Without knowing more details about your model, it is impossible to say why it's not identified. It could have to do with the number of indicators per factor, the possible lack of correlation between factors with only two indicators, or many other issues. A model may be structurally underidentified (it would not be identified for any set of data) or empirically underidentified (generally identify able but not identified in your specific case due to a specific issue in your data).
Can you provide more details? To solve the problem you need to find out why your model is underidentified and then fix the issue if possible.
@Irum Kindly provide a screenshot of your model from AMOS with the error pop-up message (if any). Then only others would be able to diagnose the problem and provide any specific suggestion.
Irum Yasmeen I totally agree with my respected colleagues that without enough information it is impossible to give any suggestions. However, the most possible reason for this can be the Quality of Data at the first place. Though quality of data depends on multiple factors even before data collection process, run a data cleaning check on your sample. Check for missing values, cases with no variance, and, most importantly, for Outliers because extreme outliers can significantly impact on model fit and variable relationships. Tq.