Base state model with amendment is a general method on spatio-temporal database construction. It accesses data with keeping base state changeless and increment variational. In order to improve querying speed, mono-class or muti-class indexes can be added. Improved base state model with amendment adopts to increase number of base states and increment indexes. This paper proposes dynamic base state model with amendment, and base state changes according to actuality, and increment is difference of actuality and database construction state. This method is of save memory space, convenient use of actuality data, and easy to resume history data.

Multiple-input multiple-output (MIMO) interference broadcast channel (IBC) plays an important role in the modern wireless communications. The upper bound of degree of freedom (DoF) and corresponding achievable schemes have been investigated. However, all the achievable schemes require perfect channel state information at transmitters (CSIT). In the absence of CSIT, the DoF value is still unknown. This paper mainly focuses on theG-cellK-user MIMO IBC, where there areMantennas at each transmitter andNantennas at each receiver. The transmitters only know channel coherent time internals rather than the values of channel coefficients. The users in the same cell are assumed to be able to share the channel information. Based on a heterogeneous block fading model, a blind interference alignment (IA) scheme is proposed for this scenario. We show that when and , then a total of degrees of freedom (DoF) can be achieved. The inner bound is same with the decomposition DoF upper bound.