International Journal of Engineering and Techniques - Volume 3 Issue 5, Sep - Oct 2017 ISSN: 2395-1303 http://www.ijetjournal.org Page 154 Reinforced Q learning for WiFi/WiMAX Network in Heterogeneous Environment Divya Parambanchary 1 , V.Malleshwara Rao 2 Department of Electronics & Communication Engineering, Gitam University,Vishakapatnam. Andra Pradesh,India 530045 I .Introduction In today’s wireless technology no single network fulfils all conditions. Advancement in technology and up gradation of system design is the requirement of the era .Due to increased data capacity and multimedia traffic the network redesign is required to implement any new technology. Maximizing the network capacity and efficiently Utilizing the network resource with unbalanced traffic load parameters should be considered before designing the network. The user yields enhanced benefits from integrated technology. Integrating various technologies such as Wi-Fi and Wi- Max should provide users with better QoS and seamless handover. In wireless communication satellite and mobile communication are commonly used and are in focus with very high requirement of data capacity. Different technologies and different coverage areas play vital role. Wireless fidelity (Wi-Fi) technology gives narrow area coverage and small-cell networks. It is available only in a smaller area called as hotspots. The worldwide interoperability for microwave access (Wi-Max) technology yields high data rate, wide area coverage, and built-in support for mobility and security [1]. Users give more priority to Wi-Fi than Wi-Max, because of its low cost and reduced power consumption. Recently, many developments are made in Wi-Fi and Wi-Max integration. The major aspect of Wi-Fi & Wi-Max integration is to handover the packet without any loss [2].Mobility produces handover data between Wi-Fi and Wi-Max technologies in a HetNet (heterogeneous network) environment. Major disturbances faced by handover process are unbalanced traffic energy of Wi- Fi and delivery through the access points (APs) and base station sub-system (BSS) covered by hotspot. There is high variation of traffic at Wi-Fi access points that fluctuates with time, as some AP are extensively used under traffic scenario but other AP remain free. Each cellular node should have two connections, one is for Wi-Fi and another is for Wi-Max in cellular network. A group of overlying APs creates a Wi-Fi inside the Wi-Max coverage area. It creates number of Wi-Fi area inside Wi-Max area to enable Wi-Fi as well as Wi-Max. In Wi-Max coverage, BS can have more than one Wi-Fi spots. Multiple Wi-Fi spots shall cover the whole Wi-Max area. If there will be multiple data flow in the network, the unbalanced traffic load may occur. In this case, based on the priority, the Wi-Fi and Wi-Max split the data and transfers to the receiver for maintaining the Quality of Service. To avoid any lack in the quality of service, the bandwidth management algorithm is used to distribute the bandwidth properly across the networks by using AP and BS. Data flow RESEARCH ARTICLE OPEN ACCESS Abstract In wireless communication maintaining QoS is a very challenging issue .In heterogeneous environment WiFi & WiMAX technologies are integrated together. Integrating different technologies such as WiFi,WiMAX ,3G,4G,5G focuses to provide users with better (QoS)and seamless mobility.QoS is determined by throughput, end to end delay, jitter and packet loss. In any integrated heterogeneous network ,the user requirements are wide coverage, high bandwidth and access cost should be low. Proposed work uses Machine level algorithm as an intelligent technique to collaborate WiFi and WiMAX in heterogeneous environment. Due to dynamic operating system designers, Q algorithm eliminates the redesign of the existing network. It also maximizes the network utilization and also helps to design various machine learning results for various applications. Keywords — HetNet, Wi-Fi,Wi-Max ,Routing, Machine learning, Cellular networks, heterogeneous