Title :
A fuzzy Q-learning approach for enhanced intercell interference coordination in LTE-Advanced heterogeneous networks
Author :
Daeinabi, A. ; Sandrasegaran, K.
Author_Institution :
Centre for Real-time Inf. Networks, Univ. of Technol., Sydney, NSW, Australia
Abstract :
Since the transmission power of macro eNodeB (eNB) is higher than pico eNB in long term evolution-advanced heterogeneous network, the coverage area of picocell is small. In order to address the coverage problem, cell range expansion (CRE) technique has been recently proposed. However, CRE can lead to the downlink interference problem on both data and control channels when users are connected to pico eNB. In order to mitigate the downlink interference problem, a new dynamic almost blank subframe (ABS) scheme is proposed in this paper. In this scheme, a fuzzy q-learning approach is used to find the optimum ABS value. Simulation results show that the system performance can be improved through the proposed scheme.
Keywords :
Long Term Evolution; fuzzy set theory; learning (artificial intelligence); radio links; radiofrequency interference; telecommunication computing; ABS; CRE technique; LTE advanced heterogeneous networks; almost blank subframe; cell range expansion; downlink interference problem; eNodeB; enhanced intercell interference coordination; fuzzy Q-learning approach; pico eNB; picocell; power transmission; Downlink; Interference; Long Term Evolution; Macrocell networks; Signal to noise ratio; System performance; Throughput; Heterogeneous networks; almost blank subframes (ABS); fuzzy q-learning; intercell interference; picocell;
Conference_Titel :
Communications (APCC), 2014 Asia-Pacific Conference on
DOI :
10.1109/APCC.2014.7091620
