Method of sequencing signal processing in the communication mode of a digital cellular radio system

Claim 1

Method of sequencing signal processing in a digital cellular radio system having a communication mode for conveying digitized speech over a communication channel between a terminal and a base station, said radio system using multiframes each comprising a sequence of frames, at least some of said frames being designated as traffic frames, said digitized speech comprising coded blocks each representing a single group of speech samples, each coded block being formed of 2·s sub-blocks each conveyed during a different respective traffic frame, a pth sub-block of a coded block and a (p+s)th sub-block of a preceding coded block being combined in a pth packet, each said multiframe comprising a number of traffic frames which is a multiple k of the number of sub-blocks 2·s, any traffic frame being identified within its respective said multiframe by the expression s·j+p, each of said traffic frames comprising at least two disjoint time slots including a first time slot for a send packet and a second time slot for a receive packet, said method comprising the steps: during each of 2·s·k traffic frames s·j+p of a current said multiframe: performing a configuration process, processing a pth receive packet in a respective receive time slot by recovering the pth receive sub-block of a (j+n) th coded receive block and the (p+s)th receive sub-block of a (j+n-1)th coded receive block, and processing a pth send packet in a respective send time slot by combining a pth send sub-block of a (j+n)th coded send block and a (p+s)th send sub-block of a (j+n-1)th coded send block; during each of 2·k send mode traffic frames s·j+a of a first subset of said 2·s·k traffic frames: shaping a next (j+n+1)th coded send block from a group of speech samples to be sent; and during each of 2·k receive mode traffic frames s·j+b of a second subset of said 2·s·k traffic frames: recovering a last (j+n-1)th coded receive block and recovering any corresponding speech samples contained therein, where s is a positive integer representative of the number of sub-blocks in a coded block divided by 2, k is a positive integer representative of the number of coded blocks in a multiframe, j is a first integer variable 0≦j≦2k representative of the position of a coded block within a multiframe, n is a second integer variable representative of the rank of the coded send-block whose first sub-block is sent on the first frame of said current multiframe, p is a third integer variable 0≦p≦s representative of the position of a sub-block within a coded block, a is first predetermined value of p representative of the position of a send frame within a coded block, and b is a second predetermined value of p representative of the position of a receive frame within a coded block.