Resource Allocation in Downlink of LTE using Bandwidth Prediction Through Statistical Information

Abstract

Long Term Evolution (LTE) is the technology used in modern third and fourth generation mobile wireless cellular networks. Due to the presence of large number of users, mobility and varying channel conditions, proper resource allocation is essential to provide a good user experience and improve the system throughput. In this paper, a resource allocation algorithm is implemented that will use the probabilistic models to predict the channel condition and allocate resources accordingly. Also, the algorithm will support QoS requirements. During the resource allocation, the channel quality information is collected and analyzed to predict the future channel conditions and resource allocation vectors are configured accordingly. The performance of the algorithm is analyzed based upon the data collected. The algorithm is able to provide a reasonable success rate for channel prediction. By using the resource allocation vectors and channel prediction, the algorithm performance also is improved considerably due to the lesser space and time complexity required.