Session Initiation Protocol is used to (set up and terminate voice or video calls of course) establish two-party (unicast) or multiparty (multicast) sessions. It also allows modification of existing calls. The modification can involve changing addresses or ports, inviting more participants, and adding or deleting media streams. SIP has also found applications in messaging applications, such as instant messaging, and event subscription and notification.

SIP works in concert with several other protocols to specify the media format and coding like Session Description Protocol, Real-time Transport Protocol and Secure Real-time Transport Protocol.

They are many network elements that use the SIP called SIP user agents. Each user agent performs the function of a user agent client (UAC) when it is requesting a service function, and that of a user agent server (UAS) when responding to a request. Thus, any two SIP endpoints may in principle operate without any intervening SIP infrastructure. However, for network operational reasons, for provisioning public services to users, and for directory services, SIP defines several specific types of network server elements. Each of these service elements also communicates within the client-server model implemented in user agent clients and servers.
User agent
A user agent is a logical network end-point used to create or receive SIP messages. The user agent manages SIP sessions. As a client (UAC), it sends SIP requests, and as a server (UAS) it receives requests and returns a SIP response. Unlike other network protocols that fix the roles of client and server, e.g., in HTTP, in which a web browser only acts as a client, and never as a server, SIP requires both peers to implement both roles. The roles of UAC and UAS only last for the duration of a SIP transaction.[5]
A SIP phone is an IP phone that implements client and server functions of a SIP user agent and provides the traditional call functions of a telephone, such as dial, answer, reject, call hold, and call transfer.[15][16] SIP phones may be implemented as a hardware device or as a softphone. As vendors increasingly implement SIP as a standard telephony platform, the distinction between hardware-based and software-based SIP phones is blurred and SIP elements are implemented in the basic firmware functions of many IP-capable devices.
In SIP, as in HTTP, the user agent may identify itself using a message header field (User-Agent), containing a text description of the software, hardware, or the product name. The user agent field is sent in request messages, which means that the receiving SIP server can evaluate this information to perform device-specific configuration or feature activation. Operators of SIP network elements sometimes store this information in customer account portals,[17] where it can be useful in diagnosing SIP compatibility problems or display of service status.

Proxy server
A proxy server is a network server with UAC and UAS components that functions as an intermediary entity for the purpose of performing requests on behalf of other network elements. A proxy server primarily plays the role of routing, meaning that its job is to ensure that a request is sent to another entity closer to the targeted user. Proxies are also useful for enforcing policy, such as for determining whether a user is allowed to make a call. A proxy interprets, and, if necessary, rewrites specific parts of a request message before forwarding it.

A registrar is a SIP endpoint that provides a location service. It accepts REGISTER requests, recording the address and other parameters from the user agent. For subsequent requests it provides an essential means to locate possible communication peers on the network. The location service links one or more IP addresses to the SIP URI of the registering agent. Multiple user agents may register for the same URI, with the result that all registered user agents receive the calls to the URI.
SIP registrars are logical elements and are often co-located with SIP proxies. To improve network scalability, location services may instead be located with a redirect server.

Redirect server
A redirect server is a user agent server that generates 3xx (redirection) responses to requests it receives, directing the client to contact an alternate set of URIs. A redirect server allows proxy servers to direct SIP session invitations to external domains.
Session border controller
Session border controllers serve as middle boxes between UA and SIP servers for various types of functions, including network topology hiding and assistance in NAT traversal.

Gateways can be used to interconnect a SIP network to other networks, such as the public switched telephone network, which uses different protocols or technologies. Wikipedia

Here is an image who can help to understand the process :

And for those who want to fully understand the process, it looks similar to a DHCP protocol (CSMA/CD etc)


There are two different types of SIP messages: requests and responses. The first line of a request has a method, defining the nature of the request, and a Request-URI, indicating where the request should be sent.[18] The first line of a response has a response code.


Main article: List of SIP request methods
Requests initiate a SIP transaction between two SIP entities for establishing, controlling, and terminating sessions. Critical methods include the following.
  • INVITE: Used to establish a dialog with media exchange between user agents.
  • BYE: Terminates an existing session.
  • REGISTER: The method implements a location service for user agents, which indicate their address information to the server.


Main article: List of SIP response codes
Responses are sent by the user agent server indicating the result of a received request. Several classes of responses are recognized, determined by the numerical range of result codes:[19]
  • 1xx: Provisional responses to requests indicate the request was valid and is being processed.
  • 2xx: 200-level responses indicate a successful completion of the request. As a response to an INVITE, it indicates a call is established.
  • 3xx: This group indicates a redirection is needed for completion of the request. The request has to be completed with a new destination.
  • 4xx: The request contained bad syntax or cannot be fulfilled at the server.
  • 5xx: The server failed to fulfill an apparently valid request.
  • 6xx: This is a global failure, as the request cannot be fulfilled at any server.

Some part I just can't deal with like some of unreliable transport protocols, transactional mechanisms, PRI circuits

So, in fact, SIP looks very interesting and surely the future of business telephone services, using internet is impressive and with the time it will be more and more stable with time.

For now, it's a bit dangerous not because of the security (it's encrypted) but of the stability example:

If we have a telephone and a PC in a house.

The PC is connected to the internet.

At the same time the telephone ring! And the incoming call is taken.

The call will be perfect but the PC will have a lot of packages loss (the call will have the priority and it take a lot)

Thank you for your attention it takes few hours and it's my first article !
If something isn't clear let me know I will try to explain it as soon as possible.