Background And Literature Review On Cognitive Radio 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.

Chapter 5

This chapter provides background on some of the fundamental concepts. The characteristics of the wireless communication channels are described in section 5.1, in addition, Section 5.2 details on multi-carrier systems and Orthogonal Frequency Division Multiplexing (OFDM). The following section contains the paradigm of cognitive radios and its physical architecture. In the end, section 5.4 points the literature review on Spectrum Sensing for cognitive radio systems and the other work done on the subject.

5.1 Wireless Channel Characteristics

The characteristics of a wireless communication channel between transmitter and receiver controls the performance of the overall system. Thus, in order to realize the effects of wireless channel on the operation of spectrum sensing we first need to realize its characteristics. The mobile radio environment characteristics which are used in this dissertation are introduced in this section.

Communication systems are operating mostly in diversified environments, including those prone to fading. For that reason, spectrum sensing must be analysed in fading environments. There are two types of fading effects called as large-scale fading and small-scale fading that characterize mobile communications (Rappaport 1999). The propagation models that characterize signal strength over large transmitter receiver separation distances several hundreds or thousands of meters are called large scale propagation models. On the other hand propagation models that is qualified by the rapid fluctuations of the received signal strength causing multiple versions of the transmitted signal to arrive at the receiver each distorted in amplitude phase and angle of arrival over very short travel distances a few wavelengths or short time durations on the order of seconds are called small scale fading models.

A relative motion between the transmitter and receiver causes significant attenuations of the signal power within a short period of time which is called Doppler shift. Considering all, to overcoming this time varying nature of the multipath wireless channel is very challenging.

Radio channels fading have been classified considering two ways. The first type of assortment discusses whether the fading is flat frequency non selective or frequency selective while the second classification is based on the rate at which the wireless channel is changing or in other words the rate of change of the impulse response of channel i.e. whether or not the fading is fast or slow. It is useful to note the following quantities in connection with these characterizations of fading channels:

• Coherence bandwidth: Coherence bandwidth: Coherence bandwidth is a statistical measure of the range of frequencies over which the channel can be considered "flat" (i.e. frequency non- selective, or in other words a channel which passes all spectral components with equal gain and phase). It is also defined as the range of frequencies through which any two frequency components have a strong potency for amplitude correlation. It has been shown that:


Bc ∝



where στ is the RMS delay spread. Also, if we use definition of the coherence bandwidth Bc as the bandwidth over which the frequency correlation function is above 0.5 (i.e. the normalized cross-correlation coefficient is higher than 0.5 for all frequencies)

5στ then Bc ∝ 1/(5 στ) . Note that if the signal bandwidth is greater than Bc , then the unlike frequency components in the signal will not be faded the same way. The channel then appears to be "frequency-selective" to the transmitted signal.

• Doppler spread and Coherence time: While στ and Bc describe the time dis- persive nature of the channel in an area local to the receiver, they do not offer any information about the time-variations of the channel due to relative motion between the transmitter and the receiver. The Doppler spread BD is defined as a measurement of spectral broadening induced by the time rate of change of the channel associated to the Doppler frequency. The coherence time Tc is a statistical measurement of time duration around which two received signals have a strong potential for amplitude correlation. Thus, if the inverse bandwidth of the baseband signal is larger in size in comparison to the coherence time of the channel then the channel alters during transmission of the baseband message. This will cause a distortion at the receiver. It is shown that:


Tc ∝



Multipath fading has a significant impact on the frangibility of wireless links. It is regarded as a small scale development in the sense that the level of attenuation of the signal changes considerably if the position of the receiver or the transmitter is deviated even slightly about half a wavelength. Shadowing is another physical phenomenon of interest; it has been considered to exhibit large scale effect as it corresponds to substantial deviations of the RF signal from

its mean due to large obstacles which create shadow zones that cause deep fades if a receiver happens to enter them.

In metropolitan areas with mega-structures like big cities, reflections, refractions and diffractions occur as a result of the fact that mobile antennas are lower than the height of the surrounding buildings. In Figure 5.1, it can be seen that the signal arrives at the receiver through various paths.

Figure 5.1 Multipath Propagation

It has been observed that Multipath fading channels have been modelled and simulated since 1950s and early 1960's. The multipath channel is a summation of the transmitted signal replicas with different amplitudes, propagation delays, phases and angles of arrival. The channel impulse response (CIR) of the multipath channel can be modelled as follows;


h(t, τ ) = ∑ αl (t)ejθl (t)δ(τ − τl (t)) (5.3)


where, αl (t) is the real amplitude and θl (t) is the phase value of the lth multipath com- ponent at time t, τl is the excess time that belongs to lth path. L is the total number of multipath components and δ is the unit impulse function that determines the specific excess delay of a multipath at time t.

If the channel impulse response is considered to be constant over the transmission,

it can be further simplified as follows,


h(τ ) = ∑ αl e−jθl δ(τ − τl ). (5.4)


The amplitudes of paths that come at the receiver at the same time delay with unlike phases could add constructively or destructively. Also, within a short period of time the phases of these paths may change. Thus, the resulting amplitude of the channel at a particular time delay could vary within a short time interval. When there are many paths, having independent amplitudes and phases, the channel impulse response h(t, Ï„ ) can be modelled as a complex Gaussian random process based on the Central limit Theorem. Furthermore, if there is no LOS component from the transmitter to the receiver, the am- plitude of the channel can be modelled as Rayleigh fading channel. Nevertheless, if there is presence of a dominant LOS element, it can be modelled as Ricean fading channel.

If a channel yields unvarying response for a bandwidth greater than the transmitted signal bandwidth then the channel is said to be a flat fading channel. The conditions for a flat fading channel are:


Ë‚Ë‚ Bc



˃˃ Tc


where, Bs and Ts are the signal bandwidth and the symbol duration respectively. In the frequency domain, a flat fading channel has a constant amplitude and linear phase response over the transmitted signal bandwidth. For this fading type, the receiver side preserved the spectral features of the transmitted signal (Rappaport 1999). In a flat fading channel, the CIR can be written as:

h(t) = α(t)ejθ(t) (5.7)

where, α(t) is Rayleigh distributed for the channel without LOS path and is Ricean distributed when LOS path exists and θ(t) is uniformly distributed.

Any channel is known to be frequency selective if the signal bandwidth is larger than the coherence bandwidth of the channel. In that case, unlike frequency components of the transmit signal undergo fading to different extents (Rappaport 1999). For a frequency-selective fading situation:

Bs > Bc (5.8)

Ts < Tc (5.9)

From the time domain view, the symbol period is shorter in comparison to the rms delay spread. The channel will spread the signal beyond the symbol period and induce intersymbol interference (ISI) onto the next transmitted symbol. Then the frequency-selective fading channels can be represented as:

Lt −1

h(t, τ ) = ∑ hl (t)δ(τ − lTs) (5.10)


where, hl(t) is the lth tap complex path coefficient at time t whose amplitude is Rayleigh and phase is uniformly distributed; Ts is the symbol period. Frequency selective fading channels are not easy to model compared to flat fading channels as each multipath signal must be modelled.

5.2 Orthogonal Frequency Division Multiplexing (OFDM)

Due to tremendous development in wireless communication systems, there is a increasing demand for higher capacity and higher performance in multi-environment communication systems within the boundaries of limited resources such as spectrum, bandwidth etcetera. The bandwidth provided by GSM is too small to allocate the high data rate applications (e.g. multimedia applications). Therefore, OFDM is proposed as a prospect modulation scheme with multiple carries allowing high data rate transmissions with the advantages of a lower data rate and minimizing the effects of the inter symbol interference (ISI)(Engels 2002 ). There is another saying that suggests the estimate behind OFDM is to convert a frequency selective channel into a collection of frequency-flat sub-channels with partially overlapping spectra.

Generally, higher data rates results serious ISI encountered in the received signals. To avoid the ISI, a large number of closely- spaced orthogonal subcarriers are used to carry data in OFDM systems.

The data are separated into various parallel data streams or channels generally one for each sub-carrier. Each of that sub-carrier is modulated with a established accepted standards of modulation scheme such as QAM PSK at a low symbol rate, holding total data rates similar to established accepted standards of single-carrier modulation schemes in the same bandwidth.

In OFDM, the complete channel is separated into many narrow sub-channels that are used in parallel transmissions. OFDM is an effective method for confronting multipath fading and for high-bit-rate transmission over mobile wireless channels. In addition, OFDM can achieve adaptive allotment of transmission load in different sub-channels to achieve optimum entire transmission rate. Furthermore, OFDM can diminish the effect of impulse noise because of the increased duration of the OFDM symbol. The summation of a number of orthogonal sub-carriers, with baseband data on each sub-carrier being severally modulated normally with the use of some sort of quadrature amplitude modulation (QAM) or phase-shift keying (PSK) resembles an OFDM carrier signal. Typically, this composite baseband signal is utilized to modulate a main RF carrier.

A serial stream of binary digits represented by s[n] is used to transmit message. The messages are first demultiplexed into N parallel streams with inverse multiplexing, and each one mapped to a symbol stream using some modulation constellation (QAM, PSK etc.). Using adaptive modulation there is a possibility that OFDM provides the opportunity to use different modulation schemes for each channel.

In the next step; inverse FFT is calculated on each set of symbols, giving a set of complex time-domain samples. These samples are then quadrature-mixed to passband in the standard way. Using digital-to-analogue converters (DACs) the real and imaginary components are first changed to the analogue domain, the analogue signals are then used

to modulate cosine and sine waves at the carrier frequency, fc, respectively. These signals are then added together to give the transmission signal, s(t).

For the implementation as seen in Figure 5.2, a message stream formed by consisting of binary digits are demultiplexed into N parallel streams using reshape function. Next, parallel streams are modulated to IFFT size must be equal to the number of sub-channels. After transformation of IFFT, by using reshape function again the consequence is then transformed into a serial stream. Finally cyclic prefix is put into the serial stream.

Figure 5.2 OFDM transmitter

Since low bit rate modulation schemes suffer less ISI caused by multipath rather than high data rate modulation schemes, the rationale is very useful to send low data bits in parallel streams instead of high data rate serial streams which is main principle of OFDM modulation scheme(Engels 2002 ). A very common way of combating the delay spread is adding of a guard prefix to every OFDM symbol. To make a successful OFDM process, the cyclic prefix length to insert must be 1/2 or 1/4 of the number of sub-channels. ISI will be averted as long as there is a cyclic prefix period which is longer than the expected delay spread. Due to the operational nature of the FFT, if the length of the cyclic prefix cannot be chosen accurately it can lead to Inter-carrier interference (ICI).

Figure 5.3 Cyclic Prefix Insertion

Through a channel the serial data stream is sent from transmitter to receiver. The receiver then picks up the signal r(t) which is further quadrature-mixed down to baseband using cosine and sine waves at the carrier frequency. This will create signals centred on 2fc, thus, use of low-pass filters reject these. Afterward, using analogue-to-digital converters (ADCs) the baseband signals are sampled and digitised, and then use of forward FFT convert back the frequency domain. Using an appropriate symbol detector each of the obtained N parallel streams are converted to a binary stream. These resulted streams are then altered into a serial stream s[n], which is an approximate of the original binary stream at the transmitter.

Figure 5.4 OFDM receiver

In the receiver side as seen in Figure 5.4, serial data is demultiplexed into N parallel streams removing the inserted cyclic prefix from transmitter. Similarly, the next step follows computing of FFT on each parallel stream and applying equalizer to reduce the channel effects. Afterward, demodulation is carried out for parallel stream in frequency domain (for PSK,QAM etc.) and then transformed from parallel to serial data stream.

As we intend to do the detection in absence of any prior knowledge on primary signal, for spectrum sensing systems, the steps from equalizer block in receiver side are not followed.

5.3 Cognitive Radios

Cognitive Radio (CR) is an epitome that enables an xG network to use spectrum in a dynamic manner. In more formal manner, the term cognitive radio can be defined as follows (FCC Report 2002):

"Cognitive Radio is a radio for wireless communications in which either a network or a wireless node changes its transmission or reception parameters based on the interac- tion with the environment to communicate efficiently without interfering with licensed


Based on the active supervising of external and internal radio environment such as radio frequency spectrum user behaviour trends and network state these changes to parameters are dependent.

Joseph Mitola IIIs in his PhD dissertation (Mitola 2000) first proposed the idea of cognitive radio as a assuring solution for the dynamic usage of the spectrum. Mitola described it as, "the point in which wireless Personal Digital Assistants (PDAs) and the

related networks are sufficiently and computationally intelligent about radio resources and related computer to computer communications to detect user communications needs as a function of use context and to provide radio resources and wireless services most appropriate to those needs".

Simply, the basic idea toward the cognitive radio paradigm is the exploitation of the idle frequency bands allocated to primary users or licensed users by the secondary users or unlicensed users without causing interference to primary users' communications. On scanning the portions of radio spectrum, we would observe that some frequency bands are largely unoccupied most of the time whilst some other frequency bands are only partially occupied and the remaining frequency bands are heavily used (Haykin 2005),(FCC Report 2002). Hence, cognitive radio targets to detect those idle bands and apportion the secondary users to those unused band. It is presumed that cognitive radios are capable of sensing the environment in a period of time and if primary user needs to commence communication, secondary users will then be moved to another idle band. In doing so, the cognitive radios vary the power level so that to caused the interference due to the secondary users insignificant. Thus, we see that recognizing the availability of the idle spectrum band is not enough to decide the usage of unused spectrum bands. Many elements such as frequency selection, modulation schemes and power level should be taken into account to conquer the variation in radio environment so as to avoid possible interference to other users.

The Federal Communication Commission (FCC) in FCC Report 2002 has identified the following features that cognitive radios can incorporate to enable a more efficient and flexible usage of the spectrum.

• Frequency Agility The cognitive radio is capable of changing its operating frequency for its adaptation to the environment.

• Dynamic Frequency To choose an optimum environment to work in signals from nearby transmitter are sensed by the cognitive radio.

• Adaptive Modulation The transmission characteristics and waveforms can be set up in a new way to manipulate all opportunities for the usage of spectrum in an efficient way.

• Transmit Power Control To allow greater dealing of spectrum the transmission power is adapted to full power limits when necessary on the one hand and to lower levels on the other hand.

• Location Awareness The cognitive radio is intelligent enough to find out its localization and the location of other devices operating in the same spectrum to optimize transmission parameters for increasing spectrum re-use.

• Negotiated Use The cognitive radio may have algorithms rendering the sharing of spectrum in terms of prearranged agreements between a licensee and a third party or on an ad-hoc/real time basis.

5.3.1 Physical Architecture

Further going into consequences on the cognitive radio, it is better to know its architecture simply in order to understand them better. As seen in Figure 5.7, the main elements of a cognitive radio transceiver are the radio front-end and the baseband pro- cessing unit. Reconfigurability of each component is provided by a control bus to able to adapt the time varying radio environment (Akyıldız, et al. 2006).

The RF front-end consists of an amplifier where the received signal is amplified and then mixed with Analog to digital (A/D) converter by A/D unit. The digitized signal is modulated/demodulated and encoded/decoded in the baseband processing unit. Since the baseband processing unit of a cognitive radio is basically alike to existent transceivers the novel part of the cognitive radio is the RF front-end. For that reason, RF front-end of the cognitive radio must be mentioned in detail. Sensing capability of the RF front-end in a cognitive radio is wideband. RF hardware for the cognitive radio should have capacity of tuning to a large range of frequency spectrum in order to manipulate all opportunities of the frequency spectrum enabling real-time measurements of spectrum information from radio environment. As shown in Figures 5.5 and 5.6, a wideband front-end architecture for the cognitive radio has the following structure.

According to (Akyıldız, et al. 2006), the components of a cognitive radio RF front-end are as follows:

• RF filter: It filters the received RF signal with a band-pass filter and selects the desired band.

• Low noise amplifier (LNA): It is responsible for amplifying the desired signal minimizing noise component at the same time.

Figure 5.5 Cognitive Radio Transceiver

Figure 5.6 Wideband RF/ Analog Front-end

• Mixer: In the mixer, the received signal is mingled with locally generated RF and converted to the baseband or the intermediate frequency (IF).

• Voltage-Controlled Oscillator (VCO): It generates a signal at a specific frequency for a given voltage to mingle with the incoming signal. Such process converts the incoming signal to baseband or an intermediate frequency.

• Phase Locked Loop (PLL): It ensures that a signal is locked on a specific frequency and can also be used to produce precise frequencies with fine resolution.

• Channel Selection Filter: It is used to pick the desired channel and to reject the adjacent channels. There are two types of channel selection filters; direct conversion receiver with a low-pass filter for the channel selection and super heterodyne receiver adopting a band pass filter.

• Automatic gain control (AGC): It preserves gain or output power level of an amplifier constant over a wide range of input signal.

Through the RF front end the wideband signals are received and then high speed analog-to-digital (A/D) converters are used to sample. Furthermore different measurements are carried out for detection of licensed user signal. But, RF antenna receives signal from various transmitters operating at different power levels in practical applications including bandwidths and locations which makes it hard to detect weak signals in those kind of range. So there is need of multi-GHz speed A/D converter consisting of high resolution which is practically unworkable to implement. Additionally the requirement of multi-GHz speed A/D converter calls for the dynamic range of the signal to be reduced before A/D conversion. Filtering stronger signals directs this simplification which can be located anywhere in the wide spectrum range through use of tuneable notch filters. Similarly, implementation of multiple antennas can also be used to reduce these signals.

As mentioned earlier, the major test of the physical architecture of the cognitive radio lies on sensing of frail signals from licensed users over a wide spectrum range as accurately as possible. For this reason, the carrying out of RF wideband front-end and A/D converter are critical issues in xG networks (Akyıldız, et al. 2006).

5.3.2. Characteristics of Cognitive Radio

The dictionary meaning of word cognition means becoming acquainted with, mental process of knowing through perception or reasoning or intuition or knowledge or we can simply say learning by understanding. There are two main characteristics of cognitive radio which is worth mentioning; according to (Akyıldız, et al. 2006) they are cognitive capability and reconfigurability which is describe in detail as follow: Cognitive Capability

Cognitive capability can be defined as capability to obtain information about the unused spectrum in the radio environment, that provides, the cognitive users with best operating

parameters to use the spectrum efficiently without any interference to the primary users. This feature makes it resourceful and efficient to cooperate with the real radio environment in order to sense appropriate communication parameters that are essential for cognitive users to use. Consequently, cognitive radio has to execute some tasks which is referred as cognitive cycle and is shown in Figure 5.7 (Akyıldız, et al. 2006). Reconfigurability

The reconfigurability in cognitive radio is the potential of programming the radio enthusiastically without any alteration in hardware components. Cognitive radio can be structured to be used as transmitter or receiver. Also, in different frequency they can use different modulation techniques with variable transmission power with respect to the communication link. Such capability of CR is achievable as an intrinsic result of the development of software-defined radio (SDR) platform which is completely reconfigurable wireless device capable to adjust its communication parameters in response to either network or user demands. A software-defined radio (SDR) is a radio communication system that can adjust to any frequency band over a large spectrum and obtain any modulation through programmable hardware which is controlled by software.

Figure 5.7 Cognitive Cycle

Since parameters for each of the functional modules are fixed a complete hardware based radio system has limited utility. On the contrary, SDR broadens the usefulness of the system for a wide range of applications that use different link-layer protocols and modulation/demodulation proficiencies. Thus, the cognitive radio systems gets better reconfigurability through software based approach. The reality of reconfiguring the transmission parameters during the transmission indicates that the cognitive radio is competent of configuring both transmitter and receiver parameters in order to switch to different spectrum bands with use of suitable protocols and modulation schemes by passing on appropriate power level of the signal.

5.4 Literature Review

(Arslan and Yu¨cek 2007), concentrates spectrum sensing as the most significant mission among others for the formation of cognitive radio. Some challenges found in spectrum sensing are also presented in the paper. Most common spectrum sensing methods such as Matched Filtering, Energy Detection and Cyclostationary Feature Detection are also studied and equated in terms of advantages and drawbacks.

(Akyıldız, et al. 2006) is one of them among the most elaborated papers written on Cognitive Radios. Here, first portion of the paper concentrates on cognitive radio and its architecture. Similarly, it discusses present and future of the frequency spectrum succeeded by an elaborated analysis of spectrum sensing task. The section covers the discussion on spectrum sharing and its challenges along with details on spectrum sensing methods are studied and compared.

(Srinavasa and Jafar, 2007) concentrates on dissimilar analysis of cognitive radio that underlay, overlay, and interweave the transmissions of the cognitive user with those of licensed users. Underlay approach permits simultaneous primary and secondary transmissions aiming to protect primary users by imposing a spectral mask on the

secondary user's signals so that unlicensed users' interference to licensed users is below the satisfactory noise threshold put for safer communication of licensed users. Overlay approach allows simultaneous primary and secondary transmissions but on contrary to underlay approach, secondary users can use part of their power to communicate and continues to power to assist primary user. If the power split is selected cautiously, the raise in a primary user's SNR because of the help from secondary user is accurately offset by the decrease in primary users' SNR as a result of intervention. Interweave approach is based on Opportunistic Communication (Mitola 2000) which is about using the spectrum holes that specify the truth that they are not in use by primary users. These approaches are also known as Interference Avoidance approach.

The papers presented by (Hur, et al. 2006) , (Tian and Giannakis 2006) and

(Quan, et al. 2008) examines wideband spectrum sensing. On the other hand, further study on wideband sensing is required particularly when bandwidths of the primary signals are unknown within the frequency range of interest along with center frequencies. Energy Detection of unknown signals was focused by (Urkowitz 1967). The paper is one of the most significant and cited paper in the literature on Energy Detection. It considers unknown signal as deterministic although the signal is actually unknown in detail. This paper assumes Deterministic Signal inserted with signal present is Gaussian but not zero mean. The noise present is assumed to be Gaussian and additive with zero mean along with known spectral region. The rationale explains detection in white noise considering whether the noise is a Low-pass or Bandpass such that random process and operating receiver characteristics are specified theoretically.

Furthermore, study reveals that a cyclic periodogram detection for CR environments was focused and implemented by Zhang and Xu on Agilent VEE Pro. Similarly, toward cyclostationary based spectrum sensing systems, (Sadeghi and Azmi 2008) and (Rajarshi and Krusheel 2008) suggested some multiple antenna techniques. (Gardner, et al. 2006) is a reappraisal on Cyclostationary which has been studied by scientists for more than 50 years in many research areas. They initially referred the general attributes of cyclostationary processes and then gave information on implementation in many applications.

In order to palliate the hidden node problem, (Cabric, et al. 2004), (Cabric, et al. 2006) and (Digham, et al. 2003 ) proposed co-operation concept among cognitive radios. But then, in the paper published by (Digham, et al. 2007), (Ashish and Linnartz 2007), (Kuppusamy and Mahapatra 2008) proposed to employ some multiple antenna techniques in order to get over some spectrum sensing challenges.