diff --git a/README.rst b/README.rst
index 48be488..57b9fdb 100644
--- a/README.rst
+++ b/README.rst
@@ -1,12 +1,9 @@
 SPECTRUM : Spectral Analysis in Python
 ==========================================
 
-
-
 .. image:: https://badge.fury.io/py/spectrum.svg
     :target: https://pypi.python.org/pypi/spectrum
 
-
 .. image:: https://secure.travis-ci.org/cokelaer/spectrum.png
     :target: http://travis-ci.org/cokelaer/spectrum
 
@@ -16,13 +13,11 @@ SPECTRUM : Spectral Analysis in Python
 .. image:: https://landscape.io/github/cokelaer/spectrum/master/landscape.png
     :target: https://landscape.io/github/cokelaer/spectrum/master
 
-.. image:: https://badge.waffle.io/cokelaer/spectrum.png?label=ready&title=Ready 
-    :target: https://waffle.io/cokelaer/spectrum
 
 
 :contributions: Please join https://github.com/cokelaer/spectrum
 :issues: Please use https://github.com/cokelaer/spectrum/issues
-
+:documentation on RTD: http://pyspectrum.readthedocs.io/ 
 
 
 
@@ -50,6 +45,7 @@ For Linux and MAC users, if you prefer to use conda (avoiding
     conda config --add channels conda-forge 
     conda install spectrum
 
+To install the **conda** executable, please see https://www.continuum.io/downloads .
 
 Contributions
 ==================
diff --git a/doc/joss/paper.bib b/doc/joss/paper.bib
new file mode 100644
index 0000000..9ac2917
--- /dev/null
+++ b/doc/joss/paper.bib
@@ -0,0 +1,41 @@
+
+@book{marple:1987,
+  address = {Australia, Sydney},
+  author = {Marple, S. L.},
+  publisher = {Prentice Hall},
+  title = {Digital Spectral Analysis With Applications},
+  year = 1987
+}
+
+@book{percival:1993,
+  title={Spectral analysis for physical applications: Multitaper and Conventional Univariate Techniques.},
+  author={Percival, Donald B and Walden, Andrew T},
+  year={1993},
+  publisher={cambridge university press}
+}
+
+@article{harris:1978,
+  author = {Harris, F. J.},
+  doi = {10.1109/PROC.1978.10837},
+  issn = {0018-9219},
+  journal = {Proceedings of the IEEE},
+  month = jan,
+  number = 1,
+  pages = {51--83},
+  publisher = {IEEE},
+  title = {On the use of windows for harmonic analysis with the discrete Fouriertransform},
+  url = {http://dx.doi.org/10.1109/PROC.1978.10837},
+  volume = 66,
+  year = 1978
+}
+
+@article{welch:1967,
+  author = {Welch, P.},
+  journal = {IEEE Transactions on Audio and Electroacoustics},
+  volume = 15,
+  number = 2,
+  pages = {70--73},
+  title = {The use of fast Fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms}
+  year = 1967
+  }
+ 
diff --git a/doc/joss/paper.md b/doc/joss/paper.md
new file mode 100644
index 0000000..9d32f74
--- /dev/null
+++ b/doc/joss/paper.md
@@ -0,0 +1,38 @@
+---
+title: 'Spectrum': Spectral Analysis in Python
+tags:
+  - spectral analysis
+  - periodogram
+  - yule-walker
+  - multi-tapering
+  - burg
+  - ARMA
+  - eigen-values
+  - tapering windows
+authors:
+ - name: Thomas Cokelaer
+   orcid: 0000-0001-6286-1138
+   affiliation: 1
+ affiliations:
+ - name: Institut Pasteur
+   index: 1
+date: 2 August 2017
+bibliography: paper.bib
+---
+
+# Summary
+
+**Spectrum** is a Python library that includes tools to estimate Power Spectral Densities. Methods
+available are based on Fourier transform, parametric methods or eigenvalues analysis. 
+
+- The Fourier methods are based upon correlogram, periodogram and Welch estimates. Standard tapering windows (Hann, Hamming, Blackman) and more exotic ones are available (DPSS, Taylor, ...)[harris:1978],[welch:1967],[marple:1987].
+- The parametric methods are based on Yule-Walker, BURG, MA and ARMA, covariance and modified covariance methods [marple:1987],[percival:1993].
+- Non-parametric methods based on eigen analysis (e.g., MUSIC) and minimum variance analysis are also implemented [marple:1987].
+- Multitapering method is also available [percival:1993]
+- Classical tools useful to spectral analysis and more generally signal processing such as window tapering [harris:1978] or transfer function are also available within the library.
+
+The following image shows the different methods of spectral estimation that are available in **Spectrum**.
+
+-![https://doi.org/10.6084/m9.figshare.5270866.v1](psd_all.png)
+
+# References
diff --git a/doc/joss/psd_all.png b/doc/joss/psd_all.png
new file mode 100644
index 0000000..4b59e80
Binary files /dev/null and b/doc/joss/psd_all.png differ