In this classical paper you can find some methods and links.
Sreerama, N., and Woody, R. W. (2000) Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set Anal Biochem 287, 252
A more recent one is from Norma J. Greenfield with links there or at her institutional website:
From practical point of view always know exactly, which protein concentration you have in your sample and which pathlength you used. These are crucial input parameters in all these softwares. Very importantly, protein concentration is related to the concentration of the amide units (= amino acid units - 1) and not the total molar mass of the protein ! (amide units are the chromophores of the circularly polarized UV/VIS light).
As an approximation you can work with a molar mass of around 100 g/mol per amino acid. Or exactly, you sum up all molar masses of the amino acids of your protein and divide by the total number of amino acids of your protein.
In addition to the comment by Martin Müller I will add the link to the CDPro site where you can download the program by Woody, Sreerama et. al.
http://lamar.colostate.edu/~sreeram/CDPro/
If you are using a JASCO CD running the Spectra Manager II program there is a free interface available that will run the CDPro programs quickly and easily in Windows. You can request that from JASCO. In addition JASCO has developed a new Multivariate secondary structure estimation program which utilizes PCR/PLS and a newly developed reference database. Unlike other methods this new method also allows the users to add their own data to the reference data sets or even create their own separate database.