Simulation of intelligent spectrum sensing in cognitive radios

Abstract

Cognitive Radios are futuristic radio transceivers under research and development at present. These radios will be designed on the principle of finding medium for themselves through dynamic and exploratory sensing of the RF Spectrum in order to find free spaces in it. These free spaces in the RF spectrum are left unused by the licensed users (to whom the bandwidth is leased) either intentionally or unintentionally i.e., absence of any user and are termed as gaps or holes. These gaps are the opportunities for Cognitive Radios hopping onto these gaps for communicating with their peers or any other radio device static or mobile using a variety of modulation, channel encoding and other digital communication techniques and protocols. This is a dynamic process needing extensive but intelligent decision making and hence the name cognitive. At present there is no single dedicated simulator present for simulating the concept in part or as a whole. This project is a step ahead simulating the intelligent spectrum sensing part of the Cognitive Radios using an optimized technique of Radio Environment Mapping and Learning. The project aims at developing a simulator ab-initio to simulates the process of Intelligent Spectrum Sensing i.e., detection of holes or spaces in the given or selected band through repeated iterations in order to map the RF environment and form a hopping sequence comprising mostly available frequencies with priority formed through learning based on weights or knowledge scores. The project is implemented in C# with SQL database at the backend to form a complete picture of the environment. Initial and repeated location and date time aware scanning using a number of iterations is carried out to assign scores or weights to the found or available frequencies. The project was developed using a hybrid software engineering methodology using Spiral with prototyping. Agile project management methodology was used for development.