« first day    last day (18 days later) » 

12:37 AM
i don;t know if you're still here but is this correct for binary XOR function?

j-001 11111101 00000111

arg1 11111101
arg2 00000111
result 11111010
 
OP here; hi sorry I'm late to the party, I'm still trying to figure out SE. Thanks for your efforts thus far. The barcodes belong to cartridges from a dispensing machine. The cartridges can be placed in any position in the machine which reads the barcode to determine what the contents of the cartridge is.
 
12, are you in? can you rip/dump data/firmware from this thing?
so the catridge read the product code?
*cartridge
 
Unfortunately I don't have access to a device that will produce any useful output. The best I have is a machine that will beep if a valid barcode is detected.
 
okay so you have to j-020 squared away?
 
Sorry?
 
12:43 AM
as per the diagram you have up to j-020 in the series reading correctly/valid?
 
I have a sample set of barcodes up to J-140, I need to figure out the pattern so I can create more up to J-999 if possible.
I know the machine will definitely read barcodes up to J-250 (however I don't have a sample of them to use / duplicate).
The sheet of barcodes I posted all read correctly (up to J-020) along with all of my other samples up to J-140.
 
ok but as of yet no identifiable pattern
i have to ask again what barcode is actually being read? product/inventory? keypad/codes?
 
The cartridges dispense small items (e.g. candy). I have 120 refillable cartridges total. The machine can fit only 50 cartridges at a time. I can put the cartridges in any position in the machine and the machine detects which cartridge is where by reading its barcode. Using the software that controls the machine I can tell it dispense 5 items from cartridge J-020 and it will dispense from the correct cartridge because of the barcode.
 
Is it possible for you to post the rest of the codes that you have?
Even one or two more sheets would be helpful.
 
okay got it, no tie-in to product, do you have or can you get diagrams to j-999?
 
12:58 AM
@Moonbutt He said he only has up to 140.
 
alternately what manufacturers are involved in issue of this tech
2012rcampion can you scroll up and tell me if i did binary XOR function correctly?
 
Yes, you correctly calculated result = arg1 (xor) arg2
 
sweet! okay 2012rcampion, i'm trying a search for binary sequence analysis tool, my question is if i found a tool for decimal sequences and was able to get a result and convert back to binary do you think it would be accurate?
12, looking at the image now
 
@user1224615 Thanks!
@moonbutt If it works then it works. There's no "right" method for analysis.
That said, I'm pretty sure you need to analyze on a bit level for this one.
 
1:10 AM
okay, define bit level, is that like that bitwise thing i was reading earlier?
 
Basically I mean that it's likely that the actual hardware is manipulating the individual bits of the barcode; it's not manipulating a higher-level representation of those bits.
 
okay, so would architecture be a factor? i'm looking at BitBlaze which is probably off the mark but meh.
 
That product is for analyzing binary executables (programs), not number sequences. Basically going from a set of low-level instructions like "copy this number to this register" or "multiply these two number" into high-level instructions like "parse this JSON file"
 
1:29 AM
ahh okay dammit
 
2:29 AM
@user1224615 I've input all the codes you sent me, and it looks like J-093 and J-101 don't follow the same pattern as the rest (the center six bits are all set, indicating a number of 129 or 130). If they both work then the center bits could be an elaborate red herring to foil exactly this type of reverse-engineering attempt.
 
3:15 AM
I would be surprised if there is any form of counter-reverse-engineering measures in place. The only other potentially useful information I can provide is that it is a Japanese manufacturer so could well reference Japanese characters (UTF-8 etc). The reader PCB also contains two 8-bit Shift Registers.
 
3:47 AM
I highly doubt UTF-8 is involved since both Hiragana and Kanji characters are each three bytes long in UTF-8. Shift JIS or one of the other encodings used in QR codes might be possible, but I highly doubt it considering the structure of the codes that I've observed so far.
The shift registers are just used for reading the sensors into the microcontroller/processor; I find it unlikely that there's any hardware processing going on.
 

« first day    last day (18 days later) »