A Spectrum Handoff Scheme based on Joint Location and Channel State Prediction in Cognitive Radio

In this paper, we present a learning based spectrum handoff based on secondary user mobility and primary user activity prediction to provide efficient spectrum access. The movement of secondary user in wireless network yields location-varying spectrum opportunities. The current approaches in most literatures which only depend on reactive selection spectrum opportunities result of waste idle spectrum. Moreover, conventional selection method tends to observe a higher handoff and delays in network performance. This inefficiency can cause transmission interruptions leading to the degradation of services. In this work, we aim to improve the performance of spectrum handoff and their operation delays. Therefore, we develop a learning based spectrum handoff that predicts the next secondary user movement and primary user activity using Hidden Markov Model (HMM). We compare our prediction strategy with a conventional sensing and location estimation method. Through simulations, we show that our strategy reduces the number of handoffs and operation delays significantly.