Title: Measuring Channel Capacity Using Machine Learning for Evaluation of Radio Receivers
Project ID: x9zD72
Domain(s): Machine Learning

Description:

Measuring channel capacity across non-AWGN channels is quite challenging as there is often not a closed form solution for fading channels. Several different methods have been proposed to estimate mutual information using modified versions of the k-nearest-neighbors algorithm. By estimating mutual information channel capacity can be calculated. These techniques can be used for evaluating the efficacy of equalizers and other receiver components by measuring the changes in mutual information. MATLAB may be used to complete this task as it has implemented many of the mathematical and machine learning functions needed to complete this task.

Desired Skills:
Information theory, machine learning, signal processing

Clearance-

US Citizenship Required: No
Active Clearance or Background Investigation Required: No
Level Needed:

Team Information-

Targeted Students: Grad,Undergrad
Team Size: 2 to 4
Details: Could be completed with 1 but is easily enough work for 2-4 since more complex receivers could be added to the scope.

Specific Requirements-

Focus on Particular University: No
Details:

Timeline-

Focus Timeline: No
Details:

Funding-

Potential Funding:No
Note: Availability of funds not guaranteed


I’m Interested In this Project