—We present an adaptive data aggregative window function (A-DAWF) for a distributed sensor network model in which nodes store data in their attribute window functions, and provide non-correlated data towards the base station (BS). Unlike previous works, namely data collection or data gathering management systems, we propose a novel approach that aims to process temporal redundant techniques in sensor nodes as well as providing spatial redundant filtration methods in cluster-head (CH) nodes. In this regard, preliminary results show that A-DAWF can suppress up to 90% of temporal redundant data among the considered sensor nodes by an optimal threshold of the window sizes, and their spatial correlations in CH node by a maximum error threshold compared to either periodic or a continuous data transmission system. I. INTRODUCTION In WSNs, sensor nodes lifetime always rely on each other. In most of the application scenarios, sensor nodes (SNs) are used as battery powered devices which have limited storages and processing capabilities. Periodically, sensors can spend a lot of energy to transmit or receive the sensor readings which shortens their lifetime as well as their network lifetime. Since data aggregation or data collection has grown as one of the promising area in energy-efficient WSNs for maximizing the network lifetime, and several well-known data aggregation techniques have been studied in the literature [1–8], however they are limited to their specificity. Especially, earlier literature works are designed based on the homogeneous application scenarios than heterogeneous, since many deployment scenarios proved that nodes have their own specific application tasks [9] rather than homogeneous tasks which are rarely used in practice. Having a better aggregation mechanism not only reduces redundant transmissions, it also saves a huge amount of receiver's energy of all other three modules such as radio receiving or listening, computation, and processing procedures of receiver nodes. Hence, it is necessary to use an effective data redundancy technique or simple prediction methods in order to avoid the redundant data transmissions to ensure the reliability in network applications. In this paper, we primarily designed our proposition based on the homogeneous application scenarios, and then heterogeneous. In this regard, we present two sorts of data filtration, one performs in sensor nodes for finding temporal data redundancies (TDRs) using both relative variation (RV) and aggregative window functions, and the another one uses in CH nodes for exploiting spatial data redundancies (SDRs) using both RV and A-DAWF. This paper is presented as work in progress, and the designed implementations of our work are discussed in section III. In section IV, we show the preliminary results to demonstrate that the proposed mechanism can suppress a