Research Of Automatic Impedance Matching In RF Antenna

In the practical application of radio frequency communication system,antenna impedance is affected sharply by the changes of environment or frequency.This makes the antenna impedance and characteristic impedance of transmission cannot maintain a good state.Mismatch decreases maximum available transfer power obtained from the antenna source,even the formed reflection will lead to the distortion of the whole system signal,which affects the stability of the system seriously.So automatic matching the impedance of system in real-time,stably and fastly is particularly important.Based on automatic impedance matching of RF antenna,this article launched the research in following aspects:The framework of automatic impedance matching system of RF antenna is gived and the process and principle are also introduced.Select the adjustable ? network to realize impedance transformation by comparing,considering the effect of passive components from high frequency characteristics,analyze the the non-ideal factors of inductance and capacitance then to establish the corresponding parasitic model.quantum particle swarm optimization algorithm and Genetic algorithm is introduced,and the automatic matching in ideal model and parasitic model are carried out,respectively.Results show that the ideal model can present a mismatch state in the practical application,whereas parasitic model can obtain better conjugate matching.Meanwhile the paper points out the problem of the two methods of intelligent algorithm faced and the fact that there are high parasitic losses in parasitic model.Then the interpretation of Smith chart theory and its matching steps provide theoretical support for the derivation of analytic method.In view of the problem that commonly used intelligent matching optimization have slow convergence speed and low accuracy even easily trapped in local minima.an improved population search algorithm is adopted for impedance matching.Compare the improved population search algorithm with genetic algorithm,quantum particle swarm optimization and population search algorithm through the typical test functions,and embed the four types of algorithm into the impedance matching to matching the ideal model.Simulations verify the effectiveness and the fast speed and high accuracy convergence properties of the proposed algorithm which can also jump out of local extremum.Then use single and double target optimization matching equivalent parasitic model based on the algorithm,and compare with the actual situation of ideal model.Results show that the importance of considering the parasitic effects,also double target matching of parasitic model has the advantages of maintaining good matching and reducing the parasitic loss.Considering the influence of parameters setting in intelligent optimization results and the complexity of the modeling iteration,the analytical method to quick solving element parameter is studied.Aiming at the low pass ? type network,propose an analytical method based on Smith chart and circuit principle,through theoretical derivation to acquire the corresponding formula for getting the element parameter values.Analytical method simulations verify the validity and compare with the three kinds of intelligent methods.The matching time of analytical method can reach microsecond,which is 0.94%,1.7%and 2.2%of millisecond matching time of intelligent methods,respectively.This proves the quickness of matching.To weaken the error effects of component values discrete operations in practical application,propose a discrete matching of neighborhood search based on the analytical method.Simulation examples indicate that the eight neighborhood search which fixing the capacitor near the antenna in the matching network can quickly get the best matching performance.