- This topic has 5 replies, 1 voice, and was last updated 13 years, 2 months ago by saman.
19th December 2007 at 11:36 #50078riverstreamerGuest
What I bring is an assumption on an idea which seems right, but I don’t have no document to rely on it. So please read my notes and approve or rject it with your reasons and guide me. If you have document please send to my mail or give link in here. my mail is:
As you know we have two subject on every data or voice channel. 1. Encoding 2. Encryption
Surely Encoding and it’s method is for increasing and using bandwidth on each channel.
BUT for Encrption, surely each encryption process increase datum volume in each pack and so because of limitation total data size, the rate of sampling decrease to some entent. In addition on bit error rate, the number of faults increases after encrytion, because for each fault, amount of net data/voice real datum which can be decryted will be decreased. SO:
I think if encrytion on a channels 1.the rate od datum in Kb/s will decrease 2. C/I will be get lower value in test 3. the voice quality will be damaged to some extent.
Please give me your idea and guide me on this theory of you have any idea by yourself or on document.
PS: for enabling or disabling encrytion on channels, you can assume TETRA, GSM or other neyworks.
Thanks you and waiting for your replies.
riverstreamer [AT] gmail [DOT] com19th December 2007 at 20:49 #50079pixGuest
i could check tomorrow, but i’m pretty sure that the encryption comes beore the encoding.
So the encoded frames contains encrypted data. After decoding, either the frame is valid or not. If it is, then it means most important bits are fully recovered, therefore the data can be decrypted successfully.20th December 2007 at 06:59 #50080riverstreamerGuest
Thanks for your reply. your answer showed me two important subject. I’m waiting for your more checks and guides.
excuse me, I wrote my email wrong. My Email is:
If it was possible please contact me on Email too. (For other subjects, you are great)
Please bring your comments too.22nd December 2007 at 09:14 #50081YousefGuest
1. Althought data encrypts before encoding as PIX said, but there is no different in amount of datum, because in anyway, encryption increases the amount of a net datum to bigger size and this catch more capcity of bandwidth so may effect on quality in distance.
2. it is right that corrupted datum will be retransmitted and real datum will be recovered but this subject get more capacity in use too.
Maybe other friends can help more.22nd December 2007 at 22:19 #50082pixGuest
Just one thing : corrupted data (datum, really ???) get retransmitted only if we’re in EGPRS. For voice call, there is absolutely no retransmission.23rd December 2007 at 08:17 #50083samanGuest
Accroding to docs from GSM fundamentals and practical behavior which I have read, there are two types of “ERROR Correction Coding”:
Backward error correction & Forward Error Correction
-Backward Error Correction: if check bits are correct, it’s assumed all the information are correct, but if check bits have problem, then AUTOMATIC REPEAT REQUEST enabled -> It’s more used for dat and not suitable ofr speech
-Forward ERRO Correction:
it’s in two types and in GSM used a combination of them: *Block codes *convolutional codes