P25 Link Level Encryption

[krokus]

It appears that the P25 standard has encrypted control channels proposed.

Updates on P25 Security, Wireline Interfaces Among January Work


Sent via Tapatalk

 

[mikewazowski]

Thread moved.

 

[ScannerSK]

This is not good at all...

If the control channels become encrypted then properly monitoring/following conversations of even non-encrypted talkgroups will be impossible for all scanners.

I assume encrypted control channels will make P25 systems even worse in areas with less than ideal coverage. However, my concern is loosing the ability to properly follow non-encrypted conversations on our scanners.

Shawn

 

[balibago]

Will this effect all the old P25 equipment or just new manufacture? Will this effect the non encrypted systems? I'm looking at a home patrol 2 and would like to get at least 3 years out of it. This will be a horrible development for those who monitor mundane things like public works, utilities etc. Since when have we become the enemy?

 

[trumpetman]

This is not good at all...

If the control channels become encrypted then properly monitoring/following conversations of even non-encrypted talkgroups will be impossible for all scanners.

I assume encrypted control channels will make P25 systems even worse in areas with less than ideal coverage. However, my concern is loosing the ability to properly follow non-encrypted conversations on our scanners.

Shawn

Encryption on P25 doesn't decrease range like it does in the old analog SecureNet days, so that's a moot point. Plus I imagine if a system is using an encrypted control channel there won't be any non-encrypted talkgroups in use, but that's just my two cents.

I seem to recall a while back when looking at the documentation there was a flag/packet that indicated the data stream was encrypted but was unable to find any literature on encrypted the control channel...I guess that was the precursor to this new ability.

 

[krokus]

Thread moved.

I had posted in scanning, so the impact on scanning could be discussed. I should have explained that.

Sent via Tapatalk

 

[krokus]

This is not good at all...

If the control channels become encrypted then properly monitoring/following conversations of even non-encrypted talkgroups will be impossible for all scanners.

I assume encrypted control channels will make P25 systems even worse in areas with less than ideal coverage. However, my concern is loosing the ability to properly follow non-encrypted conversations on our scanners.

Shawn

The only way that comes to mind, for use at that point, would be to scan the discrete repeater outputs. Assuming the encryption would be implemented after going TDMA, I wonder how well the scanners will process the time slots.

Sent via Tapatalk

 

[balibago]

I assume that only new equipment will have this capability. Louisiana at this time can barely pay their electric bill.

 

[RayAir]

I don't do much P25 monitoring but have read some security papers regarding it. I believe there already exists a way to scramble TG and RID numbers. So they would just be adding encryption to the CC.
If the voice is already well encrypted, encrypting the CC too wouldn't add much more security.

Sure, by not being able to read TG and radio ID's you wouldn't be able to do traffic analysis, but that is limited anyway. And if the CC encryption is being done at the repeater, couldn't you just scan the repeater inbound for TG and RID info?

Anyway, the local FBI shop is P25 AES 256 voice encrypted 24/7 and even they dont scramble TG's or RIDs.

Even if they did, not really a big deal. Can't hear wha'ts going on anyway.

I don't see this becoming a widespread issue. A few new systems "might" pay extra for it.

 

[slicerwizard]

I don't do much P25 monitoring but have read some security papers regarding it. I believe there already exists a way to scramble TG and RID numbers.

Pretty sure that isn't available yet.

So they would just be adding encryption to the CC.
If the voice is already well encrypted, encrypting the CC too wouldn't add much more security.

Sure, by not being able to read TG and radio ID's you wouldn't be able to do traffic analysis, but that is limited anyway.

It would let one see when a canine officer switched to a given dispatch channel (manhunt in progress) or when the ETF did (barricaded suspect) or when those special event groups start loading up with radio affiliations at 4 am (drug raids at 5 am); these are all very useful things to know when you're the overnight ENG.

I'm sure others can find uses.

And if the CC encryption is being done at the repeater, couldn't you just scan the repeater inbound for TG and RID info?

If the outbound CC is encrypted, why wouldn't the inbound also be?

Anyway, the local FBI shop is P25 AES 256 voice encrypted 24/7 and even they dont scramble TG's or RIDs.

Pretty sure that isn't available yet.


What additional Security Interfaces and Services are being worked on in TIA/P25?

Key Fill Interface to the KMF, Authentication Facility and between Key Fill devices.

Link Layer Encryption for protection of Control signaling and group/individual IDs on the trunking control channel, trunked and conventional voice channels and trunked and conventional data channels.

 

[RayAir]

Read (attached) 3.2 about P25 metadata. They mention some of the identifier fields can already be optionally encrypted?

 

[slicerwizard]

Link Control Words have a protected bit - as do control channel TSBKs. Those bits are there to support signalling encryption when it arrives. It hasn't arrived yet, has it?

 

[RayAir]

I saw an algorithm for P25 called AES-GCM.

From what I could find it seems to be a 128 bit key used as an authentication to a P25 system.

Is this the cipher they might use for CC encryption?

Do you happen to know about AES-GCM in regards to P25?

Thanks.

 

[kv5e]

I saw an algorithm for P25 called AES-GCM.

From what I could find it seems to be a 128 bit key used as an authentication to a P25 system.

Is this the cipher they might use for CC encryption?

Do you happen to know about AES-GCM in regards to P25?

Thanks.

https://en.wikipedia.org/wiki/Galois/Counter_Mode

This method is used for packet data and authentication of sending party.

Embedded authentication to thwart MITM attacks for packet data is the reason for GCM.

Control plane encryption will likely use a symmetric key method and could have some attributes of Galois Counter Mode.

Due to the amount of traffic on a CC, keys would be changed very frequently on the control plane encryption, perhaps daily.