateless Protocol (GPSR): Most popular protocol is Greedy Perimeter Stainless Routing protocol. It chooses the path with hop count due to extensive use of geographical coordinates. It reduces the routing overheads. The throughput is shown to be increased. It consists of two parts. One is greedy forwarding and another is perimeter forwarding. In greedy forwarding the source node transmit the packet to the neighbors that is closest to the destination. This will be the optimal choice of next hop. Each node requires knowing the coordinates of the nearest node. This decreases routing table dimension. Its dimension depends on the density of the network and the radio range of the node but not on the dimensions of the network. When greedy forwarding is not possible (if void exists between the node and the destination) perimeter forwarding is performed. This can be done by calculation of planar graphs.

Geographical and Energy Aware Routing (GEAR): Here the next hop is decided taking into account of the available energy of each node. Each node has to inform its neighbors about the level of its remaining energy periodically. Using computer simulations (GEAR) is shown to extend the network lifetime up to 30% compared to GPSR.

Probabilistic Geographic Routing (PGR): Here the approach is the probabilistic forwarding. In order to forward a message, the node selects a set f candidate nodes based on geographical information to guarantee that the packet will be forwarded and not travel backwards. The candidate nodes are then assigned a probability proportional to their residual energy. This algorithm gives high throughput and longer system lifetime but slightly longer path than GPSR

Other pronounced algorithms are Geographical Forwarding using Adaptive Sleeping (EnGFAS), Directional Location based Randomized Routing (DLR), Blind Geographic Routing (BGR).


Sensors establish and maintain routes either proactively or reactively. Proactive protocols periodically monitor peer connectivity to ensure the ready availability of any path amongst active nodes. Sensors advertise their routing state to the entire network to maintain a common (partially) complete topology of the network. On the other hand, reactive protocols establish paths only upon request, e.g. in response to a query, or an event; meanwhile, sensors remain idle in terms of routing behavior. Sensors forward each routing request to peers. Reactive routing protocols to their simplicity, and inherent support for data on demand, they have been the predominant design choice in wireless sensor networks.[7]

The routing protocols incurred by proactive routing are very high. Reactive protocols on the other hand try to delay preparatory actions as long as possible. Here routing occurs on demand only. A node wishing to send a message has flood the network in order to find the destination.[12]


The reactive routing protocols such as AODV do not need to send hello packet to its neighbor nodes frequently to maintain to maintain the coherent between the nodes. And it need not distribute routing information and to maintain the routing information which indicates that the routing links have been already broken. These tables are created when the message needed to be forwarded and nodes maintain this information just for a certain lifetime. When the life time of information is over, nodes discard all these routing and neighbor information. If another message is to be forwarded nodes will create new routing and neighbor information for the next time.[12]


Greedy forwarding makes use of neighborhood beacon that sends a node's ID and its position. However, instead of sending this beacon periodically and adding to network congestion, this approach piggybacks the neighborhood beacon on every message that is ever sent or forwarded by the node. Greedy Perimeter Stateless Routing (GPSR) is the earlier geographical routing protocol. Whenever a message needs to be sent, the GPSR tries to find a node that is closer to destination than itself and forwards the message to that node.

Disadvantages of GPSR: Every packet transmitted in GPSR has a fixed number of retransmits. If the message is unable to be transmitted within these stipulated attempts even in the perimeter mode the protocol will fail.[FIGURE 1.4] GPSR disallows the use of periodic broadcast of the neighborhood beacons, and piggybacks these beacons on the messages sent by each node. This leads to an increase in the message size by 12 bytes, which is a lot in case of resource constrained nodes. The overhead introduced in planarizing the graph ever time the topology changes may render the algorithm unusable in scenarios involving a highly dynamic topology/ link interface between nodes.

Greedy forwarding with success Greedy forwarding failed at node S



This protocol tries to delay the preparatory beginnings as long as possible. Instead of keeping routing table maintenance the routing occurs on demand only in Reactive protocol. A node willing to send a message has to flood the network in order to find the destination. It does not send the 'hello packets' frequently. And the routing information need not be send periodically which saves lot of bandwidth and traffic congestion is avoided. When the life time of message is over nodes can discard all this information. For another message nodes will create new routing and neighbor information.


The overview of the wireless sensor networks gives the basics of Wireless Sensor Networks, the design factors, challenges and hurdles in implementation. The applications of the Wireless Sensor Networks are mentioned. The different WSN protocols are discussed giving importance to geographical protocols. The significance of localization of nodes is discussed and the advantages of geographical routing are explained. Different approaches in forwarding are narrated and the practical constraints about greedy approach are mentioned. And lastly the novel approach in geographical routing 'Reactive Geographical Routing Protocol' is explained.