Many applications maintain temporal & spatial features in their databases. These features cannot be treated as any other attributes and need special attention. Temporal data mining has the capability to infer casual and temporal proximity relationships among different components of data. In this work a model is going to be developed which helps in measuring traffic data distributed over a wide area. This model considers the assumption that the data follow an ordered sequence. The area is divided into a set of grid points. Each grid point is identified by a set of coefficients. The traffic data at a particular location is measured. The coefficient at the identified location is mapped to measured traffic data value. Thus coefficient at the measured location is calculated. This coefficient is used to generate coefficient values at the other grid points by tridiagonal matrix algorithm. The procedure is repeated till the values cease to change for a unit time. The procedure is repeated for different intervals of time. Thus traffic data is obtained over the wide area for different times and at different locations.