1

Athina Bourdena1, Constandinos X. Mavromoustakis2, George Kormentzas1, Evangelos

Pallis3, George Mastorakis3, Muneer Bani Yassein4

1University of the Aegean, Department of Information and Communication Systems

Engineering, Samos, Greece

Tel: +302273082235, Fax: +302273082009, E-mail: abourdena@icsd.aegean.gr,

gkorm@aegean.gr 2University of Nicosia, Department of Computer Science, Nicosia, Cyprus

Tel: +35722841730, Fax: +357-22357530, E-mail: mavromoustakis.c@unic.ac.cy 3Technological Educational Institute of Crete, Department of Informatics Engineering,

Estavromenos, Heraklion, Crete, Greece

Tel: +302810379828, Fax: +302810370311, E-mail: pallis@pasiphae.eu,

gmastorakis@staff.teicrete.gr 4Jordan University of Science and Technology, P.O. Box 3030, Irbid 22110, Jordan

Tel: +962-2-7201000 Ext: 43403, Fax: +962-2-7201000, E-mail: masadeh@just.edu.jo

Abstract

This paper proposes a resource intensive traffic-aware scheme, incorporated into an

energy-efficient routing protocol that enables energy conservation and efficient data flow

coordination, among secondary communicating nodes with heterogeneous spectrum

availability in distributed cognitive radio networks. The proposed scheme associates the

backward difference traffic moments with the sleep-time duration to tune the activity

durations of a node for achieving optimal energy conservation and alleviating the uncontrolled energy consumption of wireless devices. Efficient routing protocol operation, as

a matter of maximum energy conservation, maximum-possible routing paths establishments

and minimum delays is obtained, by utilizing a signalling mechanism, developed based on a simulation scenario that includes a number of secondary communication nodes. The validity

of the proposed resource intensive traffic-aware scheme and the energy-efficient routing

protocol is estimated and verified, by conducting experimental simulation tests and obtaining performance evaluation results. The simulation results validated the efficiency of the

proposed scheme and the effectiveness of the routing protocol, in terms of minimizing the

energy consumption and maximizing resources exchange between secondary communication

nodes in a distributed cognitive radio network.

Keywords: Cognitive Radio, Routing Protocols, TVWS, Traffic-aware Energy Conservation,

Energy-Efficient scheme, Ubiquitous Computing and Communications, Ad-hoc networks.

1. INTRODUCTION Cognitive Radio (CR) technology [1] is an emerging communication paradigm that

efficiently exploits radio spectrum resources to enable the deployment of future wireless networks. CR networks are comprised of communication nodes, capable of adapting their

technical characteristics, based on interactions with the surrounding spectral environment.

They can sense a wide radio spectrum range, dynamically identify locally unused/unexploited frequencies and efficiently access them. This capability opens up the possibility of designing

new dynamic radio spectrum access policies with the purpose of opportunistically reusing

under-utilized frequencies at local level, such as “television white spaces” (TVWS) [2].

A Resource Intensive Traffic-Aware Scheme

using Energy-Aware Routing in Cognitive

Radio Networks

2

TVWS comprise VHF/UHF radio spectrum portions that are either resulted by switchover

process from analogue to digital terrestrial television, or are completely under-utilized due to

frequency planning principles (“Interleaved Spectrum”) [3]. Therefore, introduction of CR

networks in TVWS represents a disruption to the current “command-and-control” paradigm

of TV/UHF spectrum management. The exploitation of CR technology is highly intertwined with the regulation models that would eventually be adopted [4], [5] especially in future

computing systems. The flexibility in radio spectrum access phase by CR networks caused

new challenges along with increased complexity in the design of communication protocols at different layers. More specifically, the design and adoption of efficient routing schemes, is a

vital process for such an emerging communication paradigm. CR networks are characterized

by completely self-configuring architectures [6], where routing is challenging and different

from routing in a conventional wireless network. A key difference is that spectrum

availability in a CR network highly depends on primary communication nodes presence.

Therefore, a fixed Common Control Channel (CCC) is difficult to be exploited, towards

establishing a stable routing path between secondary communication nodes. The specific

features of CR network architectures pose new requirements in handling energy efficient

resources along with an underlying reliable routing scheme. To this end, this work considers

the association of the routing mechanism utilized by CR systems with the traffic volume and the end-to-end mechanism for efficiently sharing the requested resources by nodes.

Energy conservation figures an important aspect for the high performance deployment in

ad-hoc CR networks. On one hand, the Energy Conservation scheme has to be reactive so that

the energy levels of wireless nodes will be tuned, according to the estimated parameters (i.e.

capacity, traffic [7] of the nodes). On the other hand, an energy-efficient scheme has to take

into consideration the bounded end-to-end delays of the transmissions. As the network

lifetime is closely related to the transmission characteristics [8] of a source node to a

destination node and the underlying routing protocol used [9], a mechanism that combines the

temporal traffic-aware behavior of the node [10] and the efficient routing scheme in an end-to-end path has to be investigated. In [8] the sleep-proxy nodes evaluate the duration of the

activity periods of each node, according to the capacity and the estimated inter-cluster overall

energy consumed within a time frame. Towards further investigating the scheme proposed in [8], this work has applied the traffic model and the characteristics of the volume of the traffic

for a specified time window frame to CR systems, supported by the Backward Traffic

Difference estimation. In order to minimize the energy consumption the Backward Traffic Difference measures the volume of the incoming Traffic that is destined for each one of the

nodes within a time window frame. The Backward Traffic Difference [10, 19] takes into

consideration the repetition of the Traffic and estimates the Backward Difference for

extracting the time duration for which the node is allowed to Sleep.

In this context, this paper elaborates on the design, development and experimental

evaluation of a resource intensive traffic-aware scheme incorporated into an energy-efficient

routing protocol for distributed CR network architectures. Moreover, the joint routing and

traffic-aware methodologies were never combined in the past to offer energy conservation in

CR systems. More specifically, a signalling mechanism combined with an energy efficient scheme is proposed, based on the Backward Traffic Difference estimation methodology

initially stated in [7]. The goal of this work is to achieve energy usage that scales with

loading. This is possible by using the incoming traffic aggregation for each node to adjust the volume of the traffic to the estimation of the activity time period assigned for each node. In

addition, this paper elaborates to describe the development and assessment through

simulation, of a novel solution for linear scaling adjustment of energy usage with all loads on

each secondary node without any packet loss. The key idea is based on traffic aggregation via

a traffic-aware mechanism. This mechanism occurs on each secondary node to obtain an

estimation and maximization for the time slot when the interfaces of each node are put to

sleep. Based on the underlying routing scheme and the volume of traffic that each node

receives/transmits, the proposed scheme aims at minimizing the energy consumption, by

3

applying asynchronous, non-periodic sleep-time assignment slot to the secondary wireless

nodes. Following this introductory section, section 2 elaborates on the related work and

research motivation, while section 3 presents the design and development of a novel green-

aware routing protocol, offering energy efficient data transition, across secondary

communication nodes with different TVWS availability. In order to achieve an energy-efficient methodology, the proposed framework uses a traffic-aware Backward Traffic

Difference scheme for estimating the duration of the sleep time according to the nodal

traversed traffic. The proposed scheme can efficiently determine the ON and OFF duration/period of each node by adjusting the traffic onto the activity periods of each mobile

node. The proposed scheme then effectively provides a reflection of the activity of the traffic

to the overall energy consumed by nodes. Finally, section 4 elaborates on the performance

evaluation analysis of the proposed research approach, discussing experimental results and

section 5 concludes this paper by highlighting directions for future research.

2. RELATED WORK AND RESEARCH MOTIVATION Conventional routing algorithms exploited in wireless ad-hoc networks, enable for the

optimization of network performance metrics, such as end to end delay, switching delay and

backoff delay. A rich literature on conventional routing protocols is available based on

network-wide broadcast messages, without using any local hops information. Such

approaches are not suited for wireless CR networks, since there is no support for concurrently

considering radio spectrum availability of secondary communication nodes, as well as the

effect on other primary nodes that share spectrum resources. In a general context, several

research approaches have been recently proposed in [11], [12], [13], [14], towards addressing

routing issues in CR networking environments. In addition, a routing protocol is proposed in

[15], exploited to combine geographical routing and radio spectrum assignment, towards avoiding regions with high presence of primary communication nodes. It also determines

optimum routing path channel combinations that reduce delays in the network. A spectrum

aware data adaptive routing algorithm is proposed in [16], where the end to end route selection depends on the amount of data to be transferred. Furthermore, the proposed routing

protocol in [17] builds a forwarding mesh, based on a set of available routes to the destination

and opportunistically adapts during the forwarding process, according to the dynamic radio spectrum conditions. Moreover, a joint approach of on-demand routing and spectrum band

selection is proposed in [18] for CR networking environments and a delay based metric is

used to evaluate the quality of alternative routes. Most of the previous schemes are based on

on-demand routing protocols and discover paths between source and destination

communication nodes.

On the other hand, the routing mechanism has to be strictly associated with the Energy-efficiency when the CR networking architecture hosts wireless nodes requesting spectrum, via

which the traffic will be transferred. Therefore the routing mechanism in collaboration with

an energy-efficient scheme should guarantee the end-to-end availability of requested

resources, whereas it should be able to significantly reduce the Energy Consumption. In

addition, the mechanism should be able to maintain the requested scheduled transfers and the

entire end-to-end connectivity. Many recent measurement studies [19] have convincingly

demonstrated the impact of Traffic on the end-to-end connectivity [20], and thus showed the

impact on the Sleep-time duration and the Energy Consumption. Measures extracted in real-

time using realistic traffic [19-20] have shown that the impact of the responsiveness of the

routing scheme in regards to the end-to-end transmission reliability is significant. Real-Time communication networks and multimedia systems, exhibit noticeable burstiness over a

number of time scales [21-22]. Based on the stochastic traffic modelling, the traffic in most of

the cases can be expressed in time exhibiting fractal-like characteristics [24]. The problem of hosting a scheme where, in collaboration with the routing mechanism used, takes into account

the traffic characteristics in order to conserve energy has not yet explored. The scheme will be

4

able to tune the wireless interfaces of the nodes to the Sleep or Active state according to the

incoming Traffic and a model which considers the next Sleep-time duration. Notwithstanding,

many Sleep-time scheduling strategies were introduced that model the node transition

between ON and OFF states. Existing scheduling strategies for wireless nodes could be

classified into three categories: the coordinated sleeping [23], where nodes adjust their sleeping schedules, the random sleeping [24], where there is no certain adjustment

mechanism between the nodes in the sleeping schedule with all the pros and cons [10], and

on-demand adaptive mechanisms [25], where nodes enter into Sleep-state depending on the environment requirements whereas an out-band signalling is used to notify a specific node to

go to sleep in an on-demand manner.

Although there are many schemes developed addressing different Energy conservation

methodologies, the combination of a traffic-aware scheduling scheme with the routing

protocol supported by the CR networking architecture, has not yet been explored. The latter

poses a fertile ground for the development of new approaches with the association of different

parameters of the communication mechanisms, in order to reduce the Energy Consumption.

Such schemes are classified into active or passive mechanisms. Active techniques conserve

energy by performing energy conscious operations, such as transmission scheduling and

energy-aware routing. Mavromoustakis et al. in [19] consider the association of Energy conservation problem with different parameterized aspects of the traffic (like traffic

prioritization) and enable a mechanism that tunes the interfaces’ scheduler to sprawl in the

sleep state according to the activity of the traffic of a certain node in the end to-end path in

real-time.

The main target of the proposed scheme and the research approach of this paper, is to

exploit the incoming Traffic pattern in order to minimize Energy consumption of secondary

communication nodes. The proposed scheme aims to minimize the consumption of the

Energy of each secondary node in the CR system, by taking into consideration the repetition

pattern of the Traffic as well as the delay limitation (bounded delay) of each transmission. This reflective scheme considers the time-oriented continuity of the incoming traffic and the

communication traffic volume (data and control packets) among peers in order to provide the

energy conservation schedules of the communicating secondary nodes. To this end, the proposed scheme estimates the Backward Traffic Difference for extracting the time duration

for which the nodes are allowed to sleep, overcoming at the same time the network

partitioning problems and consolidating the delay limitations of the transmission in the scheme. The latter mechanism is performed through the modelled framework, taking into

consideration the overall volume of traffic that traverses a secondary node -within a specified

amount of time (duration window), and reflects this mechanism to the energy conservation

modelled scheme. The proposed scheme, in order to enable further recoverability and

availability of the requested resources, uses the promiscuous caching [10] methodology in an

opportunistic manner, in order to cache the packets destined for the node with turned-off

interfaces (sleep state) onto intermediate nodes. The proposed framework and the utilised

routing methodology enable, through the Backward Traffic Difference estimation, the next

Sleep-time duration of the recipient node to be adjusted according to the activity duration and the volume of the traffic in collaboration with the consolidated routing mechanism.

3. ENERGY EFFICIENT ROUTING SCHEME BASED ON

BACKWARD TRAFFIC DIFFERENCE ESTIMATION The transmission of secondary communication nodes in an ad-hoc CR network is based on

radio spectrum opportunity, where routing has to take into account the availability of

spectrum in specific geographical locations at local level. Spectrum awareness, route quality

and route maintenance issues have to be investigated for different routing schemes, in order to enable for efficient data transfer across regions with heterogeneous radio spectrum

availability, even when the network connectivity is intermittent or when an end to end path is

5

temporarily unavailable. Figure 1 illustrates a simulation scenario, where primary nodes

operate over specific channels in three geographical areas (i.e. Area A, B and C in Figure 1).

Secondary nodes (i.e. 43 nodes were defined in simulation scenario) opportunistically

operate, by utilizing remaining available channels in each geographical area (i.e. TVWS in

Figure 1). It has to be noted here that a CCC does not exist between secondary nodes, which are located in neighboring geographical areas (i.e. Area A, B and C in Figure 1). In this case,

secondary communication nodes, which are positioned in locations with higher TVWS

availability (e.g. locations outside areas A, B and C) operate as intermediate relay nodes, switching between alternative channels. Therefore, such relay nodes enable for ad-hoc

connections among secondary nodes, located inside areas A, B and C.

Figure 1 Secondary communication nodes operating over heterogeneous TVWS

Taking into account this simulation scenario, spectrum awareness has to be investigated,

regarding routing in such an ad-hoc CR network, where secondary nodes are prohibited to

operate on spectrum bands occupied by primary nodes. The main target of routing in this CR

networking environment is to provide optimal, high throughput data transfer by efficiently

selecting the best routing paths among secondary nodes. In this framework, a novel routing

protocol has to be adopted, in order to enable routing path discovery capabilities, considering

TVWS heterogeneity of different geographical areas. Route quality issues have also to be investigated, since the actual topology of such multi-hop CR networks is highly influenced by

primary nodes behaviour and classical ways of measuring/assessing the quality of end-to-end

routes (nominal bandwidth, throughput, delay, energy efficiency and fairness) should be

coupled with novel measures on path stability. In a general context, routing in an ad-hoc CR

network over TVWS constitutes a rather important but yet unexplored problem, especially

when a multi-hop network architecture is considered. Therefore, a novel routing protocol is

vital to be designed and developed, in order to overcome the above mentioned challenges,

towards establishing and maintaining optimum routing paths, among communication nodes

with different radio spectrum availability.

3.1 PROPOSED UNDERLYING ROUTING MECHANISM

Secondary nodes located outside geographical areas A, B and C in the above mentioned scenario, are able to operate over all available channels (i.e. c.40-c.60) and act as intermediate

nodes, connected with a Geo-location database that includes information regarding TVWS

availability. They are also enhanced with routing mechanisms capabilities, enabling to determine routing paths between secondary nodes with different radio spectrum availability in

such areas. Towards enabling for an optimum data transfer, among secondary communication

nodes, a novel routing protocol was designed, developed and evaluated, by conducting experimental simulations. This routing protocol is based on the exchange of AODV-style

messages [26] between secondary nodes, including two major steps in the route discovery

process (i.e. route discovery and route reply step). This selection was made due to the

unpredictable availability of the TVWS that requires hop-by-hop routing, by broadcasting

6

discovery packets only when necessary. During the route discovery step, a RREQ (route

request) message, including TVWS availability of nodes is sent by the source node to acquire a

possible route up to the destination node. Once the destination node receives the RREQ

message, it is fully aware about the spectrum availability along the route from the source node.

The destination node then chooses the optimum routing path, according to a number of performance metrics (e.g. backoff delay, switching delay, queuing delay, number of hops,

throughput) and assigns a channel to each secondary user along the route. It has to be noted

here, that the evaluation of performance metrics is conducted, by each intermediate node during the routing path of the RREQ message. More specifically, the evaluation of delay

metrics is represented as Eni (see Table 1), where E is the end-to-end delay in millisecond,

while ni represents the ith intermediate node that serves the flow. Also, En is defined as the

delay occurred during the RREQ message. In the next step of the proposed process, destination

node sends back a RREP (route reply) message to the source node that includes information

regarding channel assignment.

Figure 2 presents the detailed signalling mechanism of the proposed routing protocol for

handling both RREQ and RREP messages. A source node initiates a flow (i.e. New Flow in

Figure 2), transmitting a RREQ message to an intermediate node located in a neighbouring

location. This intermediate node determines if it is possible to accommodate the incoming flow based on information stemming from the Geo-location database. In case that it is possible to

accommodate it, performance metrics are evaluated and RREQ message is forwarded to the

next hop. When destination node receives this message, it is informed regarding TVWS

availability along the routing path from the source node. Destination node replies by sending a

RREP message to the source node that includes relevant information, concerning channel

allocation. Such data/information is mainly exploited to enable secondary nodes setting their

channel of operation along the routing path. When source node receives RREP message,

routing path has been established and useful data transmission is initiated.

Figure 2 Message exchange process of the proposed routing protocol

7

In the case when the intermediate node is not capable to accommodate the incoming flow

(i.e. new flow in Figure 2), a redirection mechanism (redirection in Figure 2) is in charge of

informing the source node, about the neighbouring node, which could possibly act as an

alternative intermediate node. The proposed routing protocol determines a route, only when a

source node wishes to initiate data flows to a destination node. Routes are maintained as long as they are needed by the source node and the exploitation of sequence numbers in the

exchange messages guarantee a loop-free routing process. Furthermore, the proposed routing

protocol is a reactive one, creating and maintaining routes only if it is necessary, on a demand basis. Routes are maintained in routing tables, where each entry contains information,

regarding destination node, next hop, number of hops, destination sequence number, active

neighbouring nodes for this route and expiration time of the flow. The number of RREQ

messages that a source node can send per second is limited, while each RREQ message

carries a time to live (TTL) value that specifies the number of times this message should be

re-broadcasted. This value is set to a predefined value at the first transmission and increased

during retransmissions, which occur if no replies are received.

Towards further optimizing the proposed routing protocol, an assigning mechanism was

designed and adopted to alleviate service load of intermediate nodes. This process is adapted

to each intermediate node, which is further able to determine if a neighbour node performs better during the process of routing paths establishment. More specifically, when a source

node initializes a new flow, by sending a RREQ, the intermediate node is informed regarding

the status of neighbouring nodes from the geo-location database through the CCC. Then, the

intermediate node evaluates the new flow (i.e. evaluation of performance metrics “Eni”) and

encapsulates the evaluation results in a message that it is forwarded to all neighbouring nodes.

Once neighbouring nodes receive a redirecting request, they check its validity with the

corresponding flow, ensuring that they are not the source/destination nodes or next-hop nodes

of that flow. Then the neighbouring nodes initiate a process, in order to evaluate the flow and

they send to the intermediate node the result of the evaluation through a redirecting reply message. Once the intermediate node receives the redirecting reply from several of its

neighbouring nodes it then selects the optimum one, in order to serve/accommodate the

incoming flow. The basic steps of the proposed message exchange process can be summarized in the pseudocode of Table 1.

Initiate New Flow “f” with evaluation En Update Intermediate Node “n” with neighbour status

k = number of intermediate nodes

//Decision of node “n”

for (i=1; i++; i=k){

if n = sending node ||next-hop node || destination

node

then discard message

else

flow evaluation Eni if Eni > En then flow accommodation

//Flow redirection

else do

generate and broadcast redirection information

message

flow evaluation Eni

flow accommodation

until(receive route acceptance)

generate and send RREP to source node

}

Table 1. Pseudocode of the basic steps of the proposed message exchange process

For enabling Energy-Efficiency in the proposed framework a Backward Traffic Difference

(BTD) estimation methodology is used. The main additional contribution is that, in the

8

proposed framework, the BTD estimation is bounded by the delay limitations of the

transmission, whereas it takes into consideration the hop-by-hop link delay as well as the total

end-to-end delay of the transmission. The later should satisfy the delay requirements of the

transmission. The designed model guarantees the end-to-end availability of requested

resources while it reduces significantly the Energy Consumption and maintains the requested scheduled transfers, in a mobility-enabled communication. The innovation adopted in this

scheme is that each secondary mobile node uses different assignment(s) of sleep-wake

schedules based on the incoming traffic difference that each node receives through time. The Sleep-time duration is assigned according to the BTD scheme in a dissimilar manner in order

to enhance node’s lifetime, whereas it avoids mutation which, will result in network

partitioning and resource sharing losses.

Assuming that a mobile secondary node has already used the depicted routing scheme of

the previous section and established an end-to-end connection in order to transmit requested

content/packets. Routing occurs on the end-to-end basis and each node separately runs the

Traffic-aware mechanism using the BTD as is described in the following section. The

mechanism measures the traffic that traverses each one of the nodes where, the BTD

estimation through the assigned time-window frame will affect the Sleep-time duration and

enable Energy conservation onto nodes as conducted simulation experiments show.

3.2 TRAFFIC-AWARE SCHEME FOR ENERGY-EFFICIENT TRANSMISSION

3.2.1 Traffic-driven Middleware and supported mechanisms

Efficient mobile sharing process is complex because its components change in time and space in terms of connectivity, portability, accessibility/availability and mobility. Towards

reducing the impact of these changes, the resource sharing application must have a context-

aware adaptive behavior. Context-aware through traffic-aware adaptation is a fundamental

concept for pervasive and ubiquitous environments. In collaboration with the proposed

routing methodology used, this paper elaborates on the traffic volume exploitation and

manipulation, and its direct impact on the EC mechanism. Traffic-aware policy requires an

active scheme to be applied, through which, the traffic will reflect a certain impact on the

nodes taking into account the EC trade-offs. Wireless devices should consider the incoming

traffic, in order to adapt and reflect a certain feedback according to the traffic volume to the energy conservation mechanism. A middleware, which hosts traffic changes and has a direct

impact through the estimated scheme presented in the next section using a collaborative

traffic-aware scheme, is shown in Figure 3. Figure 3 depicts a cross layer interaction through a mechanism for traffic-awareness in an end-to-end manner. In particular, real-time media

traffic, such as voice and video typically have high data rate requirements and stringent delay

constraints, whereas wireless nodes generally have limited or momentarily connectivity. The

proposed middleware enables data packets to be traversed and manipulated through the

utilized Wireless Data Link, Network, and Transport layers, by considering the traffic

awareness mechanism and the model for volume estimation to be reflected on these layers.

The proposed traffic-aware scheme and the associated mechanisms evaluate (after the bootstrap process of the system) the estimated (quantified as Volume/Capacity) traffic that is

destined for each node. In Figure 3 the i

kV denotes the volume of traffic destined for node k

and stored onto node i using the promiscuous caching policy [10]. In this way, it enables –

through the proposed mechanism- estimation for the next sleep duration of the node-as

presented in the next section. This traffic-aware policy and the sleep duration evaluation are

performed in an interactive way through the Backward Traffic Difference (BTD) using a

certain window frame-size. These mechanisms are performed, in order to tune the wireless

interface of each device to sleep/wake, according to the activity of each individual device in

the resource exchanging path.

9

Packet classification methodology was utilized as in [15], in order to mark the packets that

are exchanged whether they are delay sensitive or not. In turn, if packets are considered as

delay sensitive, strict deadlines are applied by the sender, according to the specifications set

in the network. In the case where packet deadlines cannot be satisfied, then cached packets of

nearby nodes, enable recovery using the promiscuous caching [18]. This mechanism enables the resources’ replication and increases the resource sharing reliability [18]. The quantitative

mechanisms shown in Figure 3, are depicted in the following sections with the quantitative

analysis.

Figure 3. The traffic-aware interactive middleware mechanism with the associated influenced layers in

the communication stack

The proposed scheme introduces high availability capabilities for resource sharing

allowing for continuous operation and smoother handling of system outages. The

promiscuous caching mechanism estimates the volume of traffic that is cached on an

intermediate (active state) node in the path, in order to measure the volume of traffic that is

outage. The traffic-aware middleware that hosts the resource intensive scheme allows a more

flexible system infrastructure that can adapt to dynamic changes in resource sharing application requirements and connectivity conditions. As the reflective middleware model is a

principled and efficient way of dealing with highly dynamic environments, the proposed

scheme yet supports a reflective and flexible adaptation of the traffic volume i

kV . The traffic

is considered in terms of the repetition pattern by estimating the Backward Difference for

extracting the time duration for which the node is allowed to reduce the Energy consumption

by entering the Sleep state during the next time slot T. The middleware in collaboration with

the proposed routing scheme enables secondary nodes to exchange efficiently the requested

resources by evaluating within a time frame window the incoming traffic volume as well as

the incoming traffic that is destined for these nodes. In order to enable recoverability of the

incoming traffic, if a node is in the Sleep state of in no connectivity range, then the traffic is

cached using the promiscuous caching concept applied onto intermediate nodes in the path.

The traffic-aware resource sharing scheme expands a cross-layer interaction (see Figure 3) for Level 2 Medium Access Level (L2/wireless MAC sleep/active time manipulation) and L3

using the proposed routing methodology. In the proposed middleware there are no strict

associations among the tasks and the layers. The traffic-aware middleware enables the data packets to be traversed and manipulated through the utilized Data Link, Network, and

Transport layers by considering the traffic awareness mechanism and the model for volume

estimation to be reflected on these layers. The proposed traffic-aware mechanism evaluates

(after the bootstrap process of the system) the estimated (quantified as Volume/Capacity)

traffic that is destined for each node. In this way it enables –through the proposed

mechanism- estimation for the next slot Sleep duration of the node as presented in the next

section.

i

kV

10

The power control is provided by determining the transmit periods and the associated

power level such that the energy consumed is steadily reduced. To this end, by using the

Backward traffic-aware mechanism presented in the next section, the scheme aims to

guarantee the resource sharing stability, whereas at the same time to offer energy

conservation. Since nodes in wireless networks typically rely on their battery energy, the proposed framework encompassed in a traffic-aware middleware, utilizes a reflective

mechanism which hosts a traffic-aware scheme for conserving energy in CR wireless

environments. The scheme evaluates the scheduled activity periods of each node, in order to measure and estimate a ‘safe’ forecast time duration for the scheduled time that each node can

safely sleep in order to conserve energy.

3.2.2 Opportunistic resource sharing using Backward Difference Traffic Estimation for Energy

Conservation

When a source needs to send requested packets or stream of packets (file) to a destination

where the destination node(s) may have moved or is/are set in the Sleep-state, then the

requested information will be missed and lost. This implies that, in a non-static multi-hop

environment, there is a need to model the activity slots that a node experiences in contrast to the requested resources in the end-to-end path such that the resources can be efficiently

shared among users, whereas any redundant transmissions and lost packets/streams are

avoided. The proposed scheme takes into consideration the incoming nodal traffic of the

secondary nodes in the CR system, and estimates the Sleep-time duration of the node

according to the Backward Traffic Difference (BTD) using a certain window frame-size.

Traffic that is being traversed in a path is being forwarded on a hop-by-hop basis from one

secondary node to another node to another until the requested traffic reaches the destination

node. On one hand, if node is available and in Active-state, it receives the transfer (for

example file) whereas, if the node receiving the file is not the destination, it forwards the packet to the destination node via other neighbouring hop-nodes in the path1. On the other

hand, if the next hop-node is not available to receive and process within a specified time-

frame the transmission to the next-hop node, then promiscuous caching [10] of the transmitted packet occurs in the path. This is performed in order to buffer the packets that are

intended for the destination node. Therefore the proposed traffic-based framework is focusing

on the Traffic that is incoming for each node and for a specific time-window T. Data packets will be transferred from a source node to a destination node, according to the proposed

routing procedure described above. The spatial characteristics and the associated modelled

traffic monofractality properties [10] were taken into consideration for modeling the energy

schedules for a certain time-frame in order to enable Energy Conservation. The traffic and the

monofractal characteristics of it were considered in order to enable greater associativity with

the self-similar behaviour expressed in [10, 19], where the window of the traffic duration is tw.

In this work we consider the window to be

}2)()(:),(lim{),( )( <∀−≈∈=⋅→

kktRtRdstFdst NNtntkt

w τ (1)

where ),( dstwis the time window measure for the multipath pair source-destination model

and where the limit of it should be bounded into kτ time duration for the determined window

size. k should be less than 2 in order to satisfy the monofractality property of the repetition

index ( )NR t of the incoming traffic [10].

When nodes in the transmission path are expecting traffic they keep their communication

network interfaces in Active-state for time t. This means that if the transmission will delay

with t active

d t> where active

t is the active time duration of the wireless nodal interface, then

1 The path is constructed according to the routing scheme explored in the previous section.

11

nodes may set their interfaces to an Energy conservation state (Sleep-state). In this respect,

the scheme enables the promiscuous caching [10] to be enabled. The packets that are destined

for the certain node can be cached for a specified amount of time (as long as the Node (i) is in

the Sleep-state) in the 1-hop neighbour node (Node(i-1) that is in Active state) in order to be

recoverable when the referred node enters the Active state. When certain node has incoming traffic then the node remains active for prolonged time. As a showcase this work takes the

specifications of the WiMax IEEE 802.16e (specifications v.2005) [27] that are

recommending the duration of the forwarding mechanism that takes place in a non-power saving mode lays in the interval 0.1 nsec < τ <1 psec. This means that approx. 80 times in a

msec the communication’s triggering action between nodes may result a problematic end-to-

end transmission reliability/accuracy. Adaptive Dynamic Caching [8] takes place and enables

the packets to be “cached” in the 1-hop neighbouring nodes. Correspondingly, if node is no-

longer available due to sleep-state in order to conserve energy (in the interval slot T=0.08

μsec), then the packets are cached into an intermediate node with adequate capacity equals to:

( ) ( ), ( ) , ,f ft k s t iC t C t> where

ft iC Cα> ⋅ ; where iα is the capacity adaptation degree [10] based on

the time duration of the capacity that is reserved on node N ofkC ; where ( ), ( )ft k sC t is the

needed capacity where i is the destination node and k is the buffering node (a hop before the

destination via different paths).

This work associates also the Backward Difference Traffic moments with the Sleep-time

duration in order to tune the Active durations of a node according to the transmissions’ activities and the expected traffic for the next time step. This is performed via the BTD

estimation which enables the capacity of the traffic ( )C t that is destined for the Node i in the

time slot (duration) t, and the traffic capacity ( )iN tC which is cached onto Node (i-1) for time

t, to directly affect the Sleep-time of a node. The one-level Backward Difference of the

Traffic is evaluated by estimating the difference of the traffic while the Node(i) is set in the

Sleep-state for a period, as follows:

(1) 2 1

(2) 3 2

( 1) 2

( ) ( 1)

( 1) ( 2)

( ( 1)) ( ( 2))

i

i

i

N

N

N n n

C T T

C T T

C T n T n

τ τ

τ τ

τ τ+

∇ = − −

∇ = − − −

∇ = − − − − −

M

(2)

where (1)iNC∇ denotes the first moment traffic/capacity difference that is destined for

Node(i) and it is cached onto Node (i-1) for time τ , 2 1( ) ( 1)T Tτ τ− − is the estimated traffic

difference while packets are being cached onto (i-1) hop for recoverability as in [19]. The

Traffic Difference is estimated so that the next Sleep-time duration can be directly affected

according to the following:

1 1( ( )) , , { 1, }total totalC T C C C C Tδ τ τ= − ∀ > ∈ − (3)

where the Traffic that is destined for Node(i), urges the Node to remain active for ( ( ))

0prev

total

C TT

C

δ⋅ > , prevT is the previous Sleep-time duration ({ 1, }τ τ− ) of the node. On the

contrary with [1, 4] this work measures the BTD within a certain transmission time-frame.

This means that each transmission is bounded by a certain delay limitation (time-duration

( , )wt s d ) which cannot be overtaken. When a node receives traffic, the traffic flow ft , can be

modelled as a stochastic process [19, 20] and denoted in a cumulative arrival form as

{ ( )}f ft t T N

A A T ∈= , where ( )ft

A T represents the cumulative amount of traffic arrivals in the

12

time space [0..T]. Then, the ( , ) ( ) ( )f f ft t t

A s T A T A s= − , denotes the amount of traffic

arriving in time interval (s, t]. Hence the next Sleep-time duration for Node (i) can be

evaluated as a function of the Traffic that traverses the Node (i) provided that the amount of

traffic arriving in time interval (s, t] is measured according to the total aggregated

Traffic/Capacity that the channel can handle at time t. The next Sleep-time duration for Node

(i) can be defined as:

iL (n +1) =δ(C (T ) | A

t f

(s ,T ))

Ctotal

⋅Tprev

,∀δ(C (T )) > 0,

max

( , ) 2ij

wt s dδ

<∆

(4)

where ijδ is the delay that the transmission experiences to reach destination j,

max∆ is the

max allowed delay-duration that the transmission cannot overtake. The aggregated traffic

destined for Node (i) should satisfy the 1

sup ( , ) ( )f f

f

N

t ts T t

A s T C T≤ =

− ∑ , for traffic flow

ft at time

T, and ( )ft

C T represents the service capacity of the Node(i-1) for this time duration. The

delay that the transmission experiences ijδ should satisfy the

ij pdδ < , where pd is the

maximum delay in the end-to-end path from a source to a destination and can be is evaluated

as:

1

0

i

p i i

i

d Tδ−

=

= +∑ (5)

where iδ is the duration where the requested data was hosted onto i-node, and T is the

transmission delay. Then for obtaining the minimised energy consumed in the path tCE the

following should be satisfied:

{ }arg min ( ) : ( ) min ( )ij p f f td t ij p t CC T d C T f Eδ δ< = < = (6)

Taking into consideration the above stochastic estimations, the Energy Efficiency ft

EE

can be defined as a measure of the capacity of the Node(i) over the Total Power consumed by

the Node, as: ( )

( )f

f

t

t ij p

C TEE T d

TotalPowerδ= ∀ < (7)

Equation 7 above can be defined as the primary metric for the lifespan extensibility of the wireless node in the system.

The basic steps of the proposed scheme can be summarized in the pseudocode of Table 2.

In Table 2 the Algorithm starts by examining the existing capacity of each node and whether

the node using the caching capacity parameter can host a delay-intensive traffic within a

bounded delay requirement (ij pdδ < ). Then, in line 2, the scheme examines whether each one

of the nodes has cached capacity (this is the traffic that is destined for a node which lays in

the sleep-state). It then evaluates the traffic difference and the difference in the traffic volume,

and measures the activity of the node according to the incoming traffic that was buffered in

the 1-hop neighbouring node. The most importance evaluation in the pseudocode comes

through the estimation of the next Sleep-time duration for Node (i) which is then evaluated as

a function of the Traffic that traverses the Node (i). This estimation is performed (line 7) and it is subject to the amount of traffic arriving in time interval (s, t], and is measured according

13

to the total aggregated Traffic/Capacity that the channel can handle at time t. The next Sleep-

time duration for Node (i) can be defined as in equation 4 above.

1: while ((Node(i) that there is ( )C t >0)&& (

arg min ( )ij p fd tC Tδ <

==True)) {

2: while (( )iN tC >0) { //cached Traffic measurement

3: Evaluate ((1)iNC∇ );

4: Calc(1 1

( ( )) , , { 1, }total total

C T C C C C Tδ τ τ= − ∀ > ∈ − )

5: if (Activity_Period= ( ( ))prev

total

C TT

C

δ⋅ >0)

6: //Measure Sleep-time duration

7: ( ( ) | ( , ))

( 1) , ( ( )) 0,ft

previ

total

C T A s Tn T C T

CLδ

δ+ = ⋅ ∀ >

8:

max

( , ) 2ij

wt s dδ

<∆

9: Sleep ( ( 1)i

nL + ); //sleep duration for the upcoming

slot 10: } //while 11: }//while

Table 2. Pseudocode of the basic steps of the proposed traffic-aware mechanism for energy conservation

4. PERFORMANCE EVALUATION ANALYSIS, EXPERIMENTAL

RESULTS AND DISCUSSION Several experimental tests were conducted, in order to validate the efficiency of the

proposed routing protocol and the resource intensive traffic-aware scheme. Performance

evaluation results were extracted, by conducting exhaustive simulation runs and

experimentation using the NS-2 [30] and the generated real traffic traces for implementing the proposed scenario. The energy consumption model used in the simulation for the calculation

of the amount of energy consumed is based theoretically on the specifications of the WiMax

IEEE 802.16e (ver. 2005) [27]. The extracted results are characterizing the trade-off issues

between the performance in deploying the discussed scenario and the Energy consumption of

each secondary CR node by using the proposed traffic-oriented scheme. Results also

encompass comparisons with other existing schemes for the throughput, the reliability and the

accuracy offered by the proposed framework as well as EC efficiency conveying an estimated

confidence interval (CI) of approximately 3%<CI<5%. All confidence intervals were found to

be less than 5% of the mean values of the certain examined parameters. The mobility model adopted in this work is based on the probabilistic mobility scenario derived by Fractional

Random Walk. The probabilistic random walk mobility model was derived from the

Brownian motion [31], where nodes are moving according to certain probabilities with respect to the location and the time.

According to such simulation scenario, a number of data flows are contending to pass

through the same intermediate node, thus evaluation of delay metrics is crucial, for an efficient performance of the proposed routing protocol. In this context, a number of delay

metrics [18], [28], are evaluated, such as end to end delay, backoff delay, switching delay and

queuing delay. End to end delay from the source node up to the destination node is computed

as the overall sum of queuing delay and node delay:

14

D���������� = �� ���� + ��� (6)

Node delay at an intermediate node i is based on switching delay and backoff delay and is

computed as follows:

D���� = ∑ (D��������� + D����� )�"

(7)

Figure represents simulation results related with the performance comparison of mean

end-to-end delay, while the number of active flows is increasing for both version of the

proposed routing protocol. It is clear that when routing protocol incorporates the assigning mechanism and the number of active flows in the network is small, there is no important

advantage, in terms of mean end-to-end delay. However, when the number of active flows is

more than three, intermediate nodes begin to suffer the accumulating queue, and flow

redirection becomes necessary. Such results also show that the mean end-to-end delay is less,

in the case of the enhanced routing protocol, in comparison to the basic version of it without

incorporating the assigning mechanism.

Figure 4: Mean end-to-end delay for different number of simultaneous flows

Figure 5 depicts simulation results of end-to-end Delay for one single flow, when the

probability of primary user presence increases, while Figure 6 presents the same metric, but in

this case each point represents the average of end-to-end Delay for ten simultaneous flows for a certain value of primary user presence probability. From both figures it is clear that when

the probability of primary user presence is getting higher, delay is increasing, while in the

case of the basic routing protocol, delay increase is more significant in comparison to the

enhanced routing protocol incorporating the assigning mechanism. This result is reasonable,

since the probability of the presence of an incumbent system is detected as a route failure,

introducing in this way additional delay.

15

Figure 5: End-to-End Delay for the 1st flow versus probability of PU presence

Figure 6: Average end-to-end Delay for ten simultaneous flows versus probability of PU presence

Figure 7 presents the average end-to-end delay that occurred among the source and

destination nodes as the distance between them is increased. From this figure it is clear that

the distance affects the delay among nodes. This result is reasonable since the longer is the

routing path, the more numerous are the primary nodes that affect the path, and the more

significant are the effects of the route range/diversity. It is further observed that the initial

version of the proposed routing protocol adds higher delays, as the distance is increasing

rather than those occurred when assigning mechanism is introduced, resulting the most optimal routing paths between the source and destination nodes. Consequently, the longer is

16

the path, the more significant are the effects of the route diversity. Finally, Figure 8 depicts

the comparison among both versions of the proposed routing protocol, under the number of

hops that are required, in order to make feasible all routing paths between source and

destination nodes, for each flow set according to the simulation scenario. This comparison

results that routing protocol, incorporating the assigning mechanism performs better, since it makes the decision for routing path establishment at every hop.

Figure 7: Average end-to-end Delay versus node distance

Figure 8: Number of Hops per flow

17

Furthermore, Figure 9 illustrates the lifespan of secondary nodes in the transmission path

in contrast to the number of hops. The proposed scheme is compared with existing similar

schemes, showing significant increment in the lifespan extensibility, particularly when the

number of hops increases. The comparative evaluation illustrates that the proposed routing

protocol with the assigned mechanism behaves gradually better, and increases the lifespan of each secondary node.

Figure 9: Lifespan of secondary nodes with the number of hops in the transmission path

Figure 10: Fraction of remaining energy comparisons for different EC schemes

The proposed scheme is also compared with other existing schemes in terms of the remaining energy dissipation of each secondary node. Figure 10 shows the fraction of the

18

remaining energy compared with different EC schemes. In the case of periodic sleep and

wake methodology, the fraction of the remaining energy is dramatically dropped whereas,

using the traffic-aware sleep-scheduling in contrast to the limitations of the delay bounds, the

scheme offers gradual consolidation of the reduction of the remaining energy of the nodes.

Figure 11 shows the Successful packet Delivery Ratio (SDR) with respect to the end-to-end streaming delay and the delay duration/total delay for k-hops in the communicating path, for

mobile secondary nodes. By comparing the proposed scheme with the scheme in [29], the

BTD scheme offers greater SDR in the end-to-end path. This outcome is evident because of the adaptivity in the traffic volume and the recoverability mechanism hosted by the proposed

scheme, which enables the packet delivery ratio to be kept at high percentage values.

Comparative results regarding the offered Throughput as well as the end-to-end latency for

different fading and mobility models are shown in Figure 12 respectively. The signal strength

and the associated fading characteristics are posing a major factor for the end-to-end reliable

transmission, whereas as long as the path might be, the greater the SDR dissipation. Figure 13

illustrates that the proposed methodology shows robustness in the presence of fading

Rayleigh model used. The fading characteristics of the channels affect vertically the

transmission rate of the channels, whereas the Throughput and the SDR both can be

negatively affected. In our conducted experiments the Rayleigh fading small scale fading of 2% with mean noise factor of 4dB causes minor Throughput degradation ranging from 7-32%

-in worst case affection. The Throughput remained at significantly high levels due to the

recoverability promiscuous caching methodology utilized. Additionally, the proposed scheme

was evaluated for the end-to-end latency experienced by the secondary users since their initial

request, for two mobility models, namely Uniform Random Mobility Pattern and Random

Waypoint mobility. The latter, Random Waypoint model, uses for the movement of mobile

nodes a certain likelihood based on their location, velocity and acceleration change over time,

whereas, Uniform Mobility uses only uniform distribution model to denote the next

mobility/movement. Figure 14 illustrates the measures of the end-to-end latency in msec with the number of mobile users during simulation. It can be easily spotted that when the number

of users increases, the end-to-end latency decreases vertically, whereas the overall SDR

remains at increasingly high levels.

Figure 11: The Successful packet Delivery Ratio (SDR) with the end-to-end streaming delay in the

transmission path (msec)

19

Figure 12: The Successful packet Delivery Ratio (SDR) with the total delay in the k-hops path (msec)

Figure 13: The effects of channel’s fading and no-fading with the average throughput

20

Figure 14: End-to-end latency in ms with the number of mobile users during simulation

Figure 15, shows the Complementary Cumulative Distribution Function (CCDF) for the

resource sharing reliability with the number of secondary users. The Complementary Cumulative Distribution Function (CCDF or –as called- the tail distribution) shows the

distribution for reliable transmission with the number of secondary nodes/terminals and the

variation by hosting fading characteristics during transmission. Figure 15 illustrates that the CCDF remains at high levels when the number of secondary users is increasing. This occurs

due to the cooperative promiscuous caching, for which, when a missed packet cannot arrive at

the destination node, it can then be can be recovered. It is evident that the CCDF can be

significantly increased when the number of secondary users increases resulting to the

awareness of an established best effort routing path, which enables the secondary users to

receive and transmit with reliability the requested data. The SDR with the total number of

nodes and with the node speed (m/s) is shown in Figures 16 and 17, respectively. Figure 17

highlights the benefits of using the BTD with delay limitation in the transmission for two

traffic-aware schemes. It is clearly illustrated that the proposed scheme offers an increment in the SDR and overall higher packet delivery rate. Figure 18 illustrates the Packet Drop Ratio

during simulation experimentation in a log-scale magnitude for two different mobility

frameworks and for nodes without mobility. The Packet Drop Ratio is kept in low levels considering that there is mobility which aggravates the success of packet delivery. It is worth

mentioning that the Packet Drop Ratio in the presence of generic probabilistic random walk

mobility model derived from the Brownian motion [31] as well as in the presence of mobility with distance broadcasting of secondary nodes, is not increased showing significant

robustness in the transmission reliability offered by the proposed scheme. Furthermore,

Figure 19 shows the system’s throughput during simulation which shows to be significantly

increased when compared with the scheme in [8]. This is caused by the potential effects of the

intercluster sleep-proxy scheme introduced in [8] which is only based on incoming nodal

traffic where the traffic volume is only considered in the measured cluster rather than in the

end-to-end exchanging path. In turn Figure 20 illustrates the average delay in dp bounded

transmission with the average number of secondary hop-nodes/users. The proposed scheme

shows to slightly outperform the scheme in [8] taking into account the variability of delays that are minimized when more secondary nodes are participating in the transmission process,

21

as depicted by Figure 14, earlier. Moreover in Figure 21 there is a comparative illustration for the SDRs with the transmission rates for three different power ratios for CR secondary nodes

in decibels (dBmW). Figure 21 highlights the benefits of using the proposed intensive traffic-

aware scheme, incorporated into an energy-efficient routing protocol, in contrast to the three

different power ratios for CR secondary nodes. Notwithstanding the proposed framework

enables high transmission rates, when the 33dBm is reached in the communication the SDR

drops slightly decreasing the transmission rates over 12Mbps. Figure 22 presents the

comparative results for the system’s overall EC (μW) with the number of participating

secondary nodes, illustrating the supremacy of the proposed scheme minimising the system’s

overall EC by an average of 33% when compared with [29], and by an average of 17% when compared with the scheme in [8]. The accuracy offered by the proposed framework for

obtaining the system’s overall EC is measured with a confidence interval of approximately

3%<CI<5%. Notwithstanding all confidence intervals were found to be less than 5% of their mean values, the system’s overall EC offered by the scheme in [8] showed to be slightly

above the proposed improved framework hosting the underlying efficient routing scheme, and

it clearly outperforms the scheme proposed in [29].

In Figure 23 the average energy consumed with the power ratio in decibels (dB) of the

measured power referenced to one mW in contrast to the simulation time, is presented. The

experimental evaluations were extracted in the presence of fading and no-fading

communicating obstacles and characteristics as indicated in the introduction. The Energy

Efficiency (bytes/mW), which is defined as the service capacity/total energy consumed as in

Eq. 7, in contrast to the delay requests is presented in Figure 24. Taking into consideration the

estimations of the previous section, the Energy Efficiency ftEE is defined as in Equation 7 as

a measure of the capacity of the Node(i) over the Total Power consumed by the Node. Results

obtained in Figure 24 show that the network lifetime can be significantly prolonged when the

Traffic-aware scheme is applied. By comparing the results obtained through simulation experiments for the scheme developed in [10] which takes into consideration the regional

capacity and the remaining capacity of each mobile node as well as the comparison with the

periodic Sleep/Wake scheduling, the proposed scheme offers greater Energy-Efficiency, while it minimizes the delay per request.

Figure 15: CCDF Sharing Reliability with the Number of sharing secondary-users in the CR system

22

Figure 16: The SDR with the total number of nodes

Figure 17 Comparative SDR with the node speed (m/s)

23

Figure 18: Packet Drop Ratio during simulation experimentation (log-scale)

Figure 19: System throughput during simulation

24

Figure 20: Average delay in dp bounded transmission

Figure 21: SDR with the transmission rates for three different power ratios for CR secondary nodes in

decibels (dBmW)

25

Figure 22: Comparative results for the system overall EC (μW) with the number of participating

secondary nodes

Figure 23: Average energy consumed (J/bit) with the simulation time and the power ratio in decibels

(dB) of the measured power referenced to one mW

26

Figure 24: Energy efficiency with respect to the delay per request (transmission) in an end-to-end

manner

5. CONCLUSIONS AND FURTHER RESEARCH This work proposes an efficient routing mechanism where, in collaboration with the

underlying BTD scheme, it enables energy conservation and reliable data flow among

secondary communication nodes with heterogeneous spectrum availability in CR systems.

The proposed routing scheme establishes an end-to-end optimal path whereas, secondary nodes in CR systems can efficiently and, in a collaborative manner, share requested

data/resources. The hosted traffic-aware scheme enables the self-tunability of the sleep

schedule of each node to be applied through the BTD, measured within a certain transmission

time-frame. Within the proposed framework, the bounded end-to-end delay of the

transmission is taken into consideration for each secondary node, aiming to impact the EC

through the modelled traffic-aware mechanism. The performance evaluation through

simulation shows that the proposed routing scheme in collaboration with the BTD

mechanism, manipulates the energy consumption of each secondary node/device effectively,

and outperforms in contrast to similar traffic-aware schemes. Moreover, the traffic-aware management scheme can significantly reduce the energy consumed and can keep the

throughput response of the system at relatively high-levels. The comparative measurements

with other similar schemes show that the proposed methodology can efficiently conserve the energy, by offering at the same time significantly high SDRs, and can significantly extend the

lifetime of each secondary node in the CR network. Furthermore, efficient routing protocol

operation, as a matter of maximum-possible routing paths establishments and minimum

delays was validated, by adopting the proposed message exchange mechanism that was

developed based on the simulation scenario defined above. Towards evaluating the

performance of the routing protocol in this respect, a large number set of experimental tests

was conducted under controlled simulation conditions, where various secondary systems were

concurrently/simultaneously communicating in ad-hoc connections, accessing the available

TVWS. The obtained experimental results verified the validity of the proposed routing mechanism, towards enabling for an efficient communication among secondary nodes located

in areas with different TVWS availability.

27

Further streams in our on-going research include the evaluation of our scheme using real-

time measurements and real-time verification using the existing infrastructure. Issues to be

considered are the topology formation, using social collaboration as well as geographical

profiles, in order to face potential partitioning problems. Moreover, the usage of traffic

engineering models, in order to explicitly express the behaviour of such dynamically changing scenarios. Additionally we are working towards the expression of our scheme with

the combined infinitesimal perturbation analysis and apply a stochastic algorithm into the

performance gradient of the system. Finally, several optimization methods will be adopted, towards minimizing delays occurred during the routing paths of data flows and maximizing

the number of established paths. This comprises an open-end research issue with many

research concepts for future examination.

6. ACKNOWLEDGEMENTS

This section is dedicated to host our thanks to the reviewers for their valuable comments,

which helped us to significantly improve our paper presentation and quality of our research

work.

References

[1] I. F. Akyildiz, W. Y. Lee, M. C. Vuran, S. Mohanty, “NeXt generation/dynamic spectrum access/cognitive

radio wireless networks: a survey”, Computer Networks, vol 50, no. 13, 2006, pp. 2127-2159. [2] A. Bourdena, E. Pallis, G. Kormentzas, G. Mastorakis, “A prototype cognitive radio architecture for TVWS

exploitation under the real time secondary spectrum market policy”, Physical Communication, Elsevier, 2013,

http://dx.doi.org/10.1016/j.phycom.2013.11.003

[3] A. Bourdena, E. Pallis, G. Kormentzas, C. Skianis, G. Mastorakis, “Real-Time TVWS Trading Based on a

Centralised CR Network Architecture”, in Proc. IEEE Globecom2011, IEEE International Workshop on

Recent Advances in Cognitive Communications and Networking, Texas, Houston, USA, 05-09 December, 2011, pp. 994-999.

[4] A. Bourdena, E. Pallis, G. Kormentzas, G. Mastorakis, "A radio resource management framework for TVWS

exploitation under an auction-based approach", in Proc. IEEE 8th International Conference on Network and

Service Management (IEEE CNSM2012), Las Vegas, USA, 22-26 October, 2012, pp. 204-208.

[5] A. Bourdena, E. Pallis, G. Kormentzas, H. Skianis, G. Mastorakis, "QoS provisioning and policy management

in a broker-based CR network architecture", in Proc. IEEE Globecom 2012, Anaheim, California, USA, 03-07 December, 2012.

[6] E. Hossain, D. Niyato, Z. Han, “Dynamic spectrum access and management in cognitive radio networks”,

first ed. Cambridge University Press, 2009.

[7] C. X. Mavromoustakis, "Using Backward Traffic Difference Estimation for Efficient Energy Saving

Schedules in Wireless Device". IEEE CommSoft E-Letters, Vol. 1, No. 1, May 2012, pp. 1-6.

[8] M. C. Charalambous, C. X. Mavromoustakis, M. B. Yassein, "A Resource Intensive Traffic-Aware Scheme

for Cluster-based Energy Conservation in Wireless Devices", to appear in the Proceedings IEEE 14th

International Conference on High Performance Computing and Communications (HPCC-2012) of the Third

International Workshop on Wireless Networks and Multimedia (WNM-2012), to be held in conjunction, 25-

27 June 2012, Liverpool, UK, pp. 879-884.

[9] H. Shpungin, "Energy Efficient Online Routing in Wireless Ad Hoc Networks", The 8th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (IEEE

SECON'11), 2011, Salt Lake City, UT, USA.

[10] C. X. Mavromoustakis and K. G. Zerfiridis, “On the diversity properties of wireless mobility with the user-

centered temporal capacity awareness for EC in wireless devices”. Proceedings of the Sixth IEEE

International Conference on Wireless and Mobile Communications, ICWMC 2010, September 20-25, 2010-

Valencia, Spain, pp.367-372.

[11] H. Khalife, S. Ahuja, N. Malouch, and M. Krunz, “Probabilistic path selection in opportunistic cognitive

radio networks,” in Proc. IEEE Globecom, November-December 2008.

[12] H. Ma, L. Zheng, X. Ma, and Y. Luo, “Spectrum aware routing for multihop cognitive radio networks with a

single transceiver,” in Proc. Cognitive Radio Oriented Wireless Networks and Comm. (CrownCom), pp.1-6,

15-17 May 2008.

[13] H.-P. Shiang and M. van der Schaar, “Delay-sensitive resource management in multi-hop cognitive radio

networks”, in Proc. IEEE DySPAN, October 2008.

28

[14] C. Xin, L. Ma, and C-C. Shen, “A path-centric channel assignment framework for cognitive radio wireless

networks,” Mobile Networks Applications (Kluwer), vol. 13, no. 5, pp. 463-476, October 2008.

[15] K.R. Chowdhury, M. Di Felice, “Search: a routing protocol for mobile cognitive radio ad-hoc networks”,

Computer Communication Journal vol. 32, no. 18, 2009, pp. 1983–1997.

[16] R. Pal, “Efficient routing algorithms for multi-channel dynamic spectrum access networks”, in: Proc. of IEEE

DySPAN 2007, Dublin, Ireland, 2007, pp. 288–291.

[17] I. Pefkianakis, S.H.Y. Wong, S. Lu, "SAMER: spectrum aware mesh routing in cognitive radio networks", in:

Proc. of IEEE DySPAN 2008, October 2008, pp. 1–5.

[18] G. Chen, W. Liu, Y. Li, W. Cheng, "Spectrum aware on-demand routing in cognitive radio networks", in: Proc. of IEEE DySPAN 2007, Dublin, Ireland, 2007, pp. 571–574.

[19] C. X. Mavromoustakis, C. D. Dimitriou and G. Mastorakis, "Using Real-Time Backward Traffic Difference

Estimation for Energy Conservation in Wireless Devices", Proceedings of the Fourth International

Conference on Advances in P2P Systems (AP2PS 2012), September 23-28, 2012 - Barcelona, Spain, pp. 18-

23.

[20] C. X. Mavromoustakis and H. D. Karatza, “Real time performance evaluation of asynchronous time division

traffic-aware and delay-tolerant scheme in ad-hoc sensor networks”, International Journal of Communication

Systems (IJCS), Wiley, Vol. 23 No. 2, February 2010, pp. 167-186.

[21] J. Yu and A. P. Petropulu, “Study of the effect of the wireless gateway on incoming self-similar traffic," IEEE

Trans. Signal Proceesing, vol. 54, no. 10, 2006, pp. 3741-3758.

[22] X. Zhuang and S. Pande, “A scalable priority queue architecture for high speed network processing," in Proc. 25th IEEE International Conf. Computer Commun. (INFOCOM’06), 2006, pp. 1-12.

[23] Q. Cao, T. Abdelzaher, T. He, and J. Stankonic, "Towards optimal sleep scheduling in sensor networks for

rare-event detection", the 4th international symposium on Information processing in sensor networks, 2005.

[24] H. Cunqing and P. Y. Tak-Shink,"Asynchronous Random Sleeping for Sensor Networks", ACM Transactions

on Sensor Networks (TOSN), 2007.

[25] I. Jawhar, J. Wu and P. Agrawal, "Resource Scheduling in Wireless Networks Using Directional Antennas", appears in: IEEE Transactions on Parallel and Distributed Systems, Sept. 2010, Volume: 21 Issue:9, pp. 1240

- 1253.

[26] RFC: http://www.ietf.org/rfc/rfc3561.txt

[27] WiMAX Forum Specification (2011-09-20), IEEE to WiMAX Forum, IEEE to WiMAX Forum, last accessed

29th Oct. 2012.

[28] G. Cheng, W. Liu, Y. Li, W. Cheng, "Joint on-demand routing and spectrum assignment in cognitive radio networks", in: IEEE International Conference on Communications, ICC ’07, 2007, pp. 6499–6503, doi:

10.1109/ICC.2007.1075.

[29] I. Rhee, A. Warrier, J. Min, and L. Xu., "DRAND distributed randomized TDMA scheduling for wireless ad-

hoc networks", in Mobile Ad Hoc Networking and Computing, 2006, pp. 190–201.

[30] NS-2 Simulator, at http://www.isi.edu/nsnam/ns/, last accessed Oct 2012. [31] T. Camp, J. Boleng, and V. Davies, "A Survey of Mobility Models for Ad Hoc Network Research", Wireless

Communication & Mobile Computing (WCMC): Special Issue on Mobile Ad Hoc Networking: Research

Trends and Applications. vol. 2, no. 5, 2002, pp. 483–502.