Iterative Reweighted Least Squares (IRLS) algorithm is added to RxCS software. This implementation of the IRLS allows to solve an l1 optimization problem using an l2 solver. You can find the IRLS algorithm implemented in Python on GitHub. An example of using IRLS to solve an underdetermined system is here.
A wrapper which employs IRLS to solve a compressed sensing problem is also availabe on GitHub, an example of usage is here.
A new module is present in RxCS software: Inverse Discrete Hartley Transform . You can find the module on RxCS’s GitHub page , an example of how to use the module is also available on GitHub.
An example of using RxCS module with Kernel Recursive Least Squares algorithm is added to RxCS toolbox . You can find the example on GitHub , or directly in the toolbox in examples/auxiliary dictionary.
A new example of using RxCS software is added. In this example a L1 reconstruction (regularized regression) module is explained. You can find the example here .
Module with Kernel Recursive Least Squares algorithm with Approximate Linear Dependency criterion is added to RxCS toolbox . You can find the module on GitHub . It is a Python implementation of an algorithm written in Matlab by Steven Van Vaerenbergh in his Kafbox toolbox .
More examples of using the RxCS toolbox are added. You can find the examples here . Read more about the RxCS toolbox here.
We have created the first example of using RxCS toolbox . It shows how to generate a multitone random signal using the toolbox. You can find the example here . More examples will appear in the near future.
“RxCS” toolbox v0.01 is ready. You can access it on its GitHub webpage . RxCS is the main tool which is used in the Numerical Simulations part of the project, and it will grow with the project. Stay tuned for the upcoming updates.