CHAPTER 2 LITERATURE SURVEY - Shodhgangashodhganga.inflibnet.ac.in/bitstream/10603/8869/7/07_chapter 2.pdf · Chapter 2. Literature Survey 24 transmission from sensor nodes to sink

CHAPTER 2

LITERATURE SURVEY

2.1. Introduction

Smart environments represent the next evolutionary development step in building,

utilities, industrial, home, shipboard, and transportation systems automation. Like any

conscious human being, the smart environment relies first and foremost on sensory

data from the real world. Sensory data comes from multiple sensors of different

modalities in distributed locations. The smart environment needs information about its

surroundings as well as about its internal workings; this is captured in biological

systems by the distinction between exteroceptors and proprioceptors.

With respect to the performance of wireless sensor networks, the data transmission

capacity and the lifetime of the sensor networks are critical and influential towards the

design of optimal deployment strategies of these sensor networks. The fundamental

limits of these two critical performance parameters lead to a few interesting open

problems. First, what is the maximum sustainable throughput of the network? Second,

what is the maximum lifetime of the network? These questions are usually considered

given a set of parameters of the sensor network, and under the assumption that

optimal network management is achievable. The set of parameters of the sensor

network under consideration includes the number of sensor nodes in the network, as

well as the area occupied by the sensor network. Issues relevant to network

management usually include packet routing, energy management, and congestion

control, which directly affect to quality of service. One of the earliest routing

protocols with Qualty of Service (QoS) impression is Sequential Assignment Routing

(SAR) [4]. SAR creates trees originating from one-hop neighborhood of sink and

takes into account two types of QoS metrics; energy resource and priority level of

each packet Multiple paths are created from a sink to source and a path may be

selected based on the QoS metrics. However, SAR is believed to suffer from overhead

for maintaining the node state. Zhi Ang Eu et al. [14] study the performance of

different medium access control (MAC) schemes based on CSMA and polling

techniques for WSNs which are solely powered by ambient energy harvesting using

energy harvesters. the study is based on (i) network throughput (S), which is the rate

Chapter 2. Literature Survey

21

of sensor data received by the sink, (ii) fairness index (F), which determines whether

the bandwidth is allocated to each sensor node equally and (iii) inter-arrival time (γ)

which measures the average time difference between two packets from a source node.

For CSMA, they compare both the slotted and unslotted variants. For polling, they

first consider identity polling. Then design a probabilistic polling protocol that takes

into account the unpredictability of the energy harvesting process to achieve good

performance. Finally, they present an optimal polling MAC protocol to determine the

theoretical maximum performance. They validate the analytical models using

extensive simulations incorporating experimental results from the characterization of

different types of energy harvesters. The performance results show that probabilistic

polling achieves high throughput and fairness as well as low inter-arrival times. Ren-

Shiou Liu et al. in [19] proposed a low-overhead MAC layer solution to address the

high contention problem to improve system throughput and reduce energy

consumption. Periods of burst transmissions with reduced contention from

neighboring nodes are exploited to efficiently clear up backlogged queues and

improve the performance of CSMA. Through analytical modeling they characterize

the expected performance improvement. Using extensive simulations on ns-2 and

experiments on the 49-node sensor network test bed (Kansei) running TinyOS it is

seen that the proposed scheme can increase the throughput by up to a factor of four.

Zhi Ang Eu et al. [53] designed a probabilistic polling protocol that takes into account

the unpredictability of the energy harvesting process to achieve good performance.

They presented an optimal polling MAC protocol to determine the theoretical

maximum performance. Validation occurred with the analytical models using

extensive simulations incorporating experimental results from the characterization of

different types of energy harvesters. The performance results showed that

probabilistic polling achieves high throughput and fairness as well as low inter-arrival

times.

2.2. Quality of Service in WSNs

Although the subject of WSNs is an active research field for the past few years, QoS

has largely been an unexplored area. QoS in a network is largely associated with

network data rate, energy efficiency in the network, and with network lifetime.

However, with the growing applications of WSNs, quality of service is now deemed

necessary to be implemented. At the network routing level, a few protocols have been


22

proposed so far. Much of the other QoS related work has focused on designing an

optimal number of sensor nodes from which a sink would ask for data information.

The approach is based on Gur Game theory, taking into consideration delays &

addition and removal of sensor nodes. This approach has been studied in [7] [8]. M.

Aykut Yigitel et al. [13] focuses on the QoS support at the MAC layer which forms

the basis of communication stack and has the ability to tune key QoS-specific

parameters, such as duty cycle of the sensor devices. They explore QoS challenges

and perspectives for wireless sensor networks, survey the QoS mechanisms and

classify the state of the art QoS-aware MAC protocols together with discussing their

advantages and disadvantages. According to this survey, they observe that instead of

providing deterministic QoS guarantees, majority of the protocols follow a service

differentiation approach by classifying the data packets according to their type (or

classes) and packets from different classes are treated according to their requirements

by tuning the associated network parameters at the MAC layer. Hyun Jung Choea

etal. [28] presented an efficient data reporting control scheme in a cluster-based

hierarchical wireless sensor network, which has two components (i) intra-cluster data

reporting control (IntraDRC) scheme and (ii) inter-cluster control (InterDRC) scheme.

The IntraDRC scheme controls the amount of traffic generated in a cluster by

selecting a certain number of data reporting nodes based on the desired throughput

specified by the end system. On the other hand, the InterDRC scheme offers

differentiated reporting paths from a cluster to a sink based on the traffic

characteristics. InterDRC considers two parameters: one is the hop counts to a sink to

deal with the end-to-end delay constraint while the other is the amount of traffic,

generated in a cluster and forwarded from its adjacent clusters, to deal with energy

consumption. The proposed scheme applies the block design concept from

combinatorial theory to design a novel data reporting node selection approach.

IntraDRC employs the node sets created by block designs as the initial reporting

schedule. This schedule can be updated by the request of a reporting node when its

queue size approaches a predefined threshold. They considered two network models

in their work, the first model considers homogeneous networks in which every node

has the same capabilities and adjacent cluster heads are connected in a multi-hop

manner. The second model considers heterogeneous networks in which the cluster

heads have high capabilities in terms of processing power and transmission range to

directly reach adjacent cluster heads in a single-hop manner. Simulation results show


23

that this scheme achieves good throughput performance while providing stable data

reporting that is independent of the network density. The scheme also allows for

energy savings by using load-balanced data reporting paths. Abinash Mahapatra et al.

[30] proposed an energy aware dual-path routing scheme for real-time traffic, which

balances node energy utilization to increase the network lifetime, takes network

congestion into account to reduce the routing delay across the network and increases

the reliability of the packets reaching the destination by introducing minimal data

redundancy. This paper also introduces an adaptive prioritized Medium Access Layer

(MAC) to provide a differentiated service model for real-time packets. Our claims are

well supported by simulation results. Jalel Ben-Othman et al. [31] proposed an

Energy Efficient and QoS aware multipath routing protocol (abbreviated shortly as

EQSR) that maximizes the network lifetime through balancing energy consumption

across multiple nodes, uses the concept of service differentiation to allow delay

sensitive traffic to reach the sink node within an acceptable delay, reduces the end to

end delay through spreading out the traffic across multiple paths, and increases the

throughput through introducing data redundancy. EQSR uses the residual energy,

node available buffer size, and Signal-to-Noise Ratio (SNR) to predict the best next

hop through the paths construction phase. Based on the concept of service

differentiation, EQSR protocol employs a queuing model to handle both real-time and

non-real-time traffic. By means of simulations, they evaluate and compare the

performance of proposed routing protocol with the MCMP (Multi-Constraint Multi-

Path) routing protocol. Simulation results have shown that this protocol achieves

lower average delay, more energy savings, and higher packet delivery ratio than the

MCMP protocol.

2.3. Network Energy Efficiency

WSNs are highly distributed self-organized systems and depend upon a particular

number of scattered low cost small devices. These devices include some strong

demerits in terms of processing, memory, communications and energy capabilities.

Sensor nodes collect measurements of interest over a given space and make them

available to external systems and networks at sink nodes. The power saving

techniques is commonly implemented to increase the independence of the individual

nodes and this technique makes the nodes to sleep most of the time. This can be

balanced with low power communications, which usually lead to multi hop data


24

transmission from sensor nodes to sink nodes and vice versa [54]. In order to collect

the data, WSN uses an event-driven model and depends upon the collective effort of

the sensor nodes in the network. Greater accuracy, larger coverage area and extraction

of localized features are some of the advantages of the event-driven model over the

traditional sensing. It is important that the preferred events are reliably transported to

the sink for realizing these potential gains [43]. Habitat monitoring, in-door

monitoring, target tracking and security surveillance are some of the applications

where WSNs can be used. WSNs have some problems to be overcome such as energy

conservation, congestion control, reliability data dissemination, security and

management of a WSN itself. These problems often take part in one or more layers

from application layer to physical layer and it can be studied separately in each

corresponding layer or collaboratively cross each layer. For example, congestion

control may involve only in transport layer but the energy conservation may be

related to physical layer, data link layer, network layer and higher layers [42].

At transport layers, notion of reliability exists in the Pump-slowly, Fetch-quickly: A

Reliable Transport Protocol for Sensor Networks (PSFQ) and Event-to-Sink Reliable

Transport in Wireless Sensor Networks (ESRT) protocols [5,6]. Whereas, PSFQ deals

with data flow with strict delivery guarantees, ESRT is a solution to achieve reliable

event detection with minimum energy expenditure and congestion resolution. Fuzzy

logic based management and control has been studied in the past for wireless

networks and ad hoc networks. Reliable data transport in sensor networks (RMST)

[11] is a transport layer paradigm designed to complement directed diffusion by

adding a reliable data transport service on top of it. It‟s a NACK based protocol like

PSFQ, which has primarily timer driven loss detection and repair mechanisms. It does

not provide with any congestion control mechanism. Jaesub Kim et al. [20] suggest a

transport-controlled MAC protocol (TC-MAC) that combines the transport protocol

into the MAC protocol with the aims of achieving high performance as well as energy

efficiency in multi-hop forwarding. Accordingly TC-MAC also works through a

periodic listen-and-sleep scheme, it lowers end-to-end latency by reserving data

forwarding schedules across multi-hop nodes during the listen period and by

forwarding data during the sleep period, all while increasing throughput by

piggybacking the subsequent data forwarding schedule on current data transmissions

and forwarding data consecutively. In addition, TC-MAC gives a fairness-aware

lightweight transport control mechanism based on benefits of using the MAC-layer


25

information. The results show that TC-MAC performs as well as an 802.11-like MAC

in end-to-end latency and throughput, and is more efficient than S-MAC in energy

consumption, with the additional advantage of supporting fairness-aware congestion

control. Sudip Misra et al. in [26] proposed a simple, least-time, energy-efficient

routing protocol with one-level data aggregation that ensures increased lifetime for

the network. This protocol was compared with popular ad hoc and sensor network

routing protocols, viz., AODV ([Royer and Perkins, 1999] and [Das et al., 2003]),

DSR (Johnson et al., 2001), DSDV (Perkins and Bhagwat, 1994), DD (Intanagon

wiwat et al., 2000) and MCF (Ye et al., 2001). It was observed that the proposed

protocol outperformed them in throughput, latency, average energy consumption and

average network lifetime. The proposed protocol uses absolute time and node energy

as the criteria for routing, this ensures reliability and congestion avoidance. Tuan Lea

et al. [29] proposes ERTP, an Energy-efficient and Reliable Transport Protocol for

Wireless Sensor Networks. ERTP is designed for data streaming applications, in

which sensor readings are transmitted from one or more sensor sources to a base

station (or sink). ERTP uses a statistical reliability metric, which ensures the number

of data packets delivered to the sink exceeds the defined threshold. Extensive discrete

event simulations and experimental evaluations shows that ERTP is significantly

more energy-efficient than current approaches and can reduce energy consumption by

more than 45% when compared to current approaches. Consequently, sensor nodes

are more energy-efficient and the lifespan of the unattended WSN is increased.

Sandip Dalvi et al. [43] have proposed a transport protocol, which provides the

desired event reliability to the application, by distributing the load at a sensor among

its children based on their residual energies and average MAC layer data rate. The

event rate distribution happens in such a way that the application at the sink gets its

required event rate and the overall energy consumption of nodes is minimized. They

have derived a method for computing average MAC data rate for these two protocols

and using simulations they have shown that our transport protocol performs close to

optimal. Damayanti Datta et al. [47] have proposed a new protocol for reliable data

transfer in time-critical applications with zero tolerance for data loss in wireless

sensor networks, which uses less time and fewer messages in comparison to an

established protocol PSFQ. The two key features of their proposed protocol are out-

of-sequence forwarding of packets with a priority order for sending different types of

messages at nodes and delaying the requests for missing packets. They have also


26

presented two methods for computation of the delay in requesting missing packets.

Ching-Wen Chen et. al. [49] proposed the use of node grouping and transmission

pipelining to reduce power consumption and transmission delay. In the design of node

grouping, there are several groups in WSNs, where nodes in different groups wake up

at different time. Each sensor node is initially set to belong to one of these groups. In

contrast to the situation in which all nodes hear the control packets during the

contention period, node grouping reduce the number of nodes that overhears the

control packets at the same time to reduce power consumption. The group table

recodes the group indices of all the neighbors of that node. With looking up a group

table in a sensor node, a sender can wake up at the group time of the receiver. As a

result, two nodes belonging to different groups can communicate with other. With

regard to transmission delay of a multi-hop path in WSNs, if a sender transmits data

to the receiver and the receiver cannot send the data to the next receiver right now, the

transmission delay increases. To reduce the transmission delay, they proposed the

transmission pipelining method. Transmission pipelining makes the group number of

the nodes on a path to be continuous. Therefore, the sensor node is thus able to

transmit data to the sink node pipelining. From the simulation results, when the

number of groups is 2, the power consumed in transmitting a byte (mJ/byte) and the

transmission delay in our proposed design are better than those of SMAC by about

50%. When the number of groups is 4, although the transmission delay is only a little

better than that of SMAC, the power consumed in transmitting a byte in this design is

much less than the power consumed in SMAC by 75%. Kwan-Wu Chin et al. [50]

proposed E2MAC, an energy efficient, distributed Medium Access Control (MAC)

protocol for identifying and monitoring tags in RFID-enhanced wireless sensor

networks. E2MAC exploits the low power capability of a ultra-wideband transceiver

and distinct pulses to address the reader collision problem. In addition, it uses

ResMon, an enhanced dynamic frame slotted Aloha protocol to read and monitor tags.

Lastly, E2MAC uses a novel load-balancing algorithm to amortize the cost of reading

and monitoring tags to multiple readers. These E2MAC features ensure that

the contention level at each reader is kept at a minimum and distributed fairly. As a

result, E2MAC has a high reading rate and low energy consumption. In addition,

E2MAC helps in minimizing the impact of the tag orientation problem, where a tag

becomes unreadable if its antenna is parallel to a reader‟s field lines. In particular, the

use of multiple readers increases spatial diversity and hence increases the likelihood


27

that a tag is readable by at least one reader. Our simulation results show E2MAC to

have very low energy consumption, reading delay and per-reader collision. More

importantly, system designers have the flexibility to lower these metrics further with

additional readers, bigger frame sizes, or by dividing tags into small groups. Marcel

Busse et al. [52] proposed two forwarding schemes termed single-link and multi-link

energy-efficient forwarding that trade off delivery ratios against energy costs. Multi-

link forwarding improves the network performance substantially by addressing

multiple receivers at once during the packet forwarding process. If the first

forwarding node does not receive a packet correctly, other nodes may act as backup

nodes and perform the forwarding instead. By means of mathematical analyses, it is

derived, how the energy efficiency of a forwarding path can be computed and how a

forwarding tree is established. Routing cycles are explicitly taken into account and

prevented by means of sequence numbers. Simulations and real-world experiments

provide a comparison to other reference strategies, showing a superior performance of

the forwarding scheme in terms of energy efficiency.

2.4. Traffic Contention and Network Congestion

Basically following factors are the key factors to control traffic contention and

network congestion.

2.4.1. Reliable Data Transport

The problem of reliable transport over wireless multi-hop networks like WSNs is not

an easy one to solve. Three main sources of packet losses can be found

1. Wireless channel is inclined to introduce transmission errors. Either transmission

from different nodes can collide or other failures in nodes can produce package

losses.

2. Packets can be discarded in the network due to congestion, i.e., intermediate

nodes‟ overload.

3. The receiver might discard packets because they arrive too quickly, implying a

failure in flow control.

2.4.2. Congestion Control

There are two major causes of congestion in WSNs. The first is when the packet

arrival rate exceeds the packet service rate. This is more likely to occur at sensor


28

nodes near the sink, since normally they carry more upstream traffic. The second

cause relates to performance aspects of the link layer such as contention, interference,

and bit error rate.

Congestion in WSNs has a direct impact on energetic efficiency and on QoS

parameters. For example, congestion may cause buffer overflow, which could lead to

large queuing delays and higher loss rates. Packet loss not only degrades the

reliability and QoS of the application, but also wastes node energy. The congestion

can also degrade the link utilization. Furthermore, link-level congestion results in

transmission collisions in contention-based link protocols such as CSMA. Collisions

during transmissions increase packet service time and waste energy, so congestion in

WSNs must be efficiently controlled, either to suppress it or to decrease its harmful

effects. Typically, there are three mechanisms for controlling congestion: congestion

detection, congestion notification, and rate adjustment.

2.4.3. Congestion detection

In TCP, the congestion is observed or deduced by end nodes based on a timeout or

redundant acknowledgments. In WSNs, proactive methods are preferred. A common

mechanism would be to concrete a queue length (Hull et al., 2004; Wan et al., 2003),

a service time (Wang et al., 2006), or the packet service time ratio over packet inter

arrival time at the intermediate nodes (Wang et al., 2006). In WSNs with collision-

based MAC protocols such as CSMA, the channel load can be measured and used as a

congestion indication.

2.4.4. Congestion notification

After detecting congestion in the network, the transport protocol needs to propagate

data about congestion from the congested nodes to the upstream or source nodes that

contribute to the congestion. The approach to disseminating congestion data can be

classified into implicit congestion notification and explicit congestion notification.

Explicit congestion notification uses special control messages to notify the involved

nodes that congestion is occurring, by means of suppression messages. Implicit

congestion notification is included in normal data packets, usually a bit inside the

packet.


29

2.4.5. Rate adjustment

When receiving a congestion indication, the node can adjust its data transmission rate.

If a single congestion notification (CN) bit is used, an additive-

increase/multiplicative-decrease (AIMD) scheme or one of its variants can be applied.

If the protocol implements additional information about congestion, more accurate

rate adjustment schemes can be adopted.

2.5. Medium Access Protocols for WSNs

Several authors have suggested medium access schemes for WSNs, some of which

are modifications of existing protocols for wireless ad hoc networks. This is still a

growing area of research calling attention to several open issues yet to be addressed.

Several recently proposed schemes are discussed below.

2.5.1. Contention-Based Protocols

Sensor MAC (S-MAC): S-MAC is a contention-based MAC protocol explicitly

designed for wireless sensor networks. While reducing energy consumption is the

primary goal in those networks, this protocol has also achieved good scalability and

collision avoidance by using a combined scheduling and contention scheme.

To achieve the primary goal of energy efficiency, the main sources that cause the

inefficient use of energy as well as what trade-offs can be made to reduce energy

consumption need to be identified. In this way, the following major sources of energy

waste are identified:

Collisions: When a transmitted packet is corrupted, it has to be discarded; the

follow-up retransmissions increase energy consumption. Collisions not only waste

energy, but they increase latency as well.

Overhearing: A node can pick up packets intended for other nodes.

Control packet overhead: Sending and receiving control packets also consumes

energy.

Idle listening: Listening to receive possible traffic that was not sent can be the

biggest cause of inefficiency, especially in many sensor network applications when

nodes are in the idle state most of the time. Most sensor networks are designed to

operate over a long period of time; since the nodes are idle for a long time, idle

listening can be a dominant factor behind energy waste in such cases.


30

Hongwei Zhanga et al. [27] addresses the challenges of bursty converge cast in multi-

hop wireless sensor networks, where a large burst of packets from different locations

needs to be transported reliably and in real-time to a base station. Via experiments on

a 49 MICA2 mote sensor network using a realistic traffic trace, they determined the

primary issues in bursty converge cast, and accordingly design a protocol, RBC (for

Reliable Bursty converge cast), to address these issues: To improve channel

utilization and to reduce ack-loss, they design a window-less block acknowledgment

scheme that guarantees continuous packet forwarding and replicates the

acknowledgment for a packet; to alleviate retransmission-incurred channel contention,

and introduce differentiated contention control. Moreover, they design mechanisms to

handle varying ack-delay and to reduce delay in timer-based retransmissions. They

evaluate RBC, again via experiments, and show that compared to a commonly used

implicit-ack scheme, RBC doubles packet delivery ratio and reduces end-to-end delay

by an order of magnitude, as a result of which RBC achieves a close-to-optimal

goodput. Hongwei Zhang et al. [39] have designed a window-less block

acknowledgment scheme to improve channel utilization and to reduce

acknowledgment loss that guarantees continuous packet forwarding and replicates the

acknowledgment for a packet They have introduced a differentiated contention

control to alleviate retransmission-incurred channel contention. Moreover, they have

designed mechanisms to handle varying ack delay and to reduce delay in timer based

retransmissions. Ajit Warrier et al. [40] have presented a new hybrid MAC scheme Z-

MAC, for sensor networks. Z-MAC is robust topology changes and clock

synchronization errors; in the worst case its performance falls back to that of CSMA.

They have implemented Z-MAC in Tiny OS and evaluated its channel utilization,

energy, latency and fairness over single-hop, two hop and multi-hop sensor network

topologies constructed using Mica2. Their results have shown that Z-MAC has

remarkably better data throughput than existing sensor MAC protocols while

consuming comparable energy. Yao-Nan Lien et al. [45] has proposed the Hop-by-

Hop TCP protocol for sensor networks aiming to accelerate reliable packet delivery.

Hop-by-Hop TCP makes every intermediate node in the transmission path execute a

lightweight local TCP to guarantee the transmission of each packet on each link. It

takes less time in average to deliver a packet in an error-prone environment.Another

QoS based routing protocol was presented by Akkaya [3], which classifies the traffic

on the basis of real-time and non-real time application data. This protocol further


31

makes use of a cumulative link-cost for each link and end-to-end delay and chooses

the least-cost link, though the weightages allocated to different QoS parameters are

not mentioned in detail. In [12], authors presented a fuzzy congestion control

approach for ad-hoc networks, in which a theoretical fuzzy logic based concept is

used to control the congestion. A good number of transport layer protocols have been

proposed for WSNs. These works aim to provide reliability guarantee either by

congestion detection & control or by congestion avoidance [1] [9] [10]. ESRT [6]

allocates transmission rate to sensors such that an application-defined number of

sensor readings is received at a base station, while ensuring that the network is not

congested. On reception of packets with congestion notification a bit high, sink node

regulates the reporting rate by broadcasting a high-energy control signal so that it

could reach the all sources. This high-powered congestion control signal may disrupt

some other transmissions. Also the assumption of congestion notification by the sink

node is very optimistic. In CODA [1], they present a detailed study on congestion

avoidance in sensor networks. The basic idea is that as soon as congestion occurs, the

source (or an intermediate node)‟s sending rates must be reduced to quickly release

the congestion. In the simple case, as soon as a node detects congestion, it broadcasts

a backpressure message upstream. An upstream node that receives the backpressure

can decide to drop packets, preventing its queue from building up and thus controlling

congestion. If multiple sources are sending packets to a sink, CODA also provides a

method of asserting congestion control over these multiple sources by requiring

constant feedback (ACKs) from the sinks. If a source does not receive the ACKs at

predefined times, it will start throttling the sending rates.CODA uses a combination of

the present and past channel loading conditions and the current buffer occupancy, to

infer accurate detection of congestion at each receiver with low cost. As long as a

node detects congestion, it sends backpressure messages to upstream nodes for

controlling reporting rate hop-by hop. It is also capable of asserting congestion

control over multiple sources from a single sink in the event of persistent congestion.

Even though it overcomes some of the limitations of ESRT, it doesn‟t consider the

event fairness and packet reliability at all. PSFQ [5] is scalable and reliable transport

protocol that deals with strict data delivery guarantees rather than desired event

reliability as it is done in ESRT. However, this approach involves highly specialized

parameter tuning and accurate timing configuration that makes it unsuitable for many

applications. As defined in Many-to-One Routing [2], event fairness is achieved when


32

equal number of packets is received from each node. Topology Aware Resource

Adaptation to Alleviate Congestion in Sensor Networks (TARA) [10] discusses the

network hotspot problem and presents a topology aware resource adaptation strategy

to alleviate congestion in sensor network. In these two key reasons of packet losses

have been taken into account: loss due to collision and loss due to buffer overflow,

taking care of hierarchical medium access and, thereby, reduces packet drops due to

collision. WRRF controls the number of packets to be received from upstream nodes

in each round (single-hop control). Estimating buffer status at each individual

downstream node using Exponential Moving Average (EWMA) controls round

operation. A downstream node allows packet from its upstream nodes only if there is

available buffer and thereby avoid drops due to buffer overflow. According to

Mohammad HosseinYaghmaeeetal.[15]to support quality of service (QoS)

requirements for multimedia applications having a reliable and fair transport protocol

is necessary. One of the main objectives of the transport layer in WMSNs is

congestion control. In their research they observe that the information provided might

have different levels of importance and argue that sensor networks should be willing

to spend more resources in disseminating packets carrying more important

information. Some applications of WMSNs may need to send real time traffic toward

the sink node. This real time traffic requires low latency and high reliability so that

immediate remedial and defensive actions can be taken when needed. Therefore,

similar to wired networks, service differentiation in wireless sensor networks is also

an important issue. They present a priority-based rate control mechanism for

congestion control and service differentiation in WMSNs. Also a distinguish theory

between high priority real time traffic from low priority non-real time traffic, and

service the input traffic based on its priority is presented. With Simulation results, the

superior performance of the proposed model with respect to delays, delay variation

and loss probability is confirmed. Guohua Zhanga et al. [16] explicitly model link

capacities to be time varying and investigate congestion control problems in multi-

hop wireless networks. They propose a primal–dual congestion control algorithm

which is proved to be trajectory stable in the absence of feedback delay. Different

from system stability around a single equilibrium point, trajectory stability guarantees

the system is stable around a time varying reference trajectory. They also obtain

sufficient conditions for the scheme to be locally stable in the presence of delay. The

key technique is to model time variations of capacities as perturbations to a constant


33

link. They also investigate the sensitivity of the control scheme and through

simulations to study the tradeoff between stability and sensitivity. Gungora et al. [17]

comprehensively investigate the interactions between contention resolution and

congestion control mechanisms as well as the physical layer effects in WSAN. An

extensive set of simulations is performed in order to quantify the impacts of several

network parameters on the overall network performance. The results of this analysis

reveal that the interdependency between network parameters calls for adaptive cross-

layer mechanisms for efficient data delivery in WSAN. Bjorn Scheuermann et al. [18]

present a novel hop-by-hop congestion control protocol that has been tailored to the

specific properties of the shared medium. In this scheme, backpressure towards the

source node is established implicitly, by passively observing the medium. A

lightweight error detection and correction mechanism guarantees a fast reaction to

changing medium conditions and low overhead. This approach is equally applicable

to TCP and UDP-like data streams. They demonstrate the performance of this

approach by an in-depth simulation study. These findings are underlined by testbed

results obtained using an implementation of our protocol on real hardware.Ben-Jye

Changae et al. [21] suggests that TCP/IP transmissions, the TCP congestion control

operates well in the wired network, but it is difficult to determine an accurate

congestion window in a heterogeneous wireless network that consists of the wired

Internet and various types of wireless networks. The primary reason is that TCP

connections are impacted by not only networks congestion but also error wireless

links. This paper thus proposes a novel adaptive window congestion control namely

Logarithmic Increase Adaptive Decrease, LIAD for TCP connections in

heterogeneous wireless networks. The proposed RTT-based LIAD has the capability

to increase throughput while achieving competitive fairness among connections with

the same TCP congestion mechanism and supporting friendliness among connections

with different TCP congestion control mechanisms. In the Congestion Avoidance

(CA) phase, an optimal shrink factor is first proposed for Adaptive Decreasing cwnd

rather than a static decreasing mechanism used by most approaches. Second, they

adopt a Logarithmic Increase algorithm to increase cwnd while receiving each ACK

after causing three duplicate ACKs. The analyses of congestion window and

throughput under different packet loss rate are analyzed. Furthermore, the state

transition diagram of LIAD is detailed. Numerical results demonstrate that the

proposed LIAD outperforms other approaches in goodput, fairness, and friendliness


34

under diverse heterogeneous wireless topologies. Especially, in the case of 10%

packet loss rate in wireless links, the proposed approach increases goodput up to

156% and 1136% as compared with LogWestwood+ and NewReno, respectively.

Vivek Raghunathan et al. in [22] studied the interaction between TCP congestion

control and wireless interference. One of the triumphs of wireline network research of

the last decade has been the casting of the Internet congestion control problem within

an optimization framework based on utility functions. Such an approach has provided

a sound theoretical understanding of the underlying stability and fairness issues, as

well as a post-facto justification of the scalability and stability of TCP-like additive-

increase multiplicative-decrease (AIMD) algorithms. This paper provides

counterexamples showing that the same result cannot be extended to wireless

networks, at least not in a straightforward manner. The fundamental difference is that

wireless networks are of a broadcast nature. There is no strict notion of a „link‟, since

transmissions from nearby nodes interfere with each other. They consider a fairly

general model of interference in wireless networks, and present a counterexample of a

wireless network in which the congestion control mechanism has an unstable

equilibrium point at the desired fair solution. NS-2 simulations of this counterexample

manifest an oscillatory throughput behavior that is orders of magnitude worse than the

corresponding wired networks. Surprisingly, this oscillatory throughput behavior

appears to be fairly typical of simulations in wireless networks, with almost all

randomly chosen network simulation examples manifesting it. This loss of stability

leads us to suggest that perhaps TCP should be modified for use in wireless networks,

and that a cross-layer redesign of wireless TCP and MAC is needed to explicitly

account for the effects of the wireless nature of interference. Iradj Ouveysia et al. [24]

proposed linearized congestion minimization schemes with working and protection

paths (LCM–WP), in which a mixed integer linear program is formulated to choose

the optimal working and protection paths for every OD pair such that the network

congestion is minimized. In particular, the objective is to minimize the maximum

amount of traffic on the links. To solve realistically sized problems, they consider a

restricted version of the LCM–WP, in which only limited sets of candidate working

and protection paths are considered. A simple algorithm is developed to find

candidate working and protection paths for each origin-destination (OD) pair.

Implementation of our LCM–WP schemes demonstrates the efficiency of this

approach in terms of the number of constraints and solution time. It also shows that


35

this approach is applicable to realistically sized networks. RamanujaVedanthama et

al. [25] focus on providing congestion control from the sink to the sensors in a sensor

field. They identify the different reasons for congestion from the sink to the sensors

and show the uniqueness of the problem in sensor network environments. They

proposed a generic framework that addresses congestion from the sink to the sensors

in a sensor network. Through ns2-based simulations, they evaluate the proposed

approach and compare its performance with three baseline approaches. Young-Duk

Kim et al. [32]proposed Distance Adaptive Contention Window (DACW) modified

IEEE 802.15.4 standard. The key mechanism of DCAW is a dynamical channel

access MAC protocol, which is to adjust Contention Window (CW) according to the

hop count distance to sink and traffic condition. With DACW, each sensor node can

achieve self-routing capability with low overhead and performance enhancement.

Furthermore, DACW can be easily applied to existing routing protocols without

additional overhead and shows that its performance is better than the existing MAC

protocol by the simulation result.Md. Mamun-Or-Rashid et al. [33] proposed an

energy efficient congestion avoidance protocol that includes source count based

hierarchical and load adaptive medium access control. The proposed mechanism

ensures load adaptive media access to the nodes and thus achieves fairness in event

detection. The results of simulation show that this scheme exhibits more than 90%

delivery ratio with retry limit 1, even under bursty traffic condition, which is good

enough for reliable event perception. According to Hull et al. [36] network congestion

occurs when offered traffic load exceeds available capacity at any point in a network.

In wireless sensor networks, congestion causes overall channel quality to degrade and

loss rates to rise, leads to buffer drops and increased delays (as in wired networks),

and tends to be grossly unfair toward nodes whose data has to traverse a larger

number of radio hops. Congestion control in wired networks is usually done using

end-to-end and network-layer mechanisms acting in concert. However, this approach

does not solve the problem in wireless networks because concurrent radio

transmissions on different “links” interact with and affect each other, and because

radio channel quality shows high variability over multiple time-scales. We examine

three techniques that span different layers of the traditional protocol stack: hop-by-

hop flow control, rate limiting source traffic when transit traffic is present, and a

prioritized medium access control (MAC) protocol. They implemented these

techniques and present experimental results from a 55-node in-building wireless


36

sensor network. They also demonstrate the combination of these techniques, Fusion,

can improve network efficiency by a factor of three under realistic workloads. Ian F.

Akyildiz et al. [37] have developed a unified cross-layer protocol, which replaces the

entire traditional layered protocol architecture that has been used so far in WSNs. The

objective of their proposed cross layer protocol is highly reliable communication with

minimal energy consumption, adaptive communication decisions and local congestion

avoidance. Their protocol operation is governed by the new concept of initiative

determination. Based on this concept, the cross-layer protocol performs received

based contention, local congestion control, and distributed duty cycle operation in

order to realize efficient and reliable communication in WSNs. Wafa Ben Jaballah et

al. [38] have enhanced the QoS in sensor networks. They have presented an approach,

which takes advantage of the standard 802.11e EDCA protocol that ensures effective

end-to-end delay and good quality of traffic. They have tried to improve the provision

of quality of service in sensor networks by offering a new approach, which aims to

improve the mechanism of service differentiation, implemented in the

802.11e.Muhammad MostafaMonowar et al. [41] have proposed an efficient scheme

to control multi-path congestion so that the sink can get priority based throughput for

heterogeneous data. They have used packet service ratio for detecting congestion as

well as performed hop-by-hop multi-path congestion control based on that metric.

Their simulation results have demonstrated the effectiveness of their proposed

approach. Chonggang Wang et al. [42] have proposed a node priority-based

congestion control protocol (PCCP) for wireless sensor networks. In PCCP, node

priority index has been introduced to reflect the importance of each node. PCCP uses

packet inter-arrival time along with packet service time to measure a parameter

defined as congestion degree and furthermore imposes hop-by-hop control based on

the measured congestion degree as well as the node priority index. PCCP controls

congestion faster and more energy-efficiently than other known techniques. Nurcan

Tezcan et al. [44] have addressed the problem of reliable data transferring by first

defining event reliability and query reliability to match the unique characteristics of

WSNs. They have considered event delivery in conjunction with query delivery. They

have proposed an energy-aware sensor classification algorithm to construct a network

topology that is composed of sensors in providing desired level of event and query

reliability. They have analyzed their approach by taking asymmetric traffic

characteristics into account and incorporating a distributed congestion control


37

mechanism. They have evaluated the performance of their proposed approach through

an ns-2 based simulation and show that significant savings on communication costs

are attainable while achieving event and query reliability. Sunil Kumar et al. [46]

have studied the performance of ESRT in the presence of over-demanding event

reliability, using both the analytical and simulation approaches. They have shown that

the ESRT protocol does not achieve optimum reliability and begins to fluctuate

between two inefficient network states. With insights from update mechanism in

ESRT, they have proposed a new algorithm, called enhanced ESRT (E2SRT), to solve

the over-demanding event reliability problem and to stabilize the network. Their

simulation results show that their E2SRT outperforms ESRT in terms of both

reliability and energy consumption in the presence of over-demanding event

reliability. It also ensures robust convergence in the presence of dynamic network

environments. Ilker Demirkol et al. [51] showed the significance of using a realistic

and application-specific packet traffic model by comparing the performance of a well-

known WSNs protocol under the Surveillance WSNs packet traffic model (SPTM), as

well as under periodic and binomial traffic models. A packet traffic framework

specific to surveillance applications is proposed which is then used for deriving

SPTM analytically. In order to be adaptable and flexible, SPTM incorporates a

probabilistic and parametric sensor detection model. Simulation results show that to

employ an application-specific packet traffic model has significant impact on the

performance evaluation of the WSN and ordinary traffic models may overestimate the

capacity of the WSN.

2.6. Network Routing

In WSNs, reliability is a design goal of a primary concern. To build a comprehensive

reliable system, it is essential to consider node failures and intruder attacks as

unavoidable phenomena. Y. Challalet et al. present a new intrusion-fault tolerant

routing scheme offering a high level of reliability through a secure multipath routing

construction. Unlike existing intrusion-fault tolerant solutions, our protocol is based

on a distributed and in-network verification scheme, which does not require any

referring to the base station. Furthermore, it employs a new multipath selection

scheme seeking to enhance the tolerance of the network and conserve the energy of

sensors. Extensive analysis and simulations using Tiny-OS showed that our approach


38

improves many important performance metrics such as: the mean time to failure of

the network, detection overhead of some security attacks, energy consumption, and

resilience [56]. WSNs with nodes spreading in a target area have abilities of sensing,

computing, and communication. Since the GPS device is expensive, Tai-Jung Chang

et al. [57] used a small number of fixed anchor nodes that are aware of their locations

to help estimate the locations of sensor nodes in WSNs. To efficiently route sensed

data to the destination, the server, identifying the location of each sensor node can be

of great help. This work adopts a range-free color-theory based dynamic localization

approach, to help identify the location of each sensor node. Since sensor nodes are

battery-powered, we propose an efficient color-theory-based energy efficient routing

(CEER) algorithm to prolong the lifetime of each sensor node. The uniqueness of this

approach is that by comparing the associated RGB values among neighboring nodes,

a better routing path with energy awareness can be efficiently chosen. Besides, the

CEER has no topology hole problem. Simulation results have shown that CEER

algorithm can save up to 50–60% energy than ESDSR in mobile wireless sensor

networks. In addition, the latency per packet of CEER is 50% less than that of

ESDSR.

Routing protocols for WSNs face two challenges. One is an efficient bandwidth

usage, which requires minimum delay between transfers of packets. Establishing

permanent routes from the source to destination addresses this challenge since the

received packet can be immediately transmitted to the next node. However, any

disruption on the established path either causes packet loss, lowering the delivery rate,

or invokes a costly process of creating an alternative path. The second challenge is the

ability to tolerate permanent and transient failures of nodes and links, especially since

such failures are frequent in sensor networks. Protocols that chose the forwarding

node at each hop of a packet are resilient to such failures, but incur the delay caused

by selection of the forwarding node at each hop of the multi-hop path. Thomas A.

Babbitt et al. [58] present a novel wireless sensor routing protocol, self-selecting

reliable path routing (SRP) for wireless sensor network (WSN) routing that addresses

both challenges at once. This protocol evolved from the self-selecting routing (SSR)

protocol, which is essentially memory-less. In the first generation of SSR protocols

each packet selects the forwarding node at each hop on its path from the source to

destination. The protocol takes advantage of broadcast communication commonly

used in WSNs as a communication primitive. It also uses a prioritized transmission


39

back-off delay to uniquely identify the neighbor of the forwarder that will forward the

packet As a result, the protocol is resistant to node or link failures as long as an

alternative path exists from the current forwarder to the destination. The second

generation of SSR protocols, called self-healing routing (SHR) added the route repair

procedure, invoked when no neighbor of the forwarder closer to the destination is

alive. In a series of transmissions, a packet trapped at the current forwarder by failures

of its neighbors is capable of backing-off towards the source to find an alternative

route, if such exists, to the destination. The main contribution of this work [58] is the

third generation of SSR protocols, termed self-selecting reliable path routing, SRP. It

preserves SHRs dynamic path selection in face of failure. Yet it also enables packets

to follow established paths without selection delay if failures do not occur. The

important change in the protocol is to make it memorize the successfully traversed

path and attempt to reuse it for subsequent packets flowing to the same destination.

The interesting behavior of SRP arising from this property is that if a path from the

source to destination exists on which no transient failures occur, SRP would converge

its routing to such a reliable path. A novel element of the SRP protocol that resulted in

the desired properties is described in this work. Using simulation, SRP protocol with

the representatives of the two other approaches: AODV as the route-based protocol,

and GRAB and SHR as the hop-selection protocols are compared. Considering severe

resources constraints and security threat of wireless sensor networks, [59] proposed a

novel hierarchical routing protocol algorithm. The proposed routing protocol

algorithm can adopt suitable routing technology for the nodes according to the

distance of nodes to the base station, density of nodes distribution, and residual

energy of nodes. Comparing the proposed routing protocol algorithm with simple

direction diffusion routing technology, cluster-based routing mechanisms, and simple

hierarchical routing protocol algorithm through comprehensive analysis and

simulation in terms of the energy usage, packet latency, and security in the presence

of node compromise attacks, the results show that the proposed routing protocol

algorithm is more efficient for wireless sensor networks.

WSNs are collection of wireless sensor nodes forming a temporary network without

the aid of any established infrastructure or centralized administration. In such an

environment, due to the limited range of each node‟s wireless transmissions, it may be

necessary for one sensor node to ask for the aid of other sensor nodes in forwarding a

packet to its destination, usually the base station. One important issue when designing


40

wireless sensor network is the routing protocol that makes the best use of the severely

limited resource presented by WSN, especially the energy limitation. Another import

factor required attention from researchers is providing as much security to the

application as possible. Nidal Nasser et al. [60] proposed routing protocols in the

literature focus either only on increasing lifetime of network or only on addressing

security issues while consuming much power. None of them combine solutions to the

two challenges. In there this work a new routing protocol called SEEM: Secure and

Energy-Efficient multipath Routing protocol is proposed. SEEM uses multipath

alternately as the path for communicating between two nodes thus prolongs the

lifetime of the network. On the other hand, SEEM is effectively resistive to some

specific attacks that have the character of pulling all traffic through the malicious

nodes by advertising an attractive route to the destination. The performance of this

protocol is compared to the Directed Diffusion protocol. Simulation results show that

this protocol surpasses the Directed Diffusion protocol in terms of throughput, control

overhead and network lifetime.

A new class of WSNs that harvest power from the environment is emerging because

of its intrinsic capability of providing unbounded lifetime. While a lot of research has

been focused on energy-aware routing schemes tailored to battery-operated networks,

the problem of optimal routing for energy harvesting wireless sensor networks (EH-

WSNs) has never been explored. The objective of routing optimization in this context

is not extending network lifetime, but maximizing the workload that can be

autonomously sustained by the network.

In [61] Emanuele Lattanzi et al. present a methodology for assessing the energy

efficiency of routing algorithms for networks whose nodes drain power from the

environment. A methodology that makes use of graph algorithms and network

simulations for evaluating the MESW starting from a network topology, a routing

algorithm and a distribution of the environmental power available at each node is

proposed. A tool flow implementing the proposed methodology is presented and

comparative results achieved on several routing algorithms is shown. Experimental

results highlight that routing strategies that do not take into account environmental

power do not provide optimal results in terms of workload sustainability. Using

optimal routing algorithms may lead to sizeable enhancements of the maximum

sustainable workload. Since environmental power sources change over time, the


41

results prompt for a new class of routing algorithms for EH-WSNs that are able to

dynamically adapt to time-varying environmental conditions.

Shun-Yu Chuang et al. [62] propose a simple and scalable approach to multisink

routing scheme in WSNs. These networks are a rapidly growing discipline, with new

technologies emerging and new applications under development. In addition to

providing light and temperature measurements, wireless sensor nodes have

applications such as security surveillance, environmental monitoring, and wildlife

watching. One potential problem in a sensor network is how to transmit packets

efficiently from single-source to multi-sinks, i.e., to gather data from a single sensor

node and deliver it to multiple clients who are interested in the data. The difficulty of

such a scenario is finding the minimum-cost multiple transmission paths. Many

routing algorithms have been proposed to solve this problem. Most current algorithms

address the reduction of power consumption, and potentially introduce a large delay.

This work proposes a novel multi-path routing algorithm, called hop count based

routing (HCR) algorithm, which considers energy cost and transmission delay

simultaneously. A hop count vector (HCV) is introduced to support routing decision.

Moreover, an additional pruning vector (PV) can further enhance routing

performance. The proposed algorithm also provides a maintenance mechanism to

handle the consequence of faulty nodes. A failure of a node leads to an inaccurate

HCV. Therefore, an efficient correction algorithm is necessary. An Aid-TREE (A-

TREE) is applied to facilitate restricted flooding. This correction mechanism is more

efficient than full-scale flooding for correcting the limited inaccurate HCVs. Finally,

the impact of failed nodes is studied, and an algorithm, called Lazy-Grouping, is

proposed to enhance the robustness of HCR.

Rumor routing [63] is a classic routing algorithm based on agents‟ random walk. This

work proposes a novel approach based on this routing algorithm. Hamid Shokrzadeh

et al. tried to improve the latency and energy consumption of the traditional algorithm

using propagation of query and event agents within straight lines, instead of using

purely random walk paths. Result showed that this method improves the delivery ratio

of the queries, which is a drawback of traditional rumor routing. Due to the reduction

of final path length between source and destination, they introduced a second layer

geographical routing. Moreover, a method is proposed to reduce the cost of

localization equipments by using cheaper equipments like AoA antennas. In order to


42

compare the performance measures with traditional algorithm, a simulation

framework is developed and extensive simulations are performed.

2.7. Security and Data Management in WSNs.

The rapid progress of wireless communication and embedded micro-sensing MEMS

technologies has made WSNs possible. WSNs normally come a large area with many

inexpensive, tiny sensor nodes, each capable of collecting, processing, and storing

environmental information, and communicating with neighboring nodes. In the past,

sensors were connected by wired lines but nowadays, ad hoc networking technologies

can much simplify the network formation task. Installation and configuration of a

wireless sensor network are thus effortless. Many applications of wireless sensor

networks have been proposed, including field data collection, remote monitoring and

control, smart home, factory automation, security, etc. Security is sometimes viewed

as a standalone component of a system‟s architecture, where a separate module

provides security. This separation is, however, usually a flawed approach to network

security. To achieve a secure system, security must be integrated into every

component, since components designed without security can become a point of attack.

Consequently, security must pervade every aspect of system design.[34]

Dai Zhi-Feng et al. [101] considered the features of limited energy and densely

deployed sensor nodes; the key challenge in wireless sensor networks is to reduce the

power consumption and the high information redundancy. Furthermore,

communication is widely viewed as the dominating power cost in many sensor

networks applications, so it is obvious that it will save energy if data is aggregated

before being sent to the sink node. As the illustrative examples are shown, rough set

theory is a useful tool for local redundancy information de-correlation and can

eliminate redundant transmission so as to provide a solution to uncertain data

management for wireless sensor networks. In [99] Steffen Peter et al. present an

overview of end-to-end encryption solutions for converges cast traffic in wireless

sensor networks that support in-network processing at forwarding intermediate nodes.

Other than hop-by-hop based encryption approaches, aggregator nodes can perform

in-network processing on encrypted data. Since it is not required to decrypt the

incoming ciphers before aggregating, substantial advantages are 1) neither keys nor

plaintext is available at aggregating nodes, 2) the overall energy consumption of the

backbone can be reduced, 3) the system is more flexible with respect to changing


43

routes, and finally 4) the overall system security increases. A qualitative comparison

of available approaches, point out their strengths, respectively weaknesses, and

investigate opportunities for further research is provided. Wireless Sensor Network

has been widely used and its real-time data processing capability is very limited. [98]

puts forward a new solution, which is the novel wireless sensor network node design

with hyperchaos encryption based on FPGA. With the wide application on Wireless

Sensor Network based on ZigBee protocol, authors encrypt the transported data in the

network using hyper chaos by FPGA, which combines their flexibility in rapid real-

time data processing with free configuration in FPGA and enhances the security of the

transported data. According to Arif Selcuk et al. [100] designing cost-efficient, secure

network protocols for WSNs is a challenging problem because sensors are resource-

limited wireless devices. Since the communication cost is the most dominant factor in

a sensor's energy consumption, we introduce an energy-efficient Virtual Energy-

Based Encryption and Keying (VEBEK) scheme for WSNs that significantly reduces

the number of transmissions needed for rekeying to avoid stale keys. In addition to the

goal of saving energy, minimal transmission is imperative for some military

applications of WSNs where an adversary could be monitoring the wireless spectrum.

VEBEK is a secure communication framework where sensed data is encoded using a

scheme based on a permutation code generated via the RC4 encryption mechanism.

The key to the RC4 encryption mechanism dynamically changes as a function of the

residual virtual energy of the sensor. Thus, a one-time dynamic key is employed for

one packet only and different keys are used for the successive packets of the stream.

The intermediate nodes along the path to the sink are able to verify the authenticity

and integrity of the incoming packets using a predicted value of the key generated by

the sender's virtual energy, thus requiring no need for specific rekeying messages.

VEBEK is able to efficiently detect and filter false data injected into the network by

malicious outsiders. The VEBEK framework consists of two operational modes

(VEBEK-I and VEBEK-II), each of which is optimal for different scenarios. In

VEBEK-I, each node monitors its one-hop neighbors where VEBEK-II statistically

monitors downstream nodes. In this work VEBEK's feasibility and performance

analytically and through simulations is evaluated. The results show that VEBEK,

without incurring transmission overhead (increasing packet size or sending control

messages for rekeying), is able to eliminate malicious data from the network in an

energy-efficient manner. It is also shown that this framework performs better than


44

other comparable schemes in the literature with an overall 60-100 percent

improvement in energy savings without the assumption of a reliable medium access

control layer. Recent advances in distributed in-network data storage and access

control have led to active research in efficient and robust data management in wireless

sensor networks (WSNs). Although numerous schemes have been proposed this far,

most of them do not provide enough attention towards exploiting user hierarchy and

sensor heterogeneity, which is quite a practical issue especially when deploying

WSNs in mission-critical application scenarios. [95] propose an efficient secret-key

cryptography-based (SKC) fine-grained data access control scheme for securing both

distributed data storage and retrieval. In this design, secret keying information for

data encryption and decryption are constructed based on the scheme of Blundo et al.

with information-theoretic security. To further enhance the security strength, author

then propose an efficient user revocation scheme based on the idea of blinded Merkle

hash tree construction. Extensive performance analysis shows that the proposed

schemes are very efficient and practical for WSNs. [92] proposes a new lightweight

authenticated encryption mechanism based on Rabbit stream cipher referred to as

Rabbit-MAC, for wireless sensor networks (WSNs) that fulfils both requirements of

security as well as energy efficiency. The proposed scheme provides data

authentication, confidentiality and integrity in WSNs. Rabbit based MAC function is

constructed, which can be used for data authentication and data integrity. A security

protocol is an idea for resource constrained WSNs is proposed, and can be widely

used in the applications of secure communication where the communication nodes

have limited processing and storage capabilities while requiring sufficient levels of

security. The features of Rabbit-MAC scheme conclude that this particular scheme

might be more efficient than the existing schemes in terms of security and resource

consumption. The security comes more to the fore. [91] presents SecSens, an

architecture that provides basic security components for wireless sensor networks.

Since robust and strong security features require powerful nodes, it uses a

heterogeneous sensor network. In addition to a large number of simple (cheap) sensor

nodes providing the actual sensor tasks, there are a few powerful nodes (cluster

nodes) that implement the required security features. The basic component of SecSens

offers authenticated broadcasts to allow recipients to authenticate the sender of a

message. To protect the sensor network against routing attacks, SecSens includes a

probabilistic multi-path routing protocol, which supports the key management and the


45

authenticated broadcasts. SecSens also provides functions to detect forged sensor data

by verifying data reports en-route. SecSens is successfully evaluated in a real test

environment with two different kinds of sensor boards.Routing in wireless sensor

networks is different from that in commonsense mobile ad-hoc networks. It mainly

needs to support reverse multicast traffic to one particular destination in a multihop

manner. For such a communication pattern, end-to-end encryption is a challenging

problem. To save the overall energy resources of the network sensed, data needs to be

consolidated and aggregated on its way to the final destination. [96] presents an

approach that 1) conceals sensed data end-to-end by 2) still providing efficient and

flexible in-network data aggregation. The aggregating intermediate nodes are not

required to operate on the sensed plaintext data. A particular class of encryption

transformations and discuss techniques for computing the aggregation functions

"average” and "movement detection” is applied. It shows that the approach is feasible

for the class of "going down” routing protocols. The risk of corrupted sensor nodes by

proposing a key pre-distribution algorithm that limits an attacker's gain and show how

key pre-distribution and a key-ID sensitive "going down” routing protocol help

increase the robustness and reliability of the connected backbone is considered.

S.Muhammad et al. [87] defined that sensor network comprises of scattered sensor

nodes with limited computational capabilities and battery power. The present security

solutions for traditional wireless networks cannot be used because of the constraints

associated with sensor network. A secure sink node architecture as two-tiered scheme

for sensor network security is presented. The architecture protects the sink node from

unauthorized access by surrounding it with two protection layers. Sink nodes listen to

only inner layer nodes and inner nodes are allowed to communicate with only outer

layer nodes. These protection layers are formed in an intelligent manner without

violating constraints specific to sensor network. In order to enhance security,

protection layers are re-adjusted in case of an attack. Statistical analysis to elucidate

the performance of proposed architecture is also presented.

Wireless sensor networks are researched extensively over the past few years. They

were first used by the military for surveillance purposes and have since expanded into

industrial and civilian uses such as weather, pollution, traffic control, and healthcare.

One aspect of wireless sensor networks on which research conducted is the security of

wireless sensor networks. These networks are vulnerable to hackers who might go

into the network with the intent of rendering it useless. An example of this would be


46

an enemy commandeering a drone and getting it to attack friendly forces. In this

paper, we review the security of wireless sensor networks. Areas that are covered

include: architectures and routing protocols, security issues that include context and

design as well as confidentiality, integrity, and authenticity, algorithms, and

performance issues for wireless sensor network design. Performance of the Self-

Originating Wireless Sensor Network (SOWSN), Practical Algorithm for Data

Security (PADS), and mechanisms for in-network processing were investigated in

further detail with SOWSN having the best performance as a result of it being based

on realistic scenarios [88]. The security solutions for generic wireless sensor networks

cannot be directly used in smart grid WSNs. In [89] the applications of sensor

networks in electric power systems are discussed and analyzed first. Then, the

characteristics of smart grid WSNs are summarized. Threats and security

requirements special for wireless sensor networks used in smart grid systems are

presented. Based on these works, reference security architecture was proposed to

guide the development and the design of the security solutions of wireless sensor

networks in smart grid systems, considering the information security requirements of

electric power systems. Moreover, open security issues needed solve to protect WSNs

applied in smart grid, and research challenges are introduced. A wireless network

solution can support mobility and flexibility of nodes in a network. Especially in

sensor networks, it has many advantages to replace cables with wireless logical links.

On the other hand, Bluetooth is generally considered as a promising short-range

wireless technology because of its inexpensive cost, low power and small size, and

thus Bluetooth has been gaining increasing interest from various industries. For the

above reasons, we adopt Bluetooth technology for a wireless sensor network, which is

designed for security systems. Since Bluetooth will continue to be a feature found in

many devices, it is worthwhile to investigate its use in wireless sensor networks. In

[90] a Bluetooth wireless sensor network for security systems is described, which

includes the implementation issues about system architecture, power management,

self-configuration of network, and routing. The methods or algorithms described in

this paper can be easily applied to other embedded Bluetooth applications for wireless

networks.

Distributed sensor data storage and retrieval have gained increasing popularity in

recent years for supporting various applications. While distributed architecture enjoys

a more robust and fault-tolerant WSNs, such architecture also poses a number of


47

security challenges especially when applied in mission-critical applications such as

battlefield and e-healthcare. First, as sensor data are stored and maintained by

individual sensors and unattended sensors are easily subject to strong attacks such as

physical compromise, it is significantly harder to ensure data security. Second, in

many mission-critical applications, fine-grained data access control is a must as

illegal access to the sensitive data may cause disastrous results and/or be prohibited

by the law. Last, sensor nodes usually are resource-constrained, which limits the

direct adoption of expensive cryptographic primitives. To address the above

challenges, in [97] a distributed data access control scheme that is able to enforce

fine-grained access control over sensor data and is resilient against strong attacks such

as sensor compromise and user colluding. The proposed scheme exploits a novel

cryptographic primitive called attribute-based encryption (ABE), tailors, and adapts it

for WSNs with respect to both performance and security requirements. The feasibility

of the scheme is demonstrated by experiments on real sensor platforms.

Wireless multimedia sensor networks (WMSNs) support many acoustic applications

for audio surveillance, animal tracking/vocalization, human health monitoring, etc.

However, resource constraints in sensor networks (such as limited battery power,

bandwidth/computation capability, etc.) pose challenges for the quality and security

of audio data transmission and processing. The security is a critical issue since audio

information can be accessed or even manipulated in WMSNs. In order to ensure

security, audio quality and energy efficiency, an index-based selective audio

encryption scheme for WMSNs is proposed. The scheme protects data transmissions

by incorporating both resource allocation and selective encryption based on modified

discrete cosine transform (MDCT). In this proposed work, the audio data importance

is leveraged using the MDCT audio index, and wireless audio data transmission

proceeds with energy efficient selective encryption. The simulation results depicts

that the proposed approach offers a significant gain in terms of energy efficiency,

encryption performance and audio transmission quality.

Some wireless sensor networks (WSNs) must transmit the data to users securely and

quickly. The sensor nodes just have limited computation, communication and storage

capabilities. Moreover adversary can easily eavesdrop in the process of message

transmission, the data is protected through encryption. In [93] present a new data

encryption scheme for the transmission, which uses the lightweight cryptography and

lets several sensor nodes to encrypt and transmit data cooperatively, thus it can


48

lighten the load of single sensor node, ensure the communication security, and

provide good load balancing and a higher lifetime of the network.

2.8. Communication Protocol Architectures

The protocol stack used in wireless sensor networks combines power and routing

awareness, integrates data with networking protocols, communicates power

efficiently through the wireless medium, and promotes cooperative efforts of sensor

nodes. The protocol stack consists of the application layer, transport layer, network

layer, data link layer, physical layer, power management plane, mobility management

plane, and task management plane as show below.

Figure 2.1 WSNs Protocol Stack

_____________________________________________________________________

2.8.1. Physical Layer

The physical layer is the first level of the protocol stack. It performs services

requested by the data link layer. The physical layer is the most basic network layer,

providing only the means for transmitting raw bits rather than packets over a physical

data link connecting network nodes. No packet headers or trailers are consequently

added to the data by the physical layer. The bit stream may be grouped into code

words or symbols and converted to a physical signal that is transmitted over a

physical transmission medium, which is the wireless medium in WSNs. The physical

layer provides an electrical, mechanical, and procedural interface to the transmission


49

medium. Broadcast frequencies, the modulation scheme used, and similar low-level

features are specified in the physical layer. The physical layer determines the bit rate,

also known as the channel capacity, digital bandwidth, maximum throughput, or

connection speed.

A variety of physical layer wireless transmission technologies are used in traditional

wireless networks. Considering the specific physical-layer requirements of wireless

sensor networks and taking into consideration the particular characteristics and usage

scenarios, it can be inferred that spread-spectrum technologies meet the requirements

much better than narrowband technologies. Besides, ultra wideband technologies are

found to be a promising emerging alternative.

2.8.2. Data Link Layer

The data link layer is responsible for multiplexing data streams, data frame detection,

medium access, and error control. It ensures reliable point-to-point and point-to-

multipoint connections in a communication network. The most important tasks of the

link layer are the formation and maintenance of direct communication associations

(„„links‟‟) between neighboring nodes and the reliable and efficient transfer of

information across these links. Reliability has to be achieved despite time-variable

error conditions on the wireless link. Nevertheless, the collaborative and application-

oriented nature of the sensor networks and the physical constraints of the nodes, such

as energy and processing limitations, determine the way in which these

responsibilities are fulfilled.

This layer is subdivided into Logical Link Control (LLC) and Medium Access

Control (MAC). In WSNs the fundamental design issue is the MAC. MAC protocols

solve a seemingly simple task of coordinating when a number of nodes access a

shared communication medium. We will explain the specific requirements and

problems of a WSNs MAC layer and present the fundamental MAC protocols.

2.8.3. MAC Requirements for WSNs

Medium Access Control design in sensor networks is very different from traditional

wireless MAC schemes due to the inherent WSNs limitation, among them the energy

constraint. The MAC protocol in a wireless multi-hop, self-organizing sensor network

must achieve two main goals


50

1. Creating network infrastructure: Since thousands of sensor nodes are densely

scattered in a sensor field, the MAC scheme should establish communication links

for data transfer. This forms the basic infrastructure needed for wireless

communication and gives the sensor network self-organizing ability.

2. Efficiently using and sharing energy and communication resources between

sensor nodes: Novel protocols and algorithms are needed to effectively tackle the

unique resource constraints and application requirements of sensor networks,

which means that MAC schemes in other wireless networks cannot be adopted into

the sensor network scenarios. Mobility also poses unique challenges to MAC

protocol design since weak mobility implies topology changes, while strong

mobility means new nodes or node failures.

WSNs requirements are different from those of traditional wireless networks. The

additional requirements come principally from the need to save energy. The

importance of energy efficiency for MAC protocols design is relatively new; thus,

many of the classical protocols like ALOHA and CSMA (Carrier Sense Multiple

Access) do not take this requirement into account. Other typical performance

characteristics such as fairness, throughput, or delay have played a minor role in

WSNs, yet recently they have been receiving more attention.

In WSNs, scalability and robustness requirements are confronted with the frequent

changes in the topology, which are generally produced by temporary power decreases

in nodes, node mobility, new node deployment, or „„death‟‟ of existing nodes. The

need for scalability is evident when considering very dense WSNs with dozens or

thousands of nodes.

Good collision management is also important since it can be useful for saving energy,

both in transmission from the source node and in reception at the destination node.

Collisions should be avoided by design (fixed assignments/ TDMA or assignments

under demand protocols) or by suitable collision suppression procedures to offset the

hidden-terminal problem in CSMA protocols.

Low complexity must be fulfilled by the MAC protocol for WSNs related to energy

savings. Because the nodes used in WSNs are simple, they should not consume an

exceptional amount of resources such as memory, energy, or processing power.

Accordingly, computationally expensive operations, such as complex scheduling

algorithms, should be discarded.


51

Most of the MAC protocols are classified in two groups: contention-based or

schedule-based. The difference is the number of contestants that have the option of

transmitting to a node at a given instant:

In contention-based protocols, any node can try to transmit with the risk of

collisions. As all nodes have to contend for the communication channel,

collisions are possible and are one of the major causes of energy inefficiency.

Consequently, these protocols have several mechanisms to suppress collisions or

to reduce the probability of occurrence. In a contention-based wireless sensor

network, since nodes can directly transmit information to the base station at any

time, idle listening can also occur. This is one of the main sources of energy

waste in these networks since the nodes normally remain inactive for a long time

without transmitting. The benefit of these protocols is their simplicity and

robustness.

In schedule-based or polling-based protocols, only one neighbor has the

opportunity to transmit at any given time, thus eliminating collisions. These

protocols usually have a TDMA component, which also provides an implicit

mechanism of passive listening suppression. When a node knows the slots it has

been assigned, it is sure that the communication, both transmission and reception,

will only be produced at these slots; otherwise, the receptor can be deactivated.

This scheme is much more complicated since the base station must poll the nodes

and then gives each one a time to transmit. The constraint of these protocols is

the large amount of data transmitted to set up the network structure. However,

once the structure is created, there is no chance of collisions and nodes can save

energy in their operation.

2.8.4. Network Layer

The network layer is the third level in the WSNs protocol stack. It responds to service

requests from the transport layer and issues service requests to the data link layer. In

essence, the network layer is responsible for end-to-end, i.e., source-to-destination,

packet delivery, whereas the data link layer is responsible for node-to-node, i.e., hop-

to-hop, packet delivery. The network layer provides the functional and procedural

means of transferring variable-length data sequences from a source to a destination

via one or more networks while maintaining the quality of service requested by the


52

transport layer. The network layer performs network routing, flow control, network

segmentation/de-segmentation, and error control functions.

Due to the deployment characteristics of WSNs, multi-hop communication may be a

good choice for sensor networks with strict consumption and transmission power

level requirements. In a multi-hop network, intermediate nodes have to relay packets

from the source to the destination node. Those intermediate nodes have to decide

which neighbor to forward to. The construction and maintenance of the routing tables

needed for reaching the destination node is the crucial task of a distributed routing

protocol. This section discusses some mechanisms for routing and forwarding that can

be implemented by WSNs routing protocols. These mechanisms take into account

whether a unique node identifier identifies the packet, by a set of such identifiers, or

by all nodes in the network.

The network layer of the WSNs is usually designed according to the following

principles:

Energy efficiency is always an important consideration.

WSNs are mostly data-centric. Sensors do not usually have a unique ID, because

the overhead of ID maintenance is high. The data themselves are usually more

important than knowing which nodes send data.

An ideal WSN has attribute-based addressing and location awareness.

Data aggregation: Depending on the application, this can be useful although the

energy needed for data aggregation is sometimes higher than the savings.

The routing protocol needs to be easily integrated with other networks.

In some cases, the routing protocol must be QoS-aware, thus having specific

mechanisms related to the delay and reliability of the traffic flow.

These design principles serve as a guideline when designing a routing protocol

for sensor networks and are further explained to emphasize their importance.

2.8.5. Transport Layer

The second-highest layer in the WSNs protocol stack, the transport layer responds to

the services requested from the application layer and issues service requests to the

network layer. The transport layer provides dependable data transfers between hosts.

It is usually responsible for end-to-end error recovery and flow control and for

ensuring complete data transfer.


53

The purpose of the transport layer is to provide reliable data transfer services between

end users, thus relieving the upper layers‟ responsibility for providing reliable and

cost-effective data transfer. The transport layer usually turns the unreliable and very

basic service provided by the network layer into a more powerful one. There is a long

list of services that can be optionally provided at this level, although none is

compulsory. Since not all applications require all services available, some can be

wasted overhead or even counterproductive in some cases.

Some of the particular challenges for transport protocols in WSNs include the

following:

WSNs are multi-hop wireless networks with homogeneous/heterogeneous nodes.

TCP has several drawbacks when used over wireless channels; thus, a WSNs is not

an easy environment for TCP.

Any transport protocol must adapt to the stringent energy constraints, memory

constraints or computational constraints of sensor nodes. Significant engineering

efforts would be required to run heavyweight protocols like TCP on such nodes.

Generally, transport protocols do not have good behavior with dynamic topologies.

2.8.6. Application Layer

The application layer is the last level of the WSNs protocol stack. It interfaces directly

with the application, performs common services for the application processes, and

issues requests to the transport layer. The common application layer services provide

semantic conversion between associated application processes. The application layer

of the five-layer WSN protocol stack corresponds to the application layer, the

presentation layer, and the session layer in the seven-layer OSI model.

Although many application areas for sensor networks have been defined and

proposed, potential application layer protocols for sensor networks remain a largely

unexplored region. Some application protocols for WSNs are the Sensor Management

Protocol (SMP), the Task Assignment and Data Advertisement Protocol (TADAP),

and the Sensor Query and Data Dissemination Protocol (SQDDP).

2.9. Various Middleware WSNs Approaches

Different middleware approaches were selected and classified taking the

programming models used into account.


54

Programming Wireless Sensor Networks

Programming Abstraction Programming Support

1. Global Behavior

2. Local Behavior

1. Virtual Machine

2. Data Base

3. Modules

4. Application Driven

5. Message-oriented

6. Middleware

Figure 2.2. Programming Models

Programming sensor networks includes two major classes shown in Figure 2.2.The

first one is programming support, which manages the providing systems, services, and

run-time mechanisms, such as reliable code distribution, safe code execution, and

application-specific services. The second one is programming abstraction, which is

related to the way a sensor network is viewed and presents concepts and ideas of

sensor nodes and sensor data.

2.10. Programming Support

The programming support class consists of five approaches virtual machine–based,

modular programming–based, database-based, application-driven, and message-

oriented middleware as shown in Figure 2.2.

2.10.1. Virtual Machine

This approach consists of virtual machines (VM), interpreters, and mobile agents. Its

main characteristic is flexibility, allowing developers to write applications in divided

small modules, which are injected and distributed through the network by the system

using tailored algorithms and then interpreted by the VM. Those tailored algorithms

minimize the overall energy expenditure as well as resource use. However, the

technology is complex and the instructions introduce overhead.

2.10.2. Modular Programming (Mobile Agents)

The use of mobile code facilitates the injection and distribution through the network

and leads to application modularity. Less energy is necessary when broadcasting


55

small modules instead of the complete application.

2.10.3. Database

This approach observes the entire network as a virtual database system, offering an

easy-to-use interface that permits the user to extract data of interest and issue queries

about the sensor network. Nevertheless, this approach does not support real-time

applications, as it provides only approximate results and the detection of spatial-

temporal relationships between events is not possible.

2.10.4. Application-Driven

This approach establishes a new, innovative aspect in middleware research by

complementing an architecture that accomplishes the network protocol stack,

enabling programmers to adjust the network according to the exact application

requirements. It provides a QoS advantage since the applications determine the

network operations management.

2.10.5. Message-Oriented Middleware (MOM)

This approach is essentially a communication model in a distributed-sensor network.

The system facilitates message exchange between nodes and the sink nodes by means

of a publish-subscribe mechanism. This model supports asynchronous

communication, making movable combinations between the sender and receiver

possible.

2.10.6. Characteristics of WSNs Middleware

WSNs middleware should support the implementation and basic operation of a sensor

network while taking into consideration some of the unique characteristics of WSNs:

Sensor nodes are small-scale devices (with volumes approaching a cubic

millimeter in the near future).

Sensors are limited in the amount of energy stored and/or harvested from the

environment.

Sensors are likely to fail, due to depleted batteries or to environmental

influences.

Sensors have restricted resources (CPU performance, memory, wireless


56

communication bandwidth and range).

Node mobility, node failures, and environmental obstructions cause frequent network

topology changes. Communication failures are also a typical problem in wireless

sensor networks. Another issue is heterogeneity since the network may consist of a

large number of rather different nodes in terms of sensors, computing power, and

memory. On the one hand, the large number raises scalability issues; on the other

hand, it provides a high level of redundancy. Nodes also have to be able to operate in

unattended mode since it is impossible to service a large number of nodes in remote

or inaccessible locations. In order to deal with the characteristics outlined above,

WSNs middleware must face the following challenges:

Supporting the development, maintenance, deployment, and execution of

sensing based applications. This includes mechanisms for defining complex,

high-level sensing tasks, communicating these tasks to the WSNs, coordinating

sensor nodes to split and distribute the tasks to each node, gathering data to

merge the sensor readings of the individual sensor nodes into a high-level result,

and reporting the results back to the task issuer.

Working in a network with a great number of wirelessly connected nodes

(sensors).

Providing abstraction of the network for heterogeneity among the different

components of the WSNs.

Fulfilling the main requirements of WSNs, namely, energy efficiency, reliability,

and scalability, allowing event-based or periodic communications. These

approaches represent the characteristics of the WSNs better than the traditional

scheme (based on requests and responses). Providing support for automatic

configuration and fault management, which are necessary for the unattended

way the nodes operate.

Paying attention to the concepts of time and location. These are the key elements

for unifying the information obtained by the different sensors.

Providing application knowledge in nodes. Middleware for WSNs has to provide

mechanisms for injecting application knowledge into the infrastructure and the

WSNs.


57

2.11. WSNs Topologies and Deployment Methodologies

The term „„topology‟‟ refers to the physical disposition in which the nodes of a

network (in this case, a WSNs) are connected to one another. Network topology only

refers to node connections. The distance between nodes, physical inter- connections,

transmission rates, or types of signals do not belong in this category, although they

can be influenced by the topology. However, a good WSNs design takes the topology

into account when improving several performance factors such as energy efficiency,

robustness, or general QoS parameters.

Figure 2.3. Types of Sink

To understand the topologies of a WSNs, the types of nodes that form the network

first need to be introduced. WSNs contain both sources and sinks. A source can be

any entity in the network that is able to provide information. It is usually a sensor

node, but it can also be an actuator node that provides feedback about an operation.

On the other hand, a sink is the entity that requires information. There are two

possibilities for a sink: It can belong to the WSNs and be just another sensor/actuator

node, or it can be an external entity. If the sink is an actuator belonging to the WSNs,

it could be, for example, a laptop used to interact with the sensor nodes. If it is an

external element, the sink may be a gateway to another network such as the Internet,

where the information requests come from some external device/node indirectly

connected to the WSNs. These main types of sinks are illustrated in Figure 2.3.


58

The types of network topologies can be classified according to several criteria. In

addition, the network hierarchy should be taken into account when selecting a suitable

and efficient routing scheme. In fact, the main WSN topology division is based on the

existence or absence of hierarchy among network elements.

In flat networks or those networks without hierarchy each node has the same

capabilities. Thus, control over the routes and channels must be performed in a

distributed fashion.

In hierarchical networks, some nodes will have different capabilities than others.

These capabilities are divided into two areas: physical, where the nodes or links have

different physical characteristics, and logical, in which the nodes have different

functions in the network.

The most common and representative WSNs topologies are the following

Figure 2.4. WSN topologies, (a) Ad-Hoc Network, (b) Clustered Network, (c) Overlay Network

2.11.1. Ad hoc without Hierarchy

In this case, all nodes are equal. They are their own service providers, and thus data

pass from node to node to reach a sink. A common example of this type of network is

the mobile ad hoc networks (MANETs), although this scheme can also be valid for

networks formed by nodes of low or no mobility.

2.11.2. Hierarchical network by Clustering

The idea of hierarchy implies assigning some nodes with a special role, for example,

controlling neighboring nodes. In this sense, local groups or clusters can be formed;


59

the „„controllers‟‟ of such groups are often referred to as cluster heads. The major

functions of the cluster heads are local resource arbitration (i.e., in MAC protocols),

making routing tables more stable since all traffic is routed through the cluster heads,

and making higher-layer protocols more scalable since the higher layer perceives a

less complex network due to clustering. Furthermore, cluster heads are the usual

places where the traffic is aggregated and com- pressed to converge to a single sink.

2.11.3. Overlay networks

This type of clustering network has both physical and logical hierarchies. Nodes that

assume special control functions are thus more powerful and/or have privileged

capacities with respect to the rest. This way, the more powerful nodes may form a

network on their own allowing higher scalability. An example of this type of network

is a cellular network, where base stations, which, in turn, are connected to a wired

infrastructure for, control cells inter cell routing. Another possible topology in a

network is called a mesh topology. In this case, all sensor nodes are identical and can

communicate directly with each other, providing a high level of redundancy in the

data paths between nodes. In a mesh WSNs, every node should be in the area of radio

coverage of any other node, which is a disadvantage since nodes in WSNs have,

limited power. This reduced available energy makes it unviable to implement a WSNs

with mesh topology if the coverage area exceeds certain dimensions, for example in

environmental or agriculture applications, or if it has a strong attenuation, such as

inside buildings. Therefore, in many situations it is necessary to accept a topology that

is not completely meshed, having some nodes route the information of others without

being the packet destination. This is known as multi-hop routing.

Figure 2.5. Multi-hop Routing

A topology with multi-hop routing has both advantages and disadvantages with


60

respect to a meshed topology. Some advantages are the following:

Not only is the multi-hop routing a functional solution for solving problems with

large distances or obstacles, but it also has been used for improving energy

efficiency in communications. The radio channel attenuation increases at least at

a quadratic rate with the distance in most environments, thus wasting less energy

with a multi-hop architecture than with single-hop topologies. The global power

consumption is lower if the nodes transmit to other neighboring nodes than in a

hypothetical situation in which every node transmits directly to a sink or

gateway.

If the density of intermediate nodes (relays of the information) is larger, the

reutilization frequency distance is shorter. Therefore, the global capacity for data

transmissions increases.

For several applications it is very convenient to carry out data aggregation in the

intermediate nodes instead of transmitting all the raw data generated by nodes. A

multi-hop routing allows the nodes that route the information to aggregate the

data received with their own data and transmit only the summarized or

aggregated information. Sending less information increases both the global

information transmission capacity of the network and its lifetime, thus saving

energy.

Among the disadvantages of multi-hop routing are (1) the larger delay between

generating the information and its reception by the sink and (2) some applications‟

requirement for controlled delays. The reasons for these issues are the following:

Each packet will be queued inside each of the nodes through which it is routed,

producing larger and, in general, variable delays. If the percentage of resource

use in the WSNs is low, this delay may not be significant, although the

application determines what it significant and what is not.

Even if queuing delays are not significant, the functioning of the MAC protocol

in each hop may add an important amount of time. For instance, in many MAC

protocols, the nodes switch from states of low or null activity with very low

energy consumption to activity states in which they may send or receive data.

Waiting for an active period of a neighboring node in order to send messages

may be a source of delay. Another example can be found in MAC protocols

designed for star topologies, in which a node must join the master of the

neighboring star nearest the destination prior to forwarding a packet. Allowing a


61

node to be in the influential area of more than one master node may alleviate this

problem.

2.12. Design Strategies and Operation of WSNs Software

A proposal for possible WSNs software design cycle can be found in Blumenthal et

al. (2003). They proposed a software organization with an intermediation software

layer (middleware) above the operating system. Its aim is to provide services to the

applications. Blocks represent all the components of this architecture. Figure 2.6

shows the proposed software development cycle for these types of network.

Figure 2.6 Software Development Cycle

The structure of the running software per Blumenthal et al.‟s proposal is shown in

Figure 2.7 with continuing advancements in sensor node design and increasingly

complex applications, an interest in design automation of sensor network applications

is inevitable. The objective is to eventually enable domain experts to be able to design

and analyze algorithms, and automatically synthesizes programs for an abstract

machine model of the underlying system, without requiring knowledge of low-level

networking aspects of the deployment.

Component

Design & Edit

Compilation/linked

Evaluation

Node Software

Identify and include each block component

Interconnect components & resolve dependencies. Optimize parameters

Executable creation

Run. Monitor & Evaluate


62

Figure 2.7. Software Architecture

2.13. Software Architecture in WSNs

In typical WSNs deployments, several types of applications and software coexist in

different hardware platforms, such as sensor nodes, server nodes, or gateways, and

client equipment. WSN architecture has different layers, as shown in Figure 2.7.

Several authors coincide with this layer decomposition, although there are some

differences in nomenclature.

As depicted in Figure 2.8 the software architecture consists of three different layers:

the mote layer, the server layer, and the client layer.

Figure 2.8. WSNs Software Architecture

The mote layer is composed of the motes with their sensors. In this layer, the


63

software needs to include the light operating system executed and the

corresponding applications necessary to obtain a service, i.e., environment

monitoring or intruder tracking. These applications are usually developed for the

specific hardware on which they are going to run, and the programming

language used must fit well with highly restricted devices; nesC can be an

adequate programming language. nesC was created specifically to adapt

application programming in embedded network systems, a category that includes

WSNs. The main characteristics of nesC‟s design were inspired by the TinyOS

operating system: event-based execution, incorporation of a concurrence model,

and component-based application design. In fact, TinyOS has been

reimplemented in the nesC language.

The server layer receives the information from the WSN by using proprietary

protocols and stores it in databases. It also offers services, usually by means of

TCP/IP interfaces, in order to allow interested clients to access this information.

In this case, the programming languages are selected not based on the ack of

resources, but rather on their portability to execute in machines with different

characteristics and different general-purpose operating systems for different

platforms.

The client layer includes a graphical user interface that allows the information,

topology, and state of the WSNs as well as its management to be seen. This

software must provide the user with the information needed for managing the

WSNs, interpreting the large amounts of information generated, and monitoring

the network‟s health.

2.14. Network Simulation and Commonly used Simulators

Wireless sensor networks have tremendous potential to monitor, study, and analyze

phenomena in the physical world in detail never before available, in places too far,

too deep, too high, or too dangerous for researchers to go. Simulation can be of great

help to ensure the shortest possible time to market and to minimize the overall cost of

WSNs design. Being that the cost, time, and complexity involved in deploying and

constantly changing large-scale WSNs are prohibitively high, simulation is a cost-

effective choice for the rapid exploration and validation of WSNs applications.

Simulation provides controlled and repeatable environmental conditions for


64

evaluating and optimizing the design parameters and/or the configuration alternatives.

It also offers very good insight into the effects of the various parameters and thus

helps identify those that have the greatest importance for the system‟s operation.

The simulation of wireless networks is inherently different from that of wired

networks. The signal interference and attenuation concerns are more complicated for

wireless media than for wired media. The broadcast nature of wireless radio

transmission also makes communication topology in simulation models relatively

denser than for an equivalent wired network. Specific features like node mobility,

distributed behavior, power, and terrain models dramatically increase the computation

effort of WSNs simulators. Consequently, accurate fine-grained WSNs simulations

present a significant challenge.

The sensor nodes detect the stimuli, i.e., signals, generated by the target nodes over a

sensor channel and forward the detected information to the sink nodes over a wireless

channel. Two different models for signal propagation are therefore included: a sensor

propagation model and a wireless propagation model.

The existing simulator tools are either commercial or open source and are mainly

developed in Java or C++. These simulators strongly differ with respect to features

such as

Scalability: The capacity to simulate a high number of nodes with sufficient

precision in a finite time.

Real-time requirement: Real-time or close to real-time simulation can be related to

the simulation architecture and the programming language and to some other

features as well.

Software emulation: The possibility to run the various software during simulation

as if they were running on the real processor also called „„software in the loop.‟‟

Hardware emulation: The ability to accurately simulate the behavior of various

hardware parts of the node and to evaluate some important parameters such as

consumption, memory use, collisions, etc.

Model fidelity: Availability of detailed models for various aspects of networking

such as propagation, protocols, mobility, etc.

Reliability study: Possibility to simulate the occurrence of various faults or defects,

i.e., hardware, physical, noise, etc., during the network simulation and to visualize

their effects on the network‟s behavior or on some parameters such as


65

consumption, delays, etc.

X in the loop: The possibility of connecting other (hardware and/or software)

devices to the simulator for various purposes; simulation-time reduction by

replacing a hardware simulation model by a real hardware system, either existing

or simulated on an FPGA or other hardware platform, performance analysis of

protocols by injection of real signals, reliability study in the presence of real

perturbations, etc.

In communication and computer network research, network simulation is a technique

where a program models the behavior of a network either by calculating the

interaction between the different network entities (hosts/routers, data links, packets,

etc) using mathematical formulas, or actually capturing and playing back observations

from a production network. The behavior of the network and the various applications

and services it supports can then be observed in a test lab; various attributes of the

environment can also be modified in a controlled manner to assess how the network

would behave under different conditions. When a simulation program is used in

conjunction with live applications and services in order to observe end-to-end

performance to the user desktop, this technique is also referred to as network

emulation.

Positive Train Control (PTC) refers to microprocessor-based communication

technologies that are capable of preventing train collisions, derailments, and injuries

to workers operating within the railroad system. In North America, there are 11

competing PTC projects in various stages of refinement. The North American Joint

Positive Train Control Project (NAJPTC) is one of those efforts, based on the

Advanced Train Control System (ATCS) protocol. NAJPTC is a joint development

project of the Association of American Railroads, the Federal Railroad

Administration, and the Illinois Department of Transportation. Paul Vincent et al.[63]

used a network simulation system (NS-2) to model NAJPTC ATCS communications

to determine the feasibility of that PTC system when overlaid onto geographic data

from Google Earth, and digital elevation data from the United States Geological

Survey. The Simulations were useful in observing and tracking the signal strength of

a moving train, resolving packet collisions via strategic base station placement, and

modeling and mitigating communication losses.

A network simulator is a software program that imitates the working of a computer

network. In simulators, the computer network is typically modeled with devices,


66

traffic etc and the performance is analyzed. Typically, users can then customize the

simulator to fulfill their specific analysis needs. Simulators typically come with

support for the most popular protocols in use today, such as WLAN, Wi-Max, UDP,

and TCP.

Some of the most relevant academic WSN simulators are presented below. Often

these simulators are still under development; some of them are well suited to help

with research.

OMNeT++

OMNeT++ is an open-source tool that shares many concepts, solutions, and features

with OPNET OMNeT++is a discrete-event, component-based, modular, and open-

architecture simulation environment with strong GUI support and an embeddable

simulation kernel.OMNeT++provides component architecture for models.

Components, i.e., modules, are programmed in Cþþand then assembled into larger

components and models using a high-level language (NED).

GloMoSim (Global Mobile Information System Simulator)

A simulation environment for purely wireless mobile networks, GloMoSim was

designed as a set of modules in architecture structured into eight layers. Each module

simulates a specific protocol in the protocol stack. GloMoSim has been designed

using the parallel discrete-event simulation capability provided by PARSEC

(PARSEC), a C-based sequential and parallel simulation language that can be used to

program new modules that can be added to GloMoSim. GloMoSim offers different

protocols to model node mobility and radio communication. GloMoSim has already

been used to simulate networks with thousands of wireless nodes and provides a rich

set of models for both existing and novel protocols at multiple layers of the protocol

stack. Apparently, GloMoSim does not offer environment or power models.

Ptolemy

This is an ongoing project at UC Berkeley that studies the modeling, discrete-event

simulation, and design of concurrent real-time embedded systems. The key

underlying principle in Ptolemy is the ability to use multiple computation models

(e.g., continuous-time, data flow, finite state machine) in a hierarchical heterogeneous

design environment. Ptolemy does not support network emulation but does support

both wireless network and sensor network simulations.


67

NS-2

This is a discrete-event simulator that provides support for TCP, routing, and

multicast protocols, among many others. The support for wireless and mobile network

simulation provides various modules for mobile wireless network simulation, such as

radio propagation models, the IEEE 802.11 MAC protocol, mobility models, different

ad hoc routing protocols (e.g., AODV and DSR), and Mobile IP. The latest version of

ns-2 supports the simulation of pure wireless LANs, multiple-hop ad hoc networks,

and combined simulation of wired and wireless (known as „„wired-cum-wireless‟‟)

networks. Maintaining real code in ns-2 is not transparent.

QualNet

QualNet is a network simulation tool that simulates wireless and wired packet mode

communication networks. QualNet Developer is a discrete event simulator used in the

simulation of MANET, WiMAX networks, satellite networks and sensor networks,

among others. QualNet has models for common network protocols that are provided

in source form and are organized around the OSI Stack. QualNet is a commercial tool

derived from GloMoSim that was first released in 2000 by Scalable Network

Technologies (SNT). Today, the main differences between QualNet and GloMoSim

are

QualNet is based on C++; GloMoSim is based on PARSEC C (a C based parallel

simulation language).

QualNet is a commercial product; GloMoSim is distributed under an academic open

source license.

QualNet is maintained by SNT; UCLA Parallel Computing Lab maintains

GloMoSim.

2.15 Summary

In this chapter review of related published literature of various authors of the field of

WSNs is presented. Basic theoretical concepts and factors related to field in question

is also explained. From this chapter we also get the idea of new researches and

experiments, which will be helpful to compare and evaluate our work. In the next

chapter system architecture of our 3 L-S model will be presented.

CHAPTER 2 LITERATURE SURVEY - Shodhgangashodhganga.inflibnet.ac.in/bitstream/10603/8869/7/07_chapter 2.pdf · Chapter 2. Literature Survey 24 transmission from sensor nodes to sink

Documents