Space/time switch architecture
View Patent ↗A switching core in accordance with the principles of the present invention connects N input ports to M output ports and provides multi-cast capability. The switching core includes a plurality of selection blocks, each of which receives multiple input bit packs organized in a combination of input data rails and time slots. The switching core selects one of the input bit packs from one of the rails in one of the time slots and conveys the selected bit pack to an output data position within a combination of output data rails and time slots. This operation may take place both in parallel and in sequence.
1. Apparatus for switching data from any of a plurality of inputs to any of a plurality of outputs, comprising:
at least one apparatus for receiving a plurality of input bit packs organized in a combination of input data rails and time slots and for storing said received bit packs in matrix form including a storage position for each rail and time slot combination;
at least one apparatus for selecting any of the input bit packs from any of the rails in any of the time slots of said matrix;
at least one apparatus for conveying said selected bit pack to any output data position within a first output bit map including a combination of output data rails and time slots; and
at least one apparatus for loading a second output bit map including a combination of output data rails and time slots from said first output bit map, said second output bit map configurable for being loaded in parallel with said first output bit map.
2. Apparatus of claim 1 , wherein each bit pack is one bit wide.
3. Apparatus of claim 1 , wherein a plurality of input bit packs are selected for output in a plurality of output data positions.
4. Apparatus of claim 1 , wherein a single bit pack is selected for output in a plurality of output positions.
5. Apparatus for switching data from any of N input positions arranged as T time slots on R rails to any of M output positions arranged as T 2 time slots on R 2 rails, comprising:
at least one apparatus for receiving input data arranged as bit packs in T time slots on R rails and for storing said received bit packs in matrix form including a storage position for each rail and time slot combination,
at least one apparatus for selecting any of the respective bit packs from any of the R rails and latching the selected bit packs during a predetermined time slot to thereby select a bit pack of predetermined R and T values;
at least one apparatus for conveying said selected bit pack to any output position in a first output bit map of predetermined R 2 and T 2 values; and
at least one apparatus for loading a second output bitmap of predetermined R 2 and T 2 values from said first output bit map, said second output bit map configurable for being loaded in parallel with said first output bit map.
6. Apparatus for switching data from any of N input positions arranged as T time slots on R rails to any of M output positions arranged as T 2 time slots on R 2 rails, comprising:
M selection blocks, each configured to select a bit pack for a different one of the output positions, each of the M selection blocks including:
an apparatus for receiving input data arranged as bit packs in T time slots on R rails and for storing said received bit packs in matrix form including a storage position for each rail and time slot combination;
an apparatus for selecting data from any of the R rails and latching the selected data during a predetermined time slot to thereby select a bit pack of predetermined R and T values;
an apparatus for conveying said selected bit pack to any output position of predetermined T 2 and R 2 values in a first T 2 ×R 2 output bit map, said first T 2 ×R 2 output bit map configured for receiving a selected bit pack in each location from a different one of the M selection blocks; and
an apparatus for loading a second T 2 ×R 2 output bit map from said first T 2 ×R 2 output bit map, said second T 2 ×R 2 output bit map configurable for being loaded in parallel from said first T 2 ×R 2 output bit map.
7. Apparatus of claim 6 , further comprising:
an apparatus for arranging input bit packs as an array of T time slots on R rails and conveying output bit packs from the second T 2 ×R 2 output bit map on R 2 rails in T 2 time slots.
8. Apparatus of claim 7 wherein N=M=768.
9. Apparatus for switching data from any of N input positions arranged as T time slots on R rails to any of M output positions arranged as T 2 time slots on R 2 rails, comprising:
R 2 selection blocks, each configured to select a bit pack for a different one of the output positions, each of the R 2 selection blocks including:
an apparatus for receiving input data arranged as bit packs on N rails, said apparatus for receiving input data comprising:
a first T×R input bit map configured for receiving a selected bit pack in each location from a different one of the N space/time input positions; and
a second T×R input bit map configured to be loaded in parallel from the first T×R input bit man and to convey N input bit packs to each of the R 2 selection blocks and to hold the N input bit packs available to the R 2 selection blocks during T 2 time slots;
apparatus for selecting data from said second T×R input bit map and for storing said selected bit packs in matrix form including a storage position for each rail and time slot combination; and
an apparatus for conveying said selected bit pack to any output position of predetermined T 2 and R 2 values.
10. Apparatus of claim 9 wherein N=M=768.
11. A method of switching data from any of N input positions arranged as T time slots on R rails and stored in matrix form including a storage position for each time slot and rail combination to any of M output positions arranged as T 2 time slots on R 2 rails, comprising the steps of:
(a) in each of R 2 selection blocks, selecting a bit pack from any of the storage positions of said matrix for a different one of the output positions;
(b) conveying each of the bit packs selected in step (a) to any output position in a first output bit map of predetermined R 2 and T 2 values; and
(c) loading a second output bitmap of predetermined R 2 and T 2 values, said second output bit map loaded in parallel with said conveying each of the selected bit pack to any output position in said first output bit map.
12. The method of claim 11 wherein step (a) comprises the further step of:
receiving input data arranged as bit packs on N rails.