A Novel Multiple Access Scheme In M2M Communication System Over Cellular Network With Massive MIMO

From the "Perception of China" to "Smart Cities",the Internet of Things as an important technology has been rapidly developing and carrying more and more business needs.Through the realization of direct machine-to-machine communication,M2M(Machine to Machine)communication becomes the cornerstone of the Internet of Things technology.In order to realize M2 M communication over existing mobile cellular systems,LTE-A(Long Term Evolution-Advanced)proposes the concept of hybrid deployment model.Aplication servers obtain upload information of the M2 M terminals through direct and indirect way.But because of such a large number of M2 M communication node trying to access the network simultaneously,the main users of cellular networks are bound to be influenced.In this paper,the technology of multi-access based on large-scale antenna is studied in the M2 M communication scenario,and the inter-machine communication access scheme in the cellular network is improved by involving large-scale antenna array technology.Firstly,this paper analyzes the degree of deterioration of pilot contamination when M2 M terminals access to base station in the way which are equivalent to H2H(Human to Human)terminals,and explains the necessity of cluster-based M2 M communications.Based on a simple channel estimation and two anti-pilot contamination channel estimation methods,this paper shows that the degradation of pilot pollution caused by the introduction of a large number of M2 M terminals seriously affects the accuracy of channel estimation,resulting in base station declined throughput.For the three channel estimation methods considered in this paper respectively,the capacity is decreased by 2 times,and the MSE(Mean Squared Error)is decreased by about 15 dB and 4 dB.Secondly,this paper studies the cluster-based M2 M communication which degrades the pilot contamination,and improves the multi-access technology of M2 M communication based on clustering in Massive MIMO(Multiple Input Multiple Output)scenario.Frequency resource usage ratio is improved.The cluster head layer interconnects through the D2 D links to form a connectivity map.First,the uplink data of the M2 M terminals are aggregated,then headers communicate with the base station cooperatively.In the improved SCMA access mode,Massive MIMO channels of the cluster heads are used as the orthogonal resources to encode the sparse codes.Compared with the traditional SCMA method,in which frequency is orthogonal resources,the usage ratio of resources is improved.Meanwhile,the decreases of uplink throughput and the power consumption of the allocated codeword elements are the cost.In this paper,we first get the connected graph of the minimum number of links and the total power consumption through the optimization method,then design the codebook allocation algorithm.The algorithm can approximate the optimal power cost by far less the time complexity of the violent algorithm.Finally,this paper designs timing scheme and base station beamforming based on multi-length pilot scheme to reduce the data-pilot interference caused by multi-length pilots,thus improving the throughput of cluster head layer and enhancing the capacity of M2 M terminals.Firstly,the optimal pilot length distribution in the pilot-polluted environment is analyzed,and it is pointed out that the multi-length pilots to improve the throughput of the base station are hindered by the data-pilot interference.In order to reduce the impact of the interference on the system,this paper analyzes the sources of interference in multi-cell at different time stages.By using the orthogonality of channels and downlink beamforming in Massive MIMO system,two timing schemes are designed to reduce interference.Thus enhancing the cluster head layer throughput.