Selection Algorithm In Heterogeneous Network Based On An Economic Model

Future wireless network aims to integrate different radio access technologies (RAT) to provide a seamless access and service continuity. In this paper, a new resource denotation method is proposed in the multi-code with invariable gain CDMA and OFDMA multi-service heterogeneous network based on a concept of spectral bandwidth mapping. In this method, the resources in multi-code with invariable gain CDMA network such as: spread code, received power level and bit error rate etc. and all resource attributes reflecting QoS requirement in OFDMA are mapped into the equivalent spectral bandwidth. And the equivalent bandwidth, as a type of resource denotation method, is used to denote uniformly the resource granularity occupied by each service in the heterogeneous network. Based on this resource denotation method, the corresponding resource allocation algorithm can easily satisfy the QoS requirement of each service.Additionally, since the design of call admission control algorithm depends on the comparison of system residual resources between each RAN as well as between RANs and the user's resource requirement in the heterogeneous network, the uniform denotation of resources is extremely important for the calculation of system residual capacity and moreover for the convenience of call admission control algorithm design. Therefore, this method, as a denotation of resource granularity occupied by each service, simplifies the resource expression and makes the residual capacity calculation tractable in the heterogeneous network.Based on this method, a type of call admission control algorithm relying on a novel economic model is designed in this paper, which is applied to both under-loaded and over-loaded scenario in the integrated network. When the network is over-loaded, our algorithm can continuously admit new connections and control the QoS degradation of each service in a very small region in order to increase the system capacity, thus reaching a balance between the QoS satisfaction of each service and the improvement of system resource utilization efficiency. The simulation results show that this algorithm achieves better performance than the utility function-based access selection (UFAS) method proposed in some papers through increasing system capacity and maximizing system revenue, achieving better load balancing and reducing the new session blocking probability in the multi-service heterogeneous network.