I would like to contribute to this conversation coming from a spectrum research and product design perspective. What is needed for this work is the ability to visualize and then gather statistics using a variety of metrics. For example I have recently been doing a lot of work in the 2.4 GHz band. We all know that is a crowded band. What would be very useful is to be able to go thought a set of measurements at different locations and extract different metrics, like channel utilization. Then when designing a product I want to know the probability that my product will encounter a signal above some level. I want to find opportunities for successful communication in both time and amplitude.
I want to be able to come to my collection of measurements with different questions. For example, I might have an idea for data interleaving to improve reliability. My questions are then how often will each of a collection of algorithms I am considering be overcome based on a collection of measurements that has been gathered. For example OU recently parked an instrument in a hospital room for 84 days. Let's say the data shows some device I am designing would suffer too much interference. Then I want to know how to fix the problem. Do I break the data up into shorter packets? Retransmit it more often? Change channels? If I change channels is that a solution for a lot of hospitals or only this hospital or maybe just this room in this hospital? Let's say some group of us collects data from 10 or 50 hospitals. How do we quantify and compare it? I think the metrics are an active area of discussion but certainly I want to easily go back to a data set and extract the probability of successful reception for a set of center frequencies and transmission bandwidths that my product might use. Best Regards, Stephen Berger TEM Consulting, LP
_______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
