The build up these abilities by exploiting late

The wireless revolution is feeding a greedy demand for access to the radio frequency (RF) spectrum all over the nation and the world. Furthermore, in consumer aspect, there are several devices competing for the bandwidth. Smart phones are the most well-known ones; however, wearable fitness applications and smart home applications are also wildly known.Moreover, this competition is also significant in military applications including any kind of sensors, from underwater to the space used for satellites. While this request is expanding, and cause an approaching shortage of RF range, this problem is getting more serious for the nation.So what is the problem with the present approach? Since it secludes remote frameworks by partitioning the spectrum into inflexible, solely authorized groups, at any given time, many dispensed groups are unused by licensees while different groups are overpowered, along these lines wasting the range’s tremendous limit and superfluously making states of shortage. Also, it is human-driven and not versatile to the flow of supply and demand.The aim of the Spectrum Collaboration Challenge (SC2) is mainly to grantee the access of increasing wireless devices to the electromagnetic spectrum. The challenge is designed in a way that competitors have to build a wireless model in which radio systems will independently team up and reason about how to share the RF range, subsequently avoiding interference and looking forward possibilities to share the available range.In SC2 each team will build up these abilities by exploiting late advances in Artificial intelligence (AI) and machine learning, and the growing limits of programming radioscite{SC2}.%%%%%%%%%%%https://www.darpa.mil/program/spectrum-collaboration-challenge%%%%%Unlike the past competition, in SC2, the radios have to learn how to make strategies in a way that lead them to use the optimized spectrum which is not applicable by today’s approaches which actually pre-allocate access to selected frequencies. And this will be reached in light of new machine learning and Artificial Intelligence approaches. Actually, the radios which can make strategies are making decisions as an intelligent system, so a potential application of this challenge is other applications of collaborative decision-making.As mentioned before, there are many applications that are in need of the result of this challenge, from refrigerators and automobiles to military operations. But among them all, military operations rely on this challenge outcomes to assess the dynamic environment and perform their significant purposescite{SC1}.%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%https://www.darpa.mil/news-events/2016-03-23%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%In the United States, the National Telecommunications and Information Administration, which is inside the Department of Commerce, manages radio frequency allocations.This project is a follow-up research of the project of our Collaborative and Intelligent Radio Network class last year. In the previous project, we implemented one of the algorithms that are presented in this paper. The results we obtained from the previous work was not satisfying, meaning that the radio agent could not learn the behavior of its competent completely so these agents happened to interfere most of the time. The average scores were not increasing while the number of rounds accumulated. In this project, we improved the previous work in an outstanding manner.