Cognitive Radio AD HOC Networks Computer Science Essay

Published: Last Edited:

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

The spectrum of electromagnetic waves is the range of all possible measured radiation. Historically, the most used radio range for radio communications is the low end of the radio waves with a frequency between 0 and 30 GHz. Currently the most widely used band is between 30 MHz and 10 GHz bands which are assigned for AM/FM radio, analog and digital television, GSM/UMTS communication systems, Bluetooth and Wi-Fi, and satellites, as shown in Figure 2.1.

Figure 2.1: Wireless Spectrum Usage

The increasing request for multimedia services and internet access in mobile devices such PDA, Smartphone, Tablet PC, Laptop, has brought to a growing demand for spectrum access in certain bands.

Some central authorities who assign frequencies to the wireless devices and networks, like Federal Communications Commission (FCC) and Shared Spectrum Company (SSC) have studied the occupancy level of radio spectrum.

The analysis of the FCC in some American cities shows that over 70% of the allocated spectrum is not used continuously.

The following figure shows the results of the campaign conducted by SSC over seven American cities [37].

The Figure 2.2 shows the inefficiency of the today's fixed spectrum assignment policies

where certain portion of the frequency spectrum is heavily utilized whereas a huge amount of the spectrum remain sparsely used.

Figure 2.2: Spectrum Occupancy for different bands [37]

The limited available radio spectrum and the inefficiency in spectrum usage necessitate a new communication paradigm to exploit the existing spectrum dynamically.

One of the most efficient paradigms is the Cognitive Radio Network (CRN).

Figure 2.3: Cognitive Radio Ad-Hoc Networks Architecture [2]

The components of a CRAHN can be classified into "Primary Network" and the "CR Network" components.

The first one is referred to as an existing network, where the Primary Users (PUs) have a license to operate in a certain spectrum band. Due to their priority in spectrum access, the PUs should not be affected by Unlicensed Users.

The CR Network does not have a license to operate in a desired band and additional functionality is required for CR Users (Secondary Users) to share the licensed spectrum band. Also, CR users are mobile and can communicate with each other in a multi-hop manner on both licensed and unlicensed spectrum bands.

The Cognitive Radio (CR) could be used in many contexts to improve the exploitation of the spectrum. For example, a Cognitive Radio could negotiate cooperatively with others and thus improve the sharing of spectrum. CR was therefore conceived as the objective of the evolution of software defined radio platforms. It should become completely reconfigurable wireless system that was able to adjust its communication parameters automatically depending on the demands of the network and its users.

A CR can also identify and occupy the "hole" in the spectrum so-called "white space", avoiding interference with other transmissions.

"Spectrum holes represent the potential opportunities for non-interfering (safe) use of spectrum and can be considered as multidimensional regions within frequency, time, and space." [30]

To better understand these gaps in the spectrum is very useful to refer to Figure 2.4, where the three axes represent the Time, the Signal Strength and the Frequency.

Figure 2.4: Spectrum Hole Concept [5]

The Spectrum Holes are regarded as partial or total absence of power in the time-frequency plane, and that their dynamic utilization over time, jumping from one frequency range to another, corresponds to a Dynamic Spectrum Access (DSA) that improves the use of the common wireless channel.

In a CRN, an Unlicensed User (SU) can:

sense idle spectrum (Spectrum Sensing)

select the best available channel (Spectrum Decision)

coordinate access to this channel with other Unlicensed User (SU) (Spectrum Sharing)

vacate the channel when a Licensed User (PU) needs that channel (Spectrum Mobility)

These features are highlighted in the CR Cognitive Cycle shown in Figure 2.5.

Figure 2.5: Cognitive Radio Cycle [2]

To implement CRAHNs, each basic function needs to be incorporated into the classical layering protocols. The Figure 2.6 shows that Sensing and Sharing are low-layer functions, while Decision and Mobility take place in all levels of the protocol stack.

Figure 2.6: Spectrum Management Framework [2]

2.1 Spectrum Sensing

The Spectrum Sensing is a low-layer function of the CRNs Protocol Stack.

Figure 2.7: Spectrum Management Framework: Spectrum Sensing [2]

The Spectrum Sensing allows to know the spectrum and the existence of PUs in a certain geographical area.

In CR, it involves the determination of features that are not the simple measure of the Power Spectral Density, but also includes Modulation, Waveform, Bandwidth, Carrier Frequency of all transmissions, in order to have a clear view of the environment.

Generally, Spectrum Sensing techniques can be classified into four groups:

Primary Transmitter Detection

Cooperative Detection

Primary Receiver Detection

Interference Temperature Management

2.1.1 Primary Transmitter Detection

In Primary Transmitter Detection, in order to distinguish between used and unused spectrum bands, CR users detect the signal from a primary transmitter through only the local observations of CR users.

Figure 2.8: Primary Transmitter Detection [20]

Three schemes are generally used for the transmitter detection:

Matched Filter Detection

Energy Detection

Feature Detection

The Matched Filter Detection requires "a priori" knowledge of the characteristics of the PU signal, but also synchronization between the primary transmitter and the CR user. If this information is not accurate, then the matched filter performs poorly.

In the Energy Detection Scheme, CR users sense the presence of the PUs through the energy of the received primary signal. In order to measure this amount of energy, the received signal is squared and integrated over the observation interval. Finally, the output of the integrator is compared with a threshold to decide if a PU is present. This scheme is easy to implement, but it can only determine the presence of the signal and cannot differentiate signal types. Thus, the energy detector often generates false detection triggered because it depends only on the SNR of the received signal. So, its performance is susceptible to uncertainty in noise power.

Modulated signals are, in general, characterized by cyclostationarity, since their mean and autocorrelation exhibit periodicity.

The Feature Detector exploits this inherent periodicity in the PU's signal by analyzing a Spectral Correlation Function. It distinguishes between the noise energy and the modulated signal energy, because that the noise is a wide-sense stationary signal with no correlation, while the modulated signals are cyclostationary with spectral correlation due to the built-in periodicity. For this reason, the main advantage of the Feature Detection Scheme is its robustness to the uncertainty in noise power.

The Feature Detector is also capable of differentiating different types of signals. Therefore, it can perform better than an Energy Detector in differentiating different signal types. However, it is computationally complex and requires significantly longer observation time.

2.1.2 Cooperative Detection

Because of the lack of interaction between the PUs and the CR users, the only Transmitter Detection Techniques cannot avoid causing interference to primary receivers because of the lack of primary receiver information as depicted in Figure 2.9.

Figure 2.9: Receiver Uncertainly [20]

Moreover, transmitter detection models cannot prevent the "Hidden Terminal" problem.

A CR user may not be able to detect the primary transmitter due to shadowing as shown in Figure 2.10.

Figure 2.10: Shadow Uncertainly [20]

Cooperative detection is theoretically more accurate, since the uncertainty in a single user's detection can be minimized through collaboration.

Figure 2.11: Cooperative Transmitter Detection [20]

Moreover, multipath fading and shadowing effects can be mitigated so that the Detection Probability is improved in a heavily shadowed environment as shown in Figure 2.11.

In traditional cooperative detection, the spectrum band is decided to be available only if no primary user activity is detected. Even if only one primary user's activity is detected, CR users cannot use this spectrum band.

The choice of type of cooperation in the spectrum sensing affect the final results.

There are three type of cooperation schemes:




In the Centralized scheme, a central unit collects information from all other radios, analyzes them, and then inform all devices on the results.

In the Distributed scheme, every radio share this information with any other, but it takes itself the decision about which part of the spectrum to use.

The last possibility is the External scheme where an external agent analyzes the environment and provide the information on employment of the spectrum to all radios.

2.1.3 Primary Receiver Detection

The most efficient way to detect spectrum holes is to detect the PUs that are receiving data within the communication range of a CR user.

As depicted in Figure 2.12, the primary receiver usually emits Local Oscillator (LO) leakage power from its RF front-end when it receives signals from the primary transmitter.

Figure 2.12: Primary Receiver Detection [20]

In order to determine the spectrum availability, a Primary Receiver Detection method exploits this LO leakage power instead of the signal from the primary transmitter, and detects the presence of the primary receiver directly.

2.1.4 Interference Temperature Management

The FCC has introduced a new model for measuring the interference: the Interference Temperature.

The idea is to limit the interference measured by the receiver to a maximum value calculated from the power of background noise.

This "Interference Temperature Limit", is the amount of new interference that the primary receiver could tolerate. As long as CR users do not exceed this limit, they can use the spectrum band.

In this way, as shown in Figure 2.13, there may be transmission of SUs, whose powers, added to the power of the background noise, remain below the temperature threshold of interference.

Figure 2.13: Interference Temperature Management [5]

Although this model is best fitted for the objective of spectrum sensing, the difficulty lies in accurately determining the interference temperature limit and the position of the primary receivers and transmitters.

Also, with the increase in the Interference Temperature Limit, the SNR at the primary receiver decreases, resulting in a decrease in the primary network's capacity and coverage.

2.2 Spectrum Decision

Figure 2.14: Spectrum Management Framework: Spectrum Decision [2]

Once the available spectrums are identified, the SUs have the knowledge of the channels availability. Note that such availabilities are Location-Dependent and Time-Varying, which is incurred by the activities of the PUs.

These two characteristics are unique to Cognitive Radio Systems.

Due to the sharing agreement, channels used by PU cannot be utilized by SUs in the vicinity. Therefore, that nodes within a certain range of each PU cannot reuse the same frequency (Location-Dependence).

Figure 2.15: Sample of Network Topology [22]

For example in Figure 2.15, Node 2 is within the interference range of PU 1 who

uses channel B. Therefore, Node 2 can use only the set (A,C).

Furthermore, due to the traffic load of the PUs, a SU user may observe different channel availability over time (Time-Variance).

The "Resource Allocation" problem is how the SUs should select the most appropriate band according to:

QoS Requirements

Channels Availability

Moreover, CR users should also consider the available "Spectrum Fluctuations", determined by specific channel characteristics such as path loss, interference and link errors.

When channel availabilities change, SUs need to adjust their channel allocation accordingly. They may also need to exchange information with neighboring nodes.

2.3 Spectrum Sharing

Spectrum Sharing as Spectrum Sensing, lie in the low-levels of CRN Protocols Stack as depicted in Figure 2.16.

Figure 2.16: Spectrum Management Framework: Spectrum Sharing [2]

The shared nature of the wireless channel necessitates coordination of transmission attempts between CR users.

In this respect, in CRNs, the vacant spectrum bands of PUs are opportunistically shared by the CR users. This is a fundamental requirement since it allows:

A new way for spectrum management

The coexistence of PUs, who have the right to use frequencies that are licensed, and SUs who want to exploit the holes in the spectrum in both time domain frequency

In CRNs, the Spectrum Sharing function must also consider the fluctuations in the spectrum bands due to the following reasons:

CR users can transmit data only if the available spectrum bands are accurately detected

Heterogeneous QoS requirements of the CR users

Spectrum Sharing can be regarded to be similar to generic Medium Access Control (MAC) problems in existing systems.

The analysis of CR spectrum sharing techniques has been investigated through three major theoretical approaches.

The first classification for Spectrum Sharing techniques in CRNs is based on the "Architecture", which can be described as follows:

Centralized Spectrum Sharing

Distributed Spectrum Sharing

In the Centralized Spectrum Sharing technique, a centralized entity controls the spectrum allocation and access procedures. Each entity in the CRN forward their measurements about the spectrum allocation to the central entity and this entity constructs a "Spectrum Allocation Map".

Distributed techniques are mainly proposed for cases where the construction of an infrastructure is not preferable. Each node is responsible for the spectrum allocation and access is based on local policies.

The second classification for Spectrum Sharing techniques in CRNs is based on the "Access Behavior":

Cooperative Spectrum Sharing

Non-Cooperative Spectrum Sharing

In the Cooperative Spectrum Sharing, the interference measurements of each node are shared among other nodes. With this technique an efficient usage of the spectrum is possible and is mainly used for simple deployment of various applications.

The main problem in the open spectrum approach is designing an efficient way of managing the spectrum for the SUs. This approach also requires frequent coordination and information exchange among the users that increases the stress on the communication resources of a particular network.

All the centralized solutions can be regarded as cooperative, there also exist distributed cooperative solutions.

The Non-Cooperative Spectrum Sharing is a "device-centric" spectrum management scheme in which the users do not coordinate with each other for exchanging information. The users behave independently on the local observations. This spectrum approach reduces the control traffic and simplifies the allocation behavior.

Table 2.1 shows the advantages and disadvantages of Cooperative Spectrum Sharing and Non-Cooperative Spectrum Sharing.

Cooperative Spectrum Sharing

Non-Cooperative Spectrum Sharing


All nodes work together and communicate with each other by means of a common protocol

All nodes work independently based on the local observations of the network

Efficient usage of the spectrum is possible and is mainly used for simple deployment of various applications

This technique reduces the control traffic and simplifies the allocation of the communication resources

No common channel control problem exists

Nodes work based on the interference of the neighboring nodes

This improves scalability and reduces the deployment costs

Spectrum management for the SUs is possible in this technique


Efficient spectrum management for the SUs is the main problem in this technique

Even though fair service is guaranteed for each user, the communication overhead cannot be reduced

The management scheme maximizes interference and does not provide fairness among the users

Neighbors cannot Exchange coordination information frequently

This technique normally increases the stress on the communication resources of a particular network

Common channel control problem exists

Security is a major issue in this technique

This increases the deployment costs

Table 2.1: Advantages and Disadvantages of Cooperative and Non-Cooperative Spectrum Sharing [27]

Finally, the third classification for Spectrum Sharing in CRNs based on the "Access Technology":

Overlay Spectrum Sharing

Underlay Spectrum Sharing

In Overlay Sharing, a node accesses the network using a portion of the spectrum that has not been used by licensed users. As a result, interference to the PU is minimized.

Underlay Spectrum Sharing exploits the spread spectrum techniques developed for cellular networks.

Figure 2.17: Overlay and Underlay Spectrum Sharing [38]

Once a spectrum allocation map has been acquired, an CR node begins transmission such that its transmit power at a certain portion of the spectrum is regarded as noise by the licensed users. This technique requires sophisticated spread spectrum techniques and can utilize increased bandwidth compared to overlay techniques.

2.4 Spectrum Mobility

Is the process by which a CR radio changes its frequency of transmission or reception.

CRs are designed to constantly switch bands in search of the best fitting in a way that is imperceptible to PUs.

Figure 2.18: Spectrum Management Framework: Spectrum Mobility [2]

Spectrum Mobility as Spectrum Decision needs a coordination between the different layer in the CRN Protocol Stack, as shown in Figure 2.19.

Figure 2.19: Functional blocks for Spectrum Mobility in CRNs [2]

If a SU moves, the "Spectrum Allocation Map" may change rapidly. Therefore, the spectrum allocation map constructed by the sensing algorithm may become obsolete with high mobility.

Consequently, the CR user may need to perform Spectrum Sensing as they change location.

This necessitates an adaptive spectrum sensing technology that is responsive to the mobility of the SU.

Moreover, SU are regarded as ''visitors'' to the spectrum they allocate. When a PU starts to access that particular portion of the spectrum used by SU, the CR user should vacate that portion and should move to another vacant portion of the spectrum in order to maintain the communication without any interruption.

Also, the quality degradation of the current transmission initiates Spectrum Mobility.

As a result, Spectrum Mobility is important for successful communication between CR nodes.

The following Figure 2.20 shows an example of spectrum in use by PUs and spectrum holes available for SUs.

Figure 2.20: Illustration of Spectrum in use and Spectrum Holes [31]

It gives rise to a new type of handoff in CRNs, the so-called "Spectrum Handoff" where a temporary communication break is inevitable due to the process for discovering a new available spectrum band.

The purpose of the Spectrum Mobility Management in CRAHNs is to ensure smooth and fast transition during a Spectrum Handoff. It occurs when:

PU is detected

the CR user loses its connection due to the mobility of users involved in an on-going communication

a current spectrum band cannot provide the QoS requirements

Another main functionality required for Spectrum Mobility in the CRAHNs is the "Connection Management".

2.4.1 Spectrum Handoff

For seamless communication in dynamic radio environments, Spectrum Handoff should support intelligent connection releasing and re-establishing procedures during frequencies switching.

The most crucial factor in determining the performance of Spectrum Mobility is the "Spectrum Handoff Delay".

Each time a CR user changes its frequency, the network protocols may require modifications on the operation parameters, which may cause protocol reconfiguration delay.

Another delay that must be considered, is related to the spectrum and route recovery time and the actual switching time determined by the RF Front-End reconfiguration.

To mitigate this delay effect, Connection Management needs to coordinate the spectrum switching by collaborating with upper-layer protocols.

Two different strategies are utilized to implement the Spectrum Handoff:

Reactive Spectrum Handoff

Proactive Spectrum Handoff

In the first strategy, CR users perform spectrum switching after detecting link failure due to Spectrum Mobility. The spectrum switching is immediate without any preparation time, resulting in significant quality degradation in on-going transmissions.

Reactive Spectrum Handoff is generally used in the event of a PU appearance.

In Proactive Spectrum Handoff, CR users predict future activity in the current link, determine a new spectrum while maintaining the current transmission, and then perform spectrum switching before the link failure happens.

Proactive Spectrum Handoff is suitable for the events of user mobility or spectrum quality degradation. These events do not require immediate spectrum switching, and can be easily predicted.

2.4.2 Connection Management

When the current operational frequency becomes busy in the middle of a communication by a CR user, then applications running in this node have to be transferred to another available frequency band.

An important requirement of connection management protocols is the information about the duration of a spectrum handoff.

Multi-layer mobility management protocols are required to accomplish the spectrum mobility functionalities. These protocols support mobility management adaptive to different types of applications.

2.5 Network Layer

At the network layer, the selection of the transmission bands and the routing path must be undertaken jointly.

The classical routing tables for ad-hoc networks have limited information regard the network, because they keep only the information about next hop.

For CRNs, is necessary to include other parameters, as the channel, transmission rate, modulation and such other that are unique to each link. Thus the routing tables for CRNs have full information about the environment.

A CRN must first sense the available spectrum opportunities before it can begin designating a route.

Is important to consider the "Channel Switching Delay", which affects the final end-to-end performance. So is necessary to choose the channels that minimize the number of channel switches along the path.

In literature the Routing Protocols for CRNs can be classified by:

Spectrum Decision

Joint Spectrum Decision with PU Awareness

Joint Spectrum Decision and Re-Configurability

In the Spectrum Decision Protocols,