Previous | Table of Contents | Next |
Step-by-step (Strowger switching), crossbar, and reed relay switching are called space division circuit switching, or just circuit switching, because every single telephone call is assigned a unique physical path, or connection, through the telephone system, from source to destination (end-to-end). Circuit switching is also commonly referred to as connection switching.
Figure 3-12 depicts a simple mechanical switching scheme where the Input can be switched to Output A, Output B, Output C, or Output D, depending upon the desires of the circuit operator. The diagram could represent any one of the mechanical switching types just discussed. Either an operator can physically place a cord in a jack and complete the circuit or a mechanical switch can be positioned electrically using a coded signal from the source telephone, as the Strowger switch.
Figure 3-13 shows a logical diagram of a digital switching element. As you may readily tell, any input, 0 or 1, may be switched to any output, 0 or 1. Digital switching matrices are composed of many of these types of elements.
Figure 3-14 shows a four-stage (A, B, C, D) Delta-2 Switching Matrix. The switching matrix is composed of 32 basic digital switching elements, labeled A1-A8, B1-B8, C1-C8, and D1-D8. There are 16 inputs and 16 outputs. Any input can be switched to any output. In the real world, fast 32-bit microprocessors, under program control, switch the matrices. As a packet/cell comes into the switch, the header information is read and decoded while the packet/cell is held in a buffer (waiting room). After the microprocessor decodes the header (address) information and determines that a correct switch path is available, the cell/packet is moved from the buffer into the switching matrix, to follow the path from input to output through the individual switch elements.
TDM multiplexing of channels gave us a new way to switch signals from source to destination. It is commonly called connectionless switching. Every channel in a TDM multiplexed signal is uniquely identified by its very position in the bit stream. All a particular device must do is read the bit stream during the time period of interest. The device reading the bit stream must know when to start reading and when to stop. A method of identifying the proper location that did not depend entirely upon some reference clock signal was used with PDH multiplexing. With synchronous systems, such as SONET/SDH, devices and bit streams must be operating from or have some relationship to the same fundamental timing source. Granny Bell provided that single timing source in the nationwide distribution of a master clock reference signal located at Hillsboro, Missouri. Connectionless switching development progressed from plesciosynchronous multiplexing used in PDH for conversion to a higher bit rate to synchronous multiplexing used in SONET/SDH.
Circuit switching was the original type of transfer mechanism used. A dedicated line between source and destination is allocated to the parties involved in the transfer of data over the network path. Channel capacity must be available and reserved to make the connection between each pair of nodes on the path, and each node must have a switching capability. For different applications, utilization of the line may vary enormously from no load to overload. However, there is little signal delay and effective transparency for the user. Effective transparency means the transmission path from source to destination appears to the source and destination as one homogeneous highway with no speed bumps, traffic lights, or left turns. Circuit switching is very efficient for constant bit rate (CBR) applications.
Previous | Table of Contents | Next |