Switching is the method that is used to establish connections between nodes within a network. Once a connection has been made, information can be sent. Telephone switching usually refers to the switching of voice channels.
There are a number of different types of switches including local switches, tandem switches and transit switches and any number of them can play a part in creating a connection.
Local switch: This provides switching for a specific area. Subscriber loops connect to the local switch in that area.
Tandem switch: This is used to interconnect switches at various sites within the network.
Transit switch: This is very similar to a tandem switch, except it is used for long-distance connections.
In order to carry out functions, switches need computing power. In the past this was a manual process which involved a human operator who would physically switch a call. In modern switching systems, integrated software has a wide-range of capabilities and services they can provide without the need for human interaction.
Conceptually, switches have inlets and outlets. Incoming calls are served through the inlets and outgoing calls through the outlets. When a call is received by a subscriber it enters the exchange through the inlet and connects to the call through the outlet.
Switches have three main requirements: connecting incoming calls to an array of outgoing circuits, disconnect (clear) the call after termination and prevent calls from intruding on to circuits that are already in use. Other functions may include: interconnection, control, alerts, attending, busy testing and supervising.
There are three main switching techniques: circuit switching, message switching and packet switching.
Circuit switching allows data transmission when there is a complete circuit (route or path) established between the source node and the destination node.
There are three steps required in order to establish a connection with circuit switching: setup of link, hold-up of link and release of link.
Message switching is when individual messages are separately switched at each node along its route or path, from source to destination. Circuits aren’t established exclusively for messages, instead, messages are sent with a store-and-forward method.
Message switching was the precursor to packet switching, where messages were routed in their entirety. Nowadays, message switching is often implemented over packet-switched or circuit-switched data networks.
Each message, which contains addressing information, is treated as a separate entity. At each switch the information within the message is read and the transfer path is determined.
A conversation of several messages might not always be transferred over the same path, but this is dependant on network conditions. Each message is stored (often on hard drives) before they are transmitted to the next switch. This is where the term ‘store-and-forward’ comes originates.
An example of this system is email messaging.
Packet switching is a digital networking communication method which groups all transmitted data into blocks, which are called packets, and transmitted via a medium which may be shared by multiple communication sessions.
Packets are composed of a header and payload. The header contains information that is used by the network hardware to direct the packet to its destination. The payload is extracted at the destination and used by the application software.
In order to determine the best design for a telephone switching system, a number of criteria must be determined and considered by the operator.
Perhaps the most important factor, traffic intensity of the busy hour is, simply, the calling rate + (plus) the average holding time during the 60-minute period that the traffic intensity is at its highest.
This is the average number of requests for connection per unit of time.
This is the mean amount of time that a call lasts.
Once these general properties of the traffic are established, the performance of a switching system can be stated and specified by a grade of service (GOS).
GOS is a way of measuring congestion expressed as the probability that a call will be blocked or delayed.
In order to build, maintain and improve a switch that will supply the highest quality of service to its subscribers, network operators must monitor their network hardware constantly and efficiently and be ready to repair, replace or add any parts that are required.
The ability to anticipate network downtime before it occurs and act to source parts to stop it from happening is an essential task that sets the best network providers apart from their competitors.
At Carritech, we actively source and supply, as well as repair parts for switching systems all around the world. Our customers rely on our industry knowledge, particularly in the area of hard-to-find products.
If you operate a network that relies on switching systems to serve your customers or staff, click here to find out more about our services and how we can partner with you to ensure your network is the best it possibly can be.