Interference Mitigation for Dynamic TDD Networks Employing Sounding Signals

Abstract

The requirements of fifth-generation (5G) mobile communications include services with low delay and high throughput. Network densification is pointed to as a promising method to increase the network capacity. However, this solution brings new problems, such as the fast variation in traffic demands among access nodes (ANs) and between uplink and downlink, leading to high delays. To solve the traffic issues in dense networks, dynamic time division duplex (DTDD) is pointed as a possible solution. This strategy creates a new kind of interference between ANs and user equipments (UEs) called cross-interference. Therefore, obtaining channel state information (CSI) of cross-interference channels is essential for implementing interference mitigation methods, such as interference alignment, coordinated beamforming, resource schedulers, among others. Hence, this work proposes methods to estimate the intended and interfering channels based on sounding reference signal (SRS) and/or demodulation reference signal (DMRS). A coordinated scheme is developed to assign sounding signals in the network and reduce the interference perceived during the channel sounding, which improves the channel estimation quality. Furthermore, a refined successive interference cancellation (SIC) algorithm is proposed for estimating the channel. To assess system performance, a zero-forcing beamforming algorithm has been developed based on the CSI acquired with the proposed methods. This algorithm handles the degrees of freedom issues when ANs are operating in opposite directions. The numerical results show that the improved channel quality provided by the proposed estimation algorithm increases network capacity.