Personal Information


  • Virginia Tech / ECE, Research Associate
  • DeepSig Inc, Chief Technology Officer
  • O’Shea Research LLC, Misc Consulting


  • PhD, Electrical Engineering, Virginia Tech
  • M.S. Electrical Engineering, N.C. State University
  • B.S. Computer Engineering, N.C. State University


Please contact me with inquiries related to:

  • Business or product inquiries into DeepSig Radio Machine Learning technologies (please see https://deepsig.io)
  • University research interest in Virginia Tech and related opportunities for students (oshea_at_vt_dot_edu)
  • Other consulting inquiries in the area of wireless, machine learning, and security (tim_at_oshearesearch_dot_com)


3 thoughts on “About

  1. Hi there,

    I work for a company that develope Tuner Modules for car OEMs. For testing purposes I would like to simulate an fading FM channel.

    I read your article about the Dynamic GNU Radio Channel Model Enhancements and would like to use your Fading 2 Block. Is is possible? Where can I download the Enhancements?

    Thank you very much!

    Kind regards,
    Konrad Puchalski

  2. hi sir,
    I am a post-graduate student in UESTC. recently I am studying on automatic modulation classification. I have read your paper and re-finished the experiment by adopting CNN2 as you said in the paper.
    however, the QAM16 and QAM64 seems to be hard to be recognized, which i think longer signal may improve this situation..
    so I decide to build a dataset with longer signals.. and now facing some problems..

    In “https://github.com/radioML/dataset/blob/master/transmitters.py”(dataset/transmitters.py), you import “mapper”, but the progamm faces the error “mapper has no attribute :mapper.BPSK” and I find that the mapper which i downloaded,is not that”mapper” in your code, is it right?

    could you tell me where I can get the right “mapper”…?
    hope you reply..(and i am also sry for my poor english)

Leave a Reply

Your email address will not be published. Required fields are marked *