eitcom
Seeking help with current stimulating/voltage measuring patterns step = 7 (or [0 7]) for 16 electrodes model
Hello, I am currently looking into a new stimulating/measuring patterns, back then, the adjacent method '{ad}' or [0 1] in EIDORS or dist_exc=1, step_meas=1 in pyEIT worked well.
I am trying to implement a [0 7] patterns for both stimulating and measuring patterns, and I believe it should be protocol_obj = protocol.create(n_el, dist_exc=7, step_meas=7, parser_meas="rotate_meas") in pyEIT.
For a 16 electrode models, with electrodes numbering from 0 to 15, the patterns should follow:
Stimulating 0 7
Measuring ~~0 7~~ | 1 8 | 2 9 | 3 10 | 4 11 | 5 12 | 6 13 | ~~7 14~~ | 8 15 | ~~9 0~~ | 10 1 | 11 2 | 12 3 | 13 4 | 14 5 | 15 6
(All measuring that include electrodes which are being used for current injecting will be ruled out)
Next, the stimulating electrodes step up: Stimulating 1 8 Measuring 0 7 | ~~1 8~~ | 2 9 | 3 10 | 4 11 | 5 12 | 6 13 | 7 14 | ~~8 15~~ | 9 0 | ~~10 1~~ | 11 2 | 12 3 | 13 4 | 14 5 | 15 6
And go on until the stimulating electrodes reach: Stimulating 15 6 Measuring 0 7 | 1 8 | 2 9 | 3 10 | 4 11 | 5 12 | ~~6 13~~ | 7 14 | ~~8 15~~ | 9 0 | 10 1 | 11 2 | 12 3 | 13 4 | 14 5 | ~~15 6~~
Result in a 208 values (16 rows and 13 columns) matrix.
The corresponding meas_sel matrix in EIDORS :
1 0 1 1 1 1 1 1 0 1 0 1 1 1 1 1
1 1 0 1 1 1 1 1 1 0 1 0 1 1 1 1
1 1 1 0 1 1 1 1 1 1 0 1 0 1 1 1
1 1 1 1 0 1 1 1 1 1 1 0 1 0 1 1
1 1 1 1 1 0 1 1 1 1 1 1 0 1 0 1
1 1 1 1 1 1 0 1 1 1 1 1 1 0 1 0
0 1 1 1 1 1 1 0 1 1 1 1 1 1 0 1
1 0 1 1 1 1 1 1 0 1 1 1 1 1 1 0
0 1 0 1 1 1 1 1 1 0 1 1 1 1 1 1
1 0 1 0 1 1 1 1 1 1 0 1 1 1 1 1
1 1 0 1 0 1 1 1 1 1 1 0 1 1 1 1
1 1 1 0 1 0 1 1 1 1 1 1 0 1 1 1
1 1 1 1 0 1 0 1 1 1 1 1 1 0 1 1
1 1 1 1 1 0 1 0 1 1 1 1 1 1 0 1
1 1 1 1 1 1 0 1 0 1 1 1 1 1 1 0
I believe this is the behavior of [0 7] stimulating/measuring patterns, I performed some programming on my hardware so it will follow this pattern. However, the result is just wrong, and I need to rule out the possibilities to see what is going wrong, including hardware fault and software fault.
This is my [0 7] sample data (the [0 1] patterns worked so it should not be any big issue with my circuit I believe):
4.85 2.13 2.95 0.80 3.57 6.55 5.56 5.99 3.56 0.09 2.38 6.54 4.94
6.29 6.42 4.02 1.20 3.33 6.55 5.56 6.42 4.01 0.68 2.15 6.07 5.17
1.88 1.62 2.38 1.82 2.38 2.07 5.73 2.74 2.44 1.71 2.49 2.17 6.11
5.56 5.58 3.87 1.02 3.22 6.55 5.35 6.54 4.27 0.76 2.43 6.07 2.32
5.56 5.43 3.67 1.33 3.30 6.55 5.23 6.55 3.74 0.38 2.77 1.47 5.35
5.56 5.60 3.97 1.16 3.75 6.55 5.35 6.55 4.08 0.46 1.37 6.55 5.35
5.56 5.62 3.41 0.81 3.94 6.55 5.35 6.55 3.21 1.08 3.12 6.45 5.35
5.56 5.90 2.58 0.53 4.10 6.49 5.24 6.55 1.60 0.02 2.69 6.36 5.40
5.57 5.74 2.67 0.76 3.56 6.24 5.27 1.76 3.41 0.47 2.96 6.43 4.95
5.53 6.52 2.74 0.43 3.36 6.11 5.24 6.39 4.05 0.61 2.98 6.22 5.40
5.35 6.55 3.29 0.54 3.28 6.05 5.83 6.55 3.94 0.81 2.46 6.25 5.06
5.36 6.55 3.09 0.35 3.35 6.24 5.56 6.38 4.04 1.15 2.81 6.55 5.56
5.55 5.64 3.50 0.83 3.65 3.22 5.56 6.55 4.72 1.30 2.66 6.55 5.56
5.55 5.47 3.37 1.63 1.77 6.55 5.56 6.55 4.38 1.01 2.98 6.55 5.56
5.56 5.42 3.15 2.28 4.43 6.26 5.55 6.55 4.17 0.60 3.44 6.55 5.56
5.86 5.20 1.59 0.38 4.16 5.54 5.55 6.55 3.45 0.02 3.35 5.95 5.86
difference matrix:
4.97 2.13 3.44 0.85 3.97 6.55 5.24 6.14 4.13 0.10 2.93 6.55 5.09
5.94 6.55 4.47 1.06 3.53 6.55 5.56 6.39 4.44 0.81 2.76 6.52 4.87
1.89 1.65 2.54 2.01 2.54 2.24 5.22 2.77 2.89 2.22 2.95 2.51 6.20
5.56 5.62 4.41 1.20 3.67 6.55 5.56 6.31 5.21 1.08 2.78 6.55 1.95
5.56 5.51 4.28 1.54 3.97 5.93 5.05 6.37 4.80 0.90 3.44 2.21 5.56
5.56 5.58 4.45 1.18 4.37 6.55 5.56 6.55 5.02 1.05 1.24 6.55 5.56
5.56 5.51 4.12 1.35 4.50 6.55 5.56 6.55 4.41 1.76 3.84 6.30 5.36
5.56 5.63 3.87 0.65 4.94 6.32 5.23 6.55 1.19 0.07 4.22 5.77 5.56
5.56 6.14 3.12 0.79 5.03 5.56 5.23 1.41 3.64 0.91 4.51 5.62 5.56
5.56 6.55 3.40 0.64 4.69 5.55 5.23 6.55 4.04 1.32 4.40 5.56 5.56
5.56 6.55 3.50 1.37 4.26 5.50 5.77 6.48 3.89 1.64 3.80 5.56 5.56
5.56 6.55 3.55 1.06 4.74 5.60 5.56 6.55 3.90 1.77 4.10 5.62 5.36
5.57 6.36 3.51 0.62 4.63 2.29 5.54 6.55 4.22 1.47 3.70 6.35 5.56
5.36 6.06 3.54 1.08 2.29 6.55 5.36 6.55 4.09 0.47 4.37 6.55 5.36
5.35 5.88 3.56 2.17 4.85 6.55 5.24 6.55 4.03 0.31 4.13 6.55 5.36
5.56 5.64 1.77 0.43 4.12 5.98 5.24 6.55 3.18 0.08 4.01 6.50 5.56
Those value is questionably high, but since we are measuring two faraway electrodes, the voltage drop between them might me higher than usual. And since we are not doing adjacent measuring, I guess it won't neatly follow the parabol shape rule. But I still have to state the fact that this data is not working so these reasoning are a little bit cheap. There also may be a chance that my understanding about this pattern is wrong in the first place, would you mind enlightening me about this?
Thank you!
Hello, while waiting for support, I took a look around the issues tab, and found out that the direction of current flowing in and out is specified. The ex_mat for step=7 is
[[ 0 7]
[ 1 8]
[ 2 9]
[ 3 10]
[ 4 11]
[ 5 12]
[ 6 13]
[ 7 14]
[ 8 15]
[ 9 0]
[10 1]
[11 2]
[12 3]
[13 4]
[14 5]
[15 6]]
The first element represents the electrode that current will flow to enter the object, while the second element represents the electrode that current will flow through to get out of the object and return to the ground.
My hardware is currently fixing the electrodes for injecting current into the object are the odd ones, and the even electrodes are used for leaving the object. while the correct patterns should be even-odd repeating (observing the ex_mat matrix).
When current enters an electrode, the voltage around that section should be high, while the voltage near the electrodes where current leaves will be low due to loss of energy while traveling through the solutions inside the object. Reversing this rule will cause the data to be wrong compared to the reconstructing technique.
This is my guess; I appreciate any confirmation and clarification. Thank you!