Cognitive radio system (CRS)

A Cognitive Radio System (CRS) is a radio system capable of obtaining information about the characteristics of its own operation and, on the basis of this data, adjusting its operating parameters.

Cognitive radio (CR) is a radio receiver that can be programmed and dynamically configured to use the best wireless channels in its vicinity, in order to avoid user interference and congestion. Such a radio station automatically detects available channels in the wireless spectrum and then adjusts its transmission or reception parameters accordingly, to enable simultaneous wireless communication within a given spectrum band in a single location. This process is a form of dynamic spectrum management.

Cognitive radio is an advanced technology on the way to more rational use of the radio-frequency spectrum

The mechanism of dynamic spectrum management is applied in intelligent radio systems. The term cognitive radio (CR) has been proposed to describe them. A distinctive feature of such systems is:

• the ability to extract and analyze information from the surrounding radio environment,
• to predict changes in the communication channel
• to optimally adjust their internal parameters, adapting to changes in the radio environment.

ITU-R Report 2117

“Cognitive radio is a radio system that uses radio communication technologies with programmable parameters and other technologies to automatically configure its operating mode in order to achieve desired objectives. Such a radio system is capable of accumulating knowledge about operating conditions, as well as dynamically and independently adapting its operational parameters to the corresponding environment, and can remember the results of its actions and the models used for a given environment

Cognitive radio is not a separate radiocommunication service, and can be used as a technology within any of the existing radio services. For a radiocommunication service to use cognitive radio systems in a given frequency band, that service must have an allocation in the corresponding frequency band with the right to operate on a primary or secondary basis.

Cognitive radio technologies can be used when reconfiguring connections between terminals and multiple radio systems, by radiocommunication system operators for more efficient management of the radio-frequency spectrum resources they use, as a tool for collective access to the radio-frequency spectrum, and as a tool for organizing more flexible access to the radio-frequency spectrum.

Classification of cognitive devices

The following classification of cognitive devices has been adopted, based on the method they use to access the radio-frequency spectrum, ensuring the exclusion of radio interference to existing radiocommunication facilities:

• systems with a cognitive pilot signal;
• systems based on individual or collective monitoring of channel occupancy;
• systems with a geolocation database of protected radiocommunication facilities.

A combination of these methods is also possible.

Cognitive system with a control channel

Terminal operation process with a control channel (CRS)

Main functions of the control channel:

• helps a mobile terminal select the required network depending on special conditions;
• ensures efficient use of radio resources by transmitting instructions on required actions from the network to terminals;
• provides reconfiguration capability, allowing the terminal to determine the most suitable radio access technology;
• Provides data on the state of the environment, helping the terminal determine the exact frequencies, operators and access methods in a specific area without the need to carry out a lengthy spectrum scanning process;
• helps the network provider simplify the process of making dynamic changes to network deployment by informing terminals about the availability of new radio access technologies/frequencies;
• reveals the benefits of secondary spectrum trading and flexible spectrum use, both from a technical and an economic standpoint, making terminals independent of the frequencies used for special services at any given time.

Operating principle of software-defined radio systems

Radio Access Technologies (RAT)

Cognitive radio systems can cover a range of Radio Access Technologies, and self-configuring networks of various network topologies will be able to make use of spectrum based on what is available locally. In this connection, it is necessary to determine the location and characteristics of other radio access technologies within the frequency band covered, which is reachable from a mobile terminal, as well as to scan the entire tuning range, in order to determine local spectrum usage.

Within the European Union, it was determined that devices must:

• support multiple radio access technologies
• have the capability of dynamically determining available technologies and free radio-frequency resources.

Interference

Research conducted to date has shown that to guarantee the exclusion of interference to existing radio equipment from cognitive radio systems, it is necessary to use a geolocation database containing information on the location and operation of protected radiocommunication facilities . The database ensures the selection of operating frequencies for cognitive radiocommunication systems in such a way as to ensure both the possibility of their operation and the absence of interference to other radiocommunication systems and facilities. In this case, other radio-electronic facilities (operating under the normal frequency assignment procedure) have priority in the use of radio-frequency channels over cognitive radio systems.

In practice, this means that if a certain frequency band or radio channel begins to be used by other radio equipment, the cognitive radio system using that band or channel must switch to another frequency/channel. Thus, cognitive radio technology cannot in principle be used as the sole technology for providing, for example, paid services, since the constant availability of radio-frequency spectrum for cognitive radio operation is not guaranteed.

Terminology

Depending on the transmission and reception parameters, two main types of cognitive radio are distinguished:

• Full cognitive radio (Mitola radio), in which all possible parameters observable by a wireless node (or network) are considered.
• Cognitive radio with spectrum sensing, in which only the radio-frequency spectrum is considered.

Other types depend on the parts of the spectrum available to the cognitive radio:

• Licensed-Band Cognitive Radio, capable of using bands assigned to licensed users (excluding unlicensed bands such as the U-NII band or the ISM band). The IEEE 802.22 working group is developing a standard for a wireless regional area network (WRAN) that will operate on unused television channels, also known as TV white spaces.
• Unlicensed Cognitive Radio, which can use only the unlicensed parts of the radio-frequency (RF) spectrum. One such system is described in the IEEE 802.15 Task Group 2 specifications, which focus on the coexistence of IEEE 802.11 and Bluetooth.
• Spectrum mobility: the process by which a cognitive radio user changes their operating frequency. Cognitive radiocommunication networks seek to use spectrum dynamically, allowing radio terminals to operate in the best available frequency band, ensuring seamless communication during transitions to better spectrum.
• Spectrum sharing : cognitive radio networks with spectrum sharing allow cognitive radio users to jointly use the spectrum bands of licensed-band users. However, cognitive radio users must limit their transmit power so that the interference caused to licensed-band users does not exceed a certain threshold value.
• Sensing-based spectrum sharing: ^[10] In cognitive radio networks with sensing-based spectrum sharing, cognitive radio users first listen to the spectrum allocated to licensed users in order to determine the status of the licensed users. Based on the detection results, cognitive radio users determine their transmission strategies. If the licensed users are not using these bands, the cognitive radio users will transmit in these bands. If the licensed users are using the bands, the cognitive radio users share the spectrum bands with the licensed users while limiting their transmit power.
• Database-based spectrum sharing, ^{[11] [12]},: ^[13] In this spectrum sharing modality, cognitive radio users need access to a white space database before being allowed or denied access to the shared spectrum. The white space database contains algorithms, mathematical models and local rules for predicting spectrum usage in a geographic area and for inferring the risk of interference caused to existing services by a cognitive radio user with access to the shared spectrum. If the white space database determines that harmful interference to existing operators would occur, the cognitive radio user will be denied access to the shared spectrum.

Technology

Although cognitive radio was originally conceived as an extension of software-defined radio (full cognitive radio), most of the research work has focused on spectrum-sensing cognitive radio (especially in television bands). The main challenge of spectrum-sensing cognitive radio is the development of high-quality spectrum-sensing devices and algorithms for exchanging spectrum-sensing data between nodes. It has been shown that a simple energy detector cannot guarantee accurate detection of signal presence, ^[14] requiring more sophisticated spectrum measurement methods and requiring regular exchange of spectral measurement information between nodes. Increasing the number of cooperating sensing nodes reduces the probability of false detection. ^[15]

One possible approach is adaptive filling of free radio-frequency bands using OFDMA. Timo A. Weiss and Friedrich K. Jondral of the University of Karlsruhe proposed a spectrum pooling system, in which free bands (sensed by nodes) were immediately filled with OFDMA sub-bands. Applications of spectrum-sensing cognitive radio include higher throughput for emergency networks and WLANs and increased transmission range. Cognitive radio is evolving towards cognitive networks; the concept of cognitive networks is the intelligent organization of a network of cognitive radios.

Functions

The main functions of cognitive radio:

• Power control: Power control ^[18] is typically used for CR systems with spectrum sharing to maximize the throughput of secondary users under interference power constraints, in order to protect primary users.
• Spectrum sensing: detecting unused spectrum and sharing it without harmful interference to other users; an important requirement of a cognitive radio network is sensing of empty spectrum. Detection of primary users is the most effective way of detecting empty spectrum. Spectrum sensing methods can be divided into three categories:
  • Transmitter detection: cognitive radios must be able to determine whether a signal from a primary transmitter is locally present in a given spectrum band. Several approaches to transmitter detection are proposed:
    • Matched filter detection
    • Energy detection: Energy detection is a spectrum measurement method that determines the presence/absence of a signal simply by measuring the power of the received signal. ^[19] This approach to signal detection is quite simple and convenient for practical implementation. However, implementing an energy detector requires information about the noise variance. It has been shown that imperfect knowledge of noise power (noise uncertainty) can lead to the phenomenon of the SNR wall, which represents the SNR level below which an energy detector cannot reliably detect any transmitted signal, even by increasing the observation time. ^[20] It has ^[21]also been shown that the signal-to-noise ratio boundary is caused not by the presence of the noise uncertainty itself, but by insufficient refinement of the noise power estimate as the observation time increases.
    • Cyclostationary feature detection. This type of spectrum measurement algorithm is motivated by the fact that most artificial communication signals, such as BPSK, QPSK, AM, OFDM, etc., exhibit cyclostationary behavior. ^[22] However, noise signals (typically white noise) do not exhibit cyclostationary behavior. These detectors are robust to noise variance uncertainty. The goal of such detectors is to exploit the cyclostationary nature of artificial communication signals hidden behind noise. Their main decision parameter is a comparison of the nonzero values obtained via the CSD of the primary signal. ^[23] Cyclostationary detectors can be single-cycle or multi-cycle.
• Wideband spectrum sensing: refers to spectrum sensing over a large spectral band, typically hundreds of MHz or even several GHz. Since modern ADC technology cannot afford a high sampling rate with high resolution, it requires revolutionary methods, for example, compressive sensing and sub-Nyquist sampling. ^[24]
  • Cooperative detection: refers to spectrum measurement methods in which information from multiple cognitive radio users is combined for detection of the primary user ^[25]
  • Interference-based detection
• Null-space-based CR: using multiple antennas, the CR detects the null space of the primary user and then transmits in the null space, so that its subsequent transmission causes less interference to the primary user.
• Spectrum management: capturing the best available spectrum to meet the user's communication requirements, without causing excessive interference to other (primary) users. Cognitive radios must select the best spectrum band (from all available bands) to meet quality-of-service requirements; therefore, cognitive radio receivers require spectrum management functions. Spectrum management functions are divided into:
  • Spectral analysis
  • Spectrum decision ^{[26] [27]}

Practical implementation of spectrum management functions is a complex and multifaceted matter, since it must comply with numerous technical and legal requirements. An example of the former is the choice of an appropriate sensitivity threshold for detecting other users, while an example of the latter is the need to comply with the rules and regulations established for access to the radio-frequency spectrum in international (ITU Radio Regulations) and national (telecommunications law) legislation.

Smart Antenna (SA)

Smart antennas (or smart antenna systems) are an antenna technology that uses spatial beamforming and spatial coding to suppress interference; nevertheless, applications are emerging for extending intelligent multiple or cooperative antenna arrays for use in complex communication environments. Cognitive radio, by comparison, allows user terminals to determine whether a portion of the spectrum is being used, for spectrum sharing with neighboring users. The following table compares the two:

Note that both methods can be combined, as shown in many modern transmission scenarios. ^[28]

Cooperative MIMO (CO-MIMO) combines both methods.

Applications of cognitive radio

Cognitive Radio (CR) can sense its environment and, without user intervention, can adapt to the user's communication needs while complying with FCC rules in the US. Theoretically, the amount of spectrum is infinite; in practice, for propagation and other reasons it is finite, owing to the desirability of certain portions of the spectrum. Assigned spectrum is used far from fully, and its efficient use is a growing concern; CR offers a solution to this problem. CR can intelligently determine whether any portion of the spectrum is in use, and can temporarily use it without interfering with the transmissions of other users. ^[29] According to Bruce Fette, «some of the other cognitive capabilities of a radio include determining its location, determining spectrum use by neighboring devices, changing frequency, adjusting output power, or even changing transmission parameters and characteristics. All these capabilities, along with others yet to be implemented, will give wireless spectrum users the ability to adapt to spectrum conditions in real time, offering regulators, licensees and the general public flexible, efficient and comprehensive use of the spectrum».

Examples of applications include:

• Use of CR networks for emergency and public safety communications using white space ^{[30] [31]}
• Capabilities of CR networks to perform dynamic spectrum access (DSA) ^{[32] [33]}
• Use of CR networks in military operations, such as detecting and investigating chemical, biological, radiological and nuclear attacks, command and control, obtaining battle damage assessment information, battlefield surveillance, assisting reconnaissance and target designation. ^[34]
• They have also proven useful in building body-area medical networks ^[35], which can be used for ubiquitous patient monitoring, helping to immediately notify physicians of important patient information such as blood sugar level, blood pressure, blood oxygen and electrocardiogram (ECG), etc. This provides the additional benefit of reducing the risk of infections, as well as increasing patient mobility.
• Cognitive radio is also applicable to wireless sensor networks, where packet relaying can take place using primary and secondary queues to forward packets without delay and with minimal power consumption. ^[36]

Modeling CR networks

Currently, modeling and simulation is the only paradigm that allows complex behavior in cognitive radio networks of a given environment to be modeled. Network simulators such as OPNET, NetSim, MATLAB and ns2 can be used to model a cognitive radio network. CogNS ^[37] is an open-source NS2-based modeling platform for cognitive radio networks. Research areas using network simulators include:

1. Spectrum sensing and detection of the incumbent
2. Spectrum allocation
3. Measurement and/or modeling of spectrum usage ^{[38] [39]}
4. Spectrum utilization efficiency ^{[38] [39]}

Network Simulator 3 (ns-3) is also a viable option for modeling CR. ^[40] ns-3 can also be used for simulation and experiments in CR networks using standard hardware, such as Atheros WiFi devices. ^[40]

Future plans

The success of the unlicensed band in accommodating a range of wireless devices and services has prompted the FCC to consider opening additional bands for unlicensed use. In contrast, licensed bands are underutilized due to static frequency allocation. Recognizing that CR technology can make use of inefficiently used licensed bands without causing interference to existing users, the FCC issued a Notice of Proposed Rulemaking that would allow unlicensed radios to operate in television broadcast bands. The IEEE 802.22 working group, formed in November 2004, is tasked with defining an air interface standard for wireless regional area networks (based on CR sensing) to operate in the unlicensed spectrum of the distributed television service.^[41]To comply with the FCC's later rules on unlicensed use of TV spectrum, IEEE 802.22 defined interfaces to a mandatory TV white space database, in order to avoid interference to existing services.

Further work on the development of cognitive radio:

Development of provisions for equipment certification and compliance;

• Provisions for equipment relocation;
• Consideration of different approaches that may be used by different administrations when implementing cognitive radio;
• Consideration of the need to develop regulatory measures concerning the possible use of cognitive radio system technologies;
• Consideration of the potential need for worldwide harmonization of the control channel (bandwidth less than 50 kHz);
• Consideration of candidate bands and spectrum needs;
• Deciding whether regulatory tools are needed for models without a control channel (database model, white spaces).

The E2R Consortium, together with the European Telecommunications Standards Institute (ETSI), is conducting a study of the following issues:

• overcoming design difficulties, power consumption issues, and the development of cognitive radio measurement and standards;
• the fact that existing technologies do not provide all the requirements needed for universal application of cognitive radio. It is assumed that a software-defined radio (SDR) system will:
• operate at all frequencies from 9 kHz to 300 GHz, using a harmonized control channel with a bandwidth of less than 50 kHz wherever possible;
• receive/transmit and modulate/demodulate signals in all frequency bands and all modulation modes
• have core radio parameters that can be changed via reconfigurable software

32 organizations from 14 countries, budget of 19.0 million euros

The E2R Consortium, including the largest telecommunications operators

See also

• Channel allocation schemes
• Channel-dependent scheduling
• Cognitive network
• LTE Advanced
• Network simulator
• OFDMA
• Radio Resource Management (RRM)
• White spaces (radio)
• Gaps (database)
• Software-defined radio
• SDR [[b7977]]

See also

• Software-defined radio