rough estimation can be done by per capita waste generation multiplied by population of the city. For accurate estimate....go for wardwise population record from Municipality or corporation and multiply according to per capita waste generation of particular ward. Because it changes according to the life style
If you want to include commercial waste stream in your estimate together with domestic waste per capita, you can use waste generation factors available from the link
If it is a European city, then the figures should be available online from government website. In UK municipalities have to report NAWPI - performance indicator figures. These are available on-line from government websites.
You can sample and measure waste generation and composition in different areas and sectors in the city, given than municipal waste is produced in domestic, comercial and services sources. Your samples should follow an statistical approach, and can be taken at the source or in the landfill or processing site.
Most of the developing countries have landfill sites for municipal solid waste. Visit the local site and seek the help of scavengers to sort, segregate and weigh the in coming waste. You can do this for 7 days and then extrapolate.
Another , lengthy, way is to hire assistants who can conduct door to door surveys in selected neighborhoods. Base the selection on some criterion that you might later be able to justify. You can use the results to evaluate your hypothesis.
You need to get the generation rate per person per day, and multiply by the number of days in a year. To get the daily generation rate, you need to (i) classify the estates of the city by socio-economic status (as this determines the generation rate); (ii) sample households from each estate, e.g. systematic random sampling or other convenient means; (iii) supply collection bags for each sampled household, and let them place their wastes in the container for a specified number of days, e.g. 1 week, then you collect and weigh. Supply a new container to cover the new week. Keep this record against the demographic profile of each household e.g. household size. Repeat this at least thrice for each household...ie do it for at least 3 weeks.
Do this in all the identifiable sampled socio-economic groupings of the city (estates), and then extrapolate by multiplying with the number of residents in each estate / socio-economic class. Once you do this for all classes, you are done with the residential / domestic wastes. To cover temporal variations, you can do this quarterly to ensure you even out the seasonal variations. This approach covers domestic wastes. you can do the same for other generators of waste e.g. commercial, health care etc.