RTP/RTCP Question Regarding Encryption in DES-CBC Mode - encryption

I have been working to create an RTP/RTCP client and server implementation as a project. I have hit a block when it comes to parsing incoming packets because I am having trouble understanding the encryption method. I have read RFC 3550 several times. Section 9 explains confidentiality and security. I've read about DES-CBC mode here.
Observations I've deduced about encryption
RTP/RTCP packets are encrypted as a unit, meaning ALL bytes are encrypted
Encrypted RTCP packets are prefixed with a 32-bit random number
Encrypted RTP packets are randomized by the timestamp and sequence number
DES-CBC is the default mode
DES-CBC mode requires a 64-bit key
DES-CBC mode requires a 64-bit initialization vector (IV)
DES-CBC has a block size of 64 bits
What I'm confused about:
DES-CBC states that it uses the "Privacy Enhancement for Internet Electronic Mail" (PEM) protocol, but the RTP RFC makes no mention of this. In addition, the encryption diagrams do not include any PEM headers or elements.
Diagram
UDP packet UDP packet
----------------------------- ------------------------------
[random][RR][SDES #CNAME ...] [SR #senderinfo #site1 #site2]
----------------------------- ------------------------------
encrypted not encrypted
So my questions are:
Where does the key come from or what header elements make up the key?
What is the initialization vector for RTP/RTCP?
How do I parse an encrypted packet vs an unencrypted one?
Does RTP/RTCP use the PEM protocol? If so, how?

DES-CBC is a way for encrypting data and it can be used for all different protocols that need it's data to be encrypted.
Where does the key come from or what header elements make up the key?
Encryption keys are generated during connection initialization(at connect time) and can be generated using different ways but generally using Public key cryptography(ex: RSA and DH) and using certificates to prevent MITM attacks which is where the PEM standards are used.
What is the initialization vector for RTP/RTCP?
The IV(initialization vector) is randomly generated number that will be used to ensure that same plaintext will not be encrypted to same ciphertext if using same key (That's why it's Random).
How do I parse an encrypted packet vs an unencrypted one?
For encrypted packets you decrypt first using key(SECRET) you got at connection time then use IV which is sent with data packets(NOT secret) to reverse encryption process. For unencrypted packets you just parse the data since there is no KEY and IV data is just plaintext.
Does RTP/RTCP use the PEM protocol? If so, how?
PEM is not protocol in the sense as network protocols such SSH. This definition I think is accurate enough from wikipedia
Privacy-Enhanced Mail (PEM) is a de facto file format for storing and sending cryptographic keys, certificates, and other data, based on a set of 1993 IETF standards defining "privacy-enhanced mail.
so if your client is using encryption there is a very big chance that it will be using PEM standards for symmetric key initialization.
NOTE: timestamp and sequence number are different from IV these fields are RTP header fields which can serve a similar purpose but they are different from CBC mode IV.

Related

How to send vector in cbc for decipher?

I'm expecting to cipher a string, from a client side (using aes-256-cbc and a Secret-Password already known by both server & client).
And then I'd like to decipher it (on the server side).
I have a code which does work just fine (functionnaly speaking). But it seems I need to send the IV vector from client to server.
I currently did it like this :
00000000000000000000000000000000.JAPQElUAxxxxxxxxxxxjTgJAPQElUAfgydvbY=
So, on the server I split the payload with a "dot" as separator and consider the first item to be the vector.
I believe this is not good to sendthe IV like this.
Would you please help me to understand how should it works?
In CBC encryption mode sending the IV without encrypted is normal and there is nothing to worry as long as the IV is unpredictable. See this question Why is CBC with predictable IV considered insecure against chosen-plaintext attack? from Cryptography.
A common way to send the IV is prepending it to the ciphertext. The libraries, in general, were designed to extract first 16-byte or enable offset in the byte-arrays.
CBC mode of operation is outdated. It has many problems as padding oracled attacks. You should use better alternative Authenticated Encryption (AE) modes as AES-GCM and ChaCha-Poly1305 which does not need padding. Also, in AE you will have confidentiality, integrity, and authentication where CBC only supports confidentiality.
AS modes provide an Authentication Tag that you will need to transmit with the IV. The usual practice; prepend the IV and append the tag to the ciphertext.
IV || Ciphertext||Tag

TLS handshake - Symmetric scheme

From what I can tell TLS works using both symmetric and assymmetric encryption.
The assymmetric schemes are used to exchange keys but when and what symmetric schemes are used?
The asymmetric schemes are used to exchange keys
and digital-signatures.
The symmetric schemes are used to data transfer with the agreed symnetric key during the key-exchange.
This is called Hybrid cryptosystem.
Yes you are right. Asymmetric algorithms are usually slower than the symmetric algorithms. However, symmetric algorithms require a shared secret key to encrypt and decrypt messages. Therefore, TLS allows the client and the server exchange a shared secret key using the asymmetric mechanism. Without an asymmetric algorithm, there is no way the shared secret can be exchanged between the two parties in a secured way. Once both the parties have the shared secret key, all subsequent communication between the client and the server are encrypted using the symmetric algorithm which is much faster than the asymmetric algorithm.
At a very high level, the steps in establishing a TLS connection looks like this:
Client -> Requests for secured session
Server -> Sends certificate & chain certificates
Client -> Verifies certificate
Client -> Generate random key for symmetric encryption
Client -> Encrypts the generated key with the server public key and sends the encrypted value to the server
Server -> Decrypt the client sent key with its own private key
Here onwards all subsequent communications between the server and the client will be encrypted using a symmetric algorithm.
Which specific algorithm will be used is determined by the cipher suites supported by the server and the client. During the connection setup, the cipher suite to be used is determined by the client preference.
A typical cipher suite name looks like this:
TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256
Here
ECDHE - Key exchange algorithm
ECDSA - Digital Signature algorithm used for signing the key
AES_128_GCM - Block cipher and mode with 128 bit key
ECDHE stands for Elliptic Curve Diffie Hellman Ephemeral. The Elliptic variant (the first E) is used for performance, whereas the Ephemeral variant (the last E) is for forward secrecy. Forward secrecy means that if an attacker keeps recording all the communications over TLS and at a later point of time somehow gets hold of the private key, he/she cannot decrypt the past recorded communications.
ECDSA is used for authenticating (verifying the integrity of) the shared secret. ECDSA is weaker and slower than the other authentication algorithms like HMAC. Yet it is used for shared key authentication because it does not need the verifier know the secret key used to create the authentication tag. The server can very well use its private key to verify the integrity of the message.
AES_128_GCM - Once a common secret key is shared between both the parties (usually a browser and a web server), a symmetric block cipher algorithm is used to encrypt the message exchanges between the parties. In this particular case, the block cipher AES with 128 bit key and GCM authentication mode is used.
If you open a HTTPS website in a browser, you can see the cipher suite used using the browser utilities. For e,g, in Firefox you can see the details under the Security tab in the Page Info, as shown below:

how to encrypt tls 1.0 record layer application data?

I am using TLS_RSA_WITH_3DES_EDE_CBC_SHA cipher suite, and I have sucessfully finished the handshake process, so i should have the correct KEYS for the server side and client side, but when i use the server write key and iv to encrypt the data and put under record layer(applicaiton type) send to client, but when i use wireshark ssl debug, i found out that wireshark didn't decrypt the application data correctly, wonder is the application data use a different key to do the encryption and decryption? Thanks
I have finally found the problem, TLS 1.0 used the last block of encrypted data as new IV to encrypt the data. its a little surprised, the encrypted data is visible on the network, so anyone capture that data can find out the IV.

Is is possible to fake a AES encrypted data?

Given a AES secret key, is it possible to fake a data that can be decrypt using that key? the decrypted data doesn't need to be meaningful, just want to know if can fake a encrypted data.
AES is a symmetrical cipher, i.e. the same key is used to both encrypt and decrypt data. As such, the key must be known to both ends of communication, and no one else. Authentication and key-exchange is done via public-key protocols, such as Diffie-Hellman key exchange.
Given these, if you are the third person to possess an AES key, I assume that you cannot send "fake data" to the recipient, as authentication will fail (no negotiation has taken place). If you have established such a connection, you can send whatever you want (you can encrypt gibberish and send it). If you mean to alter an encrypted packet, you can do it without decrypting it and re-encrypting it, but it's most likely to fail integrity tests (e.g. if hashes or CRC are present).

Best practices for encrypting continuous/small UDP data

I am having an application where I have to send several small data per second through the network using UDP. The application needs to send the data in real-time (no waiting). I want to encrypt these data and ensure that what I am doing is as secure as possible.
Since I am using UDP, there is no way to use SSL/TLS, so I have to encrypt each packet alone since the protocol is connectionless/unreliable/unregulated.
Right now, I am using a 128-bit key derived from a passphrase from the user, and AES in CBC mode (PBE using AES-CBC). I decided to use a random salt with the passphrase to derive the 128-bit key (prevent dictionary attack on the passphrase), and of course use IVs (to prevent statistical analysis for packets).
However I am concerned about few things:
Each packet contains small amount of data (like a couple of integer values per packet) which will make the encrypted packets vulnerable to known-plaintext attacks (which will result in making it easier to crack the key). Also, since the encryption key is derived from a passphrase, this will make the key space way smaller (I know the salt will help, but I have to send the salt through the network once and anyone can get it). Given these two things, anyone can sniff and store the sent data, and try to crack the key. Although this process might take some time, once the key is cracked all the stored data will be decrypted, which will be a real problem for my application.
So my question is, what are the best practices for sending/encrypting continuous small data using a connectionless protocol (UDP)?
Is my way the best way to do it? ...flowed? ...Overkill?
Please note that I am not asking for a 100% secure solution, as there is no such thing.
You have several choices. You can use DTLS, which is a version of TLS adapated for datagrams. It is specified in an RFC and implemented in the openssl library. You can also use the IKE/IPsec protocol and use a UDP encapsulation of the IPsec portion. Usually IPsec is available at the OS level. You can also use OpenVPN, which looks to be a hybrid of TLS for key exchange and a proprietary UDP-based packet encryption protocol.
If your problem is that the data is too small, how about extending the data with random bytes? This will make the plaintext much harder to guess.
This question is a little old, but what about using a One Time Pad type approach? You could use a secure reliable transport mechanism (like HTTPS) to transmit the one time keys from the server to your client. There could be two sets of keys -- one for client to sever, and one for server to client. Each datagram would then include a sequence number (used to identify the one time key) and then the encrypted message. Because each key is used for only one datagram, you shouldn't be exposed to the small data problem. That said, I'm not an expert at this stuff, so definitely check this idea out before using it...
Use Ecdh key exchange (use a password to encrypt the client private key; left on the client) instead of a password. This is a very strong key.
Aes cbc does not help you; the messages are too short and you want to prevent replay attacks. Pad your 64 bit message (two integers) with a counter (starting with 0) 64 bits means 2^64 messages can be sent. Encrypt the block twice (aes ecb) and send e(k;m|count)|e(k;e(k;m|count)). Receiver only accepts monotonically increasing counts where the second block is the encryption of the first. These are 32 byte messages that fit fine in a udp packet.
if 2^64 messages is too small; see if your message could be smaller (3 byte integers means the counter can be 80 bits); or go back to step 1 (new private keys for at least one side) once you are close (say 2^64-2^32) to the limit.
You could always generate a fresh pair of IVs and send them alongside the packet.
These days a good streaming cipher is the way to go. ChaCha20 uses AES for a key stream. Block ciphers are the ones that need padding.
Still that's only part of the picture. Don't roll your own crypto. DTLS is probably a mature option. Also consider QUIC which is emerging now for general availability on the web.
Consider using ECIES Stateless Encryption https://cryptopp.com/wiki/Elliptic_Curve_Integrated_Encryption_Scheme where you sending devices use the public key of the central system and an ephemeral key to generate a symmetric key pair, then a KDF, then AES-256-GCM. You end up with modest size packets which are stateless and complete. No need for an out-of-band key agreement protocol.
There are good examples on the internet, for example: https://github.com/insanum/ecies/blob/master/ecies_openssl.c
I am using such a system to deliver telemetry from mobile devices over an unsecure channel.

Resources