Cell Classification in Mobile Networks with Reservoir Computing

Cell classification in traffic analysis and modeling is an important task which is required by planning and optimization of mobile networks. Because traffic is always nonlinear, nonstationary and influenced by immeasureable factors, accurate analytical traffic model can be hardly obtained. Therefore a new classification method using reservoir computing to cell types is proposed. Analysis of field traffic data collected by China Mobile Communications Corporation (CMCC) Heilongjiang Co.Ltd is achieved. Experiments results show that new method adopting reservoir computing is feasible and effective for cell classification by traffic for mobile networks.