Home | Reports | Technical Documents | Tech-Blog | One-Shot Gallery | Korea ICT News | About Us | List of Contributors | Become a Contributor |  How to Advertise  
  KT SK Telecom LG U+ Korean Vendors Network Architectures  
Section 5G 4G LTE C-RAN/Fronthaul Gigabit Internet IPTV/Video Streaming IoT SDN/NFV Wi-Fi KT SK Telecom LG U+ Network Protocol Samsung  
CHANNELS HFRFRONTHAUL NetvisionMPTCP Springwave1588 PTP   Korea Communication Review      
Cognitive Radio's Main component
August 10, 2017 | By Astro Ahmed @ GrEEK CAMPUS (m.rahm7n@gmail.com)
Online viewer:
Comments (0)

We are pleased to share with you all an interesting article contributed by Astro Ahmed - Interesting areas [Artificial Intelligence, 5G, LTE, Cognitive Radio, Embedded Systems, IoT, Quantum physics] 


Astro Ahmed 

Fresh Graduate Telecommunication Engineer



All Articles by Astro Ahmed  

  How to contribute your article to Netmanias.com !  
  List of Contributors  




 Radios and Resources


All of us know radios like (Wifi 2.4 GHz, Wifi 5 GHz, GSM 2G, 3G, LTE 4G, Bluetooth, RFID, and more and more). All of them provide a way to communicate with each other, to share information, to entertain and to feel alive. The problem is that radios need a medium and resources to share this information and carry it from place to another, simply they are (Frequency, space and time). So Let’s figure out our most valuable resources (Frequency). Do we have a lot of such frequencies to use? 


Actually the answer yes, we are limited by these resources. But there are actually some affords to use the light band Millimeter wave (mmWave) where it lies between 30 GHz and 300 GHz, this spectrum can be used for high-speed wireless communications as seen with the latest 802.11ad Wi-Fi standard (operating at 60 GHz) and streaming high-resolution with the WirelessHD standard, also known as UltraGig. But the problem is that Millimeter wave goes through different losses and it supports only Line of Site (LOS) so it can be used defiantly for indoor only but not suitable for outdoor applications where it will be difficult to be used for mobile broadband.


 Fixing the bugs


so we need to deal our existing resources and try to fix the bugs to optimize and enhance the existing resources, first we have a big spectrum to look at (RF-Spectrum).



RF-spectrum is used by various types of services like satellite, GPS systems, Mobile protocols (2G, 3G and 4G), FM radios, WIFI, Bluetooth and more as I mentioned before. Some of them are licensed and some are not, which means that to operate on a certain band you have to pay money, like mobile services where mobile operators like Vodafone, Verizon, orange .. etc., pay a lot of money to get predetermined frequency band to share their services and applications. Where some are not like WI-FI and Bluetooth, we don’t have to pay money to use them. Each country has an organization which put set of rules of spectrum allocation process like NTRA in Egypt- the following picture contains a picture of spectrum allocation chart in Egypt.



 What is introduced by a Cognitive radio?


The spectrum is scarce and is not utilized, the following picture is a snapshot for the TV spectrum where it shows that there some portions of band are heavily used, lightly used and not used at all.


Cognitive Radio is a combination of a BODY (SDR-Software Defined Radio) and a MIND (Artificial Intelligence). We can think of radios in terms of humans as they need sensors to observe the outside world so as human ear could take sound wave “the air molecules vibrations patterns and transform them into electrical signals which are carried into the brain to analyze and understand them, similarly for radio devices, the Antenna is the front end sensor which is mainly a combination of passive components [Resistance, Inductance, Capacitance), it takes the Electromagnetic waves and convert those waves into electrical signals which are processed by Hardware: Amplification, filtering and (ADC) analog to digital converter and after dirty long process they are mapped to software in order to analyze them in intelligence way. 


So, cognitive radios can be understood as radios that gain awareness about their surroundings and adapt their behavior accordingly. For instance, a cognitive radio may determine an unused frequency band and use that for a transmission, before jumping to another unused band. The cognitive radio terminology was coined by Joseph Mitola and refers to a smart radio that has the ability to sense the external environment, learn from the history and make intelligent decisions to adjust its transmission parameters according to the current state of the environment.

 Software Defined Radio (SDR)


SDR is a radio communication technology that is based on software defined wireless communication protocols instead of hardwired implementations. In other words, frequency band, air interface protocol and functionality can be upgraded with software download and update instead of a complete hardware replacement. SDR provides an efficient and secure solution to the problem of building multi-mode, multi-band and multifunctional wireless communication devices. SDR Performs the majority of signal processing in the digital domain using programmable DSPs and hardware support, but some signal processing is still done in the analog domain, such as in the RF and IF circuits. The ultimate device, where the antenna is connected directly to an A-D/D-A converter and all signal processing is done digitally using fully programmable high speed DSPs.  All functions, modes, applications, etc. can be reconfigured by software. Where it Is Flexible and Brings Analog and Digital World Together. Software defined radio (SDR) technology brings the flexibility, cost efficiency and power to drive communications forward, with wide-reaching benefits realized by service providers and product developers through to end users.


 Machine and Deep Learning Techniques


Now in order to understand how to actually analyze signals in intelligent way we need a cognitive engine powered by a machine or Deep learning algorithm to observe, analyze, predict and make decision. But first we need to understand for now that the main difference between machine learning and deep learning is features engineering where in machine learning, the learning model decides to classify the situation based on fixed set of features (observed/measured parameters) but in deep learning, the algorithm has to decide which features fit into this situation in order to classify so learning process is based on features selection and classification.



let’s look at a very simple, yet effective, procedure called supervised learning which is simply means that the answers is known for each set data sample(we supervise the training process where the actual answer is known). Here we feed the neural network -which is one the popular learning algorithms- vast amounts of training data, labeled/marked by Human pictures examples so that a neural network can essentially fact-check itself as it’s learning. The pictures are the data; “Humans” are the labels, depending on the picture. 


As pictures are fed in, the network breaks them down into their most basic components(features), i.e. edges, textures and shapes. As the picture propagates through the network, these basic components are combined to form more abstract concepts, i.e. curves and different colors which, when combined further. At the end of this process, the network attempts to make a prediction as to what’s in the picture. At first, these predictions will appear as random guesses, as no real learning has taken place yet. If the input image is an “Human”, but “Dog” is predicted, the network’s inner layers will need to be adjusted. The adjustments are carried out through a process called backpropagation to increase the likelihood of predicting “Human” for that same image the next time around. This happens over and over until the predictions are more accurate when achieving the global minima where the error is optimized. Hence the error is simply being the actual output – the desired output, and it could be computed mathematically in many different ways. 



This learning process takes place always in Radios to create a radio which is brain empowered (cognitive radio) so we can think of it as a way to optimize resources and use the radio resources with the full capacity and in optimum way. 


 Cognitive Cycle


The figure below includes the simplified cognitive cycle where we sense the Radio Environment using some sensing techniques like doing the Fast Fourier Transform in order to detect the signal strength level in the spectrum band then the sensory Data are mapped into the learning model to find out the corresponding solution. After that the radio is dynamically reconfigured to the corresponding parameters like transmitted power, sampling rate, center frequency or the wave form. And if there is any new solution which is absolutely based on a new problem it shall be stored in the Data base and this process allows us to construct and build our system memory - learning by experience.



Thank you for visiting Netmanias! Please leave your comment if you have a question or suggestion.






Subscribe FREE >>

Currently, 47,000+ subscribed to Netmanias.

  • You can get Netmanias Newsletter

    (New contents, Korea ICT News)

  • You can view all netmanias' contents

  • You can download all netmanias'

    contents in pdf file












SK Telecom's Massive IoT Deployment through LoRa for Small Things



SK Telecom commercialized the world’s first nationwide LoRa-based, IoT dedicated network in the end of June. This report will discuss how well SK Telecom is poised for the emerging IoST sector, and where it is heading.









How to contribute articles to Netmanias!

We always welcome contributed articles. Share your expertise and inspire others!






View All (675)
4.5G (1) 5G (59) AI (2) AR (1) ARP (3) AT&T (1) Akamai (1) Authentication (4) Big Data (2) C-RAN/Fronthaul (17) CDN (4) CPRI (4) Carrier Ethernet (3) China Mobile (2) Cloud (2) CoMP (6) Connected Car (3) DHCP (5) Ericsson (1) FTTH (6) GSLB (1) GiGAtopia (2) Gigabit Internet (19) Google (7) Google Global Cache (3) HLS (5) HSDPA (2) HTTP Adaptive Streaming (5) Handover (1) Huawei (1) IEEE 802.1 (1) IP Routing (7) IPTV (20) IoST (3) IoT (44) KT (40) Korea (18) Korea ICT Market (1) Korea ICT Service (13) Korea ICT Vendor (1) LG U+ (18) LTE (70) LTE-A (16) LTE-B (1) LTE-H (2) LTE-M (3) LTE-U (4) LoRa (7) MPTCP (2) MWC 2015 (8) NB-IoT (6) Netflix (2) Network Protocol (17) Network Slicing (4) Nokia (1) OSPF (2) OTT (3) PCRF (1) QoS (3) RCS (3) SD-WAN (11) SDN/NFV (49) SK Broadband (2) SK Telecom (33) Samsung (5) Security (10) Self-Driving (1) Small Cell (2) Spectrum Sharing (2) Switching (6) TAU (2) UHD (5) VR (2) Video Streaming (12) VoLTE (8) VoWiFi (2) Wi-Fi (21) YouTube (6) eICIC (1) eMBMS (1)
Password confirmation
Please enter your registered comment password.