Call Admission Control In WCDMA/WLAN Network Based On Fuzzy Neural Network

With the development of wireless Internet and the user service diversity, usingheterogeneous network joint technology, seamless handoff and improve the fusionperformance, enhance the user experience has become a big problem, has become aresearch hotspot in recent years. Due to the current3G network and WLAN network,with the development of wireless heterogeneous network load flow increases, existinga network utilization rate is low, the ping-pong handoff rate, blocking rate high, droprate higher seriously affect the user experience of the phenomenon. One of the reasonsis the current on the heterogeneous network access control algorithm and no specificnetwork environment effectively to study, not all aspects to consider specific networkload capacity, channel quality, influence has user movement trend.In view of the WCDMA and WLAN network structure is different, the problemof signal intensity of traditional access control algorithm for direct comparison ofdifferent network is not accurate, this paper analyzes the channel model with twotypes of network, transforming the received signal strength of WCDMA and WLANtwo kind of different network to the contrast between the bit error rate of the channel,the channel access conditions for the two objective a network to do a unified in BER,formulate unified channel of heterogeneous network quality measure.In view of the existing heterogeneous access control of low utilization problemcyber source, WCDMA and WLAN limit capacity formulas are derived in this paper,analysis the effect of disturbance on the capacity, as well as WLAN because the userDCF mode between backoff competition effect on the capacity of WLAN, to establishthe relationship between the number of users and residual capacity, to develop aunified heterogeneous network load measure.In view of the existing access control algorithm for blocking rate, call drop rate ishigh, the utilization rate of resources is low, this paper presents WCDMA fuzzy neuralnetwork with heterogeneous network access control algorithm based on WLAN.Study on the conditions of heterogeneous network channel from the channel, the biterror rate as decision criteria of WCDMA and WLAN channel conditions. Secondly,combined with WCDMA and WLAN heterogeneous network capacity formula, fromthe analysis of capacity, thus the reasonable distribution of cyber source, enhance theeffective utilization rate of network bandwidth. The use of fuzzy neural networkalgorithm, analysis of access control mechanism to trigger decision algorithm,introducing the model and type of business street of Manhattan. Analysis of therealization process of fuzzy neural network control algorithm, including the off-linephase acquisition expert data sample method through simulation, training and onlinereal-time decision. Finally, through the simulation and evaluation of users in different arrival rates performances under the heterogeneous network access control algorithm.Through the analysis of the simulation results, the fuzzy neural network controlalgorithm considering the location information of channel conditions, the networkcapacity conditions, user movement, movement speed and the cell residence time ofthe user, the fuzzy neural network artificial intelligence algorithm, compared to otherexisting heterogeneous network access control algorithm in the blocking probability,call drop rate, average the ping-pong handoff number and cyber source utilizationperformance have improve.