IILM College of Engineering and Technology

Why Cognitive radio will be lifebloodfor future 5G wireless systems?

The attention of telecom operators and companies have finally shifted towards 4G wireless systems to provide broadband mobile services to widespread users any time, any place and anywhere. However, surge in mobile traffic and demand for more sophisticated broadband mobile services has attracted the attention of think-tanks and major telecom industry players towards tighter integration of wireless technology with higher speed, requiring the development of new generation of mobile communication : 5G. Evolution towards 5G requires convergence of Internet services with existing mobile networking standards referred commonly as ‘mobile Internet over Heterogeneous networks’ with very high broadband speeds. Some of the design targets include 10-100x peak data rate,10x energy efficiency, 10-30x lower latency and 1000x network capacity paving the path towards Gigabit wireless.

However, there lies no single technological advancement to meet the projected future traffic demand of 1000x. Instead different amalgamation of spectrum (Hertz), spectral efficiency (bits/Hertz/cell) and small cell (cells/km2) is required. A key interest lies in making efficient use of spectrum as it acts as transportation system for all wireless communication. According to investigations, there lies 85% of current allocated spectrum partially or completely unused at different times across varied geographical regions. To meet the growing hunger for data, 5G is looking to exploit lower frequency spectrum (below 6 GHz) as well as higher frequency spectrum (millimeter wave bands). This enable 5G to support wide range of spectrum from 400 MHz to 90 GHz.

The behavior of wireless spectrum in terms of coverage region, transmission power, power consumption and bandwidth can be depicted from the graphical picture. Figure 1 depicts lower data rates, large coverage region and higher transmission power for lower frequency signals while higher frequency signals can carry higher data rates but coverage region is smaller with correspondingly lower transmission power.

Fig 1: Wireless radio spectrum behavior (Courtesy : F. Akhtar, M.H. Rehman, M. Reisslein “ White space: Definitional perspectives and their role in exploiting spectrum opportunities”, Telecommunications Policy 40 (2016), 319-331).

You can imagine the role of spectrum in 5G as the role of teacups in a tea party. It is assumed that there is going to be unlimited tea to serve in the party but the number of teacups is going to be limited and has to be dynamically shared among various persons. This requires fast sensing to locate the idle cups which can be reused to serve other users and the sizes of these tea cups can vary based on the availability and demand. The success of tea party depends on how well managed the teacups were to serve each and every person. This analogy paves the background for ‘Cognitive radio’.

Cognitive radio (CR) coined by Mitola in 1999 is a promising technology that has the potential to meet stringent spectrum demands of future 5G wireless systems. A CR transceiver senses the radio environment to locate partially or unused frequency bands (spectrum holes)by primary (or licensed) users and dynamically adapts its system parameters such as transmit power, carrier frequency, modulation technique etc for employment of unlicensed secondary users. This opportunistic usage of spectrum by secondary users on a non-interfering or leasing basis is done based on agreed policies with primary users or defined by regulating authorities such as FCC (Federal Communication Commission) in the United States and Ofcom (Office of Communication) in the United Kingdom.

CR is intelligent radio where the intelligence is facilitated through awareness, perception, reasoning and decision making. The learning from radio environment to make the radio trainable instead of programmable, can be done by employing different machine learning algorithms for maximizing the utility function. The following figure depicts the learning process in CR.

Fig2: Learning process in CR (Courtesy: N. Abbas, Y. Nasser and K. Ahmad, “Recent advances on artificial intelligence and learning techniques in cognitive radio networks”, EURASIP journal on Wireless Communications and Networking, 2015).

In the end, CR can be seemed as the heart of 5G wireless systems, pumping the system with necessary spectral resources to satisfy the accelerating demand of mobile data among users.

Dr. Astha Sharma

Assistant Professor

Department of Electronics and Communication Engineering

IILM College of Engineering & Technology, Greater Noida



Leave a Comment