Re: Hello Wideband/UWB Gurus, how to measure "energy samples" from Energy Detector?
First, the energy detector:
I don't know anything about it myself (yet)-- I decided to use this because literature say that it is less complex and thus good for an application in my mind.
Found the following through Google.
(a) For low system complexity and power consumption, we focus on the noncoherent reception of PPM signals, which is akin to GML detection. The symbol decision is based on
finding the pulse position that contains the maximum energy. Source: **broken link removed**
(b) For non-coherent receivers, the data is decoded by comparing the total energy in the two different burst slots. The window size for the energy collection
is adjusted by the transmitted burst duration plus the channel spread. Source: **broken link removed**
See? They both talk about "energy". Back to my question (4), how does one measure energy in this context?
To answer your question what I am trying to do: I want to attach air/inflation pressure sensor in a tire wheel (don't know if it is feasible and people would want to do that, but right now I am concerned with its signal processing aspect), and when low inflation is sensed, some kind of signal (UWB-PPM? or OOK?) will be sent to a receiver. I want to record signals from many (hundreds? thousands?) different position of the wheel as it rotates. Say, rotate a few cm and measure/record signal from there, rotate another few cm, etc. In the training phase, I will also record the tire id. this will be done for each tire of the vehicle. Later in the localization phase, a signal originating from some unknown tire will be compared with the database and the low pressure tire will be identified. That's the idea I am after. Just identifying a "region" (a tire) is sufficient for this kind of problem.
Not only I have to figure out how I measure "energy", I also need to implement the wave propagation part with ray tracing.
Thank you for your interest.
---------- Post added at 01:38 ---------- Previous post was at 01:22 ----------
pstuckey:
Thank you for the comments.
I am taking con-coherent approach (no need to know channel information) to reduce complexity, so may be I should have asked: can I neglect (2) and (3) when two "regions" (=tires) are so closely placed? (5 cm?)
If by neglecting (2) and (3) results in the application's accuracy to less than 5cm (for example), then I will have to consider (2) and (3).
I admit that I am learning as I am asking.. I am sorry and I ask for your and other reader's understanding.
With or without (2) and (3), ray tracing still looks formidable to me.