Abstract: In this study, we present a method in which the dynamic spectrum gets optimized for its usage in cognitive radio networks. Firstly, based on the interference temperature model, we determine the interference constraints and there is a check on the transmission power of the secondary users. We then formulate the SINR by considering the noise spectral density of the environment and the transmission powers of the surrounding neighbors. Using the Shannon formula, we calculate the co-channel interference between potential links on each channel. Next, using Genetic algorithm, we formulate the spectrum assignment problem. Genetic algorithms are adaptive heuristic search algorithm based on previous and historic results obtained earlier. Although, random, Genetic algorithms exploit historical information to direct the search into a set of better solutions. It is observed that the total network capacity is improved by a significant margin bio inspired algorithms have always been promising in giving optimum solutions.