Addressing strategy for Viterbi metric computation
There is provided an addressing architecture for parallel processing of recursive data. A basic idea is to store a calculated new path metric at the memory location used by the old path metric, which old metric was employed to calculate the new metric. If m metric values are read and m metric values are simultaneously calculated in parallel, it is possible to store the new, calculated metrics in the memory position where the old metrics were held. This is advantageous, since the size of the storage area for the path metrics is reduced to half compared to the storage area employed in prior art Viterbi decoders for the same performance with regard to path metric computations.
1. An addressing architecture for parallel processing of recursive data, which architecture comprises:
memory arranged with a first storage area for storing a first set of path metrics and a second storage area for storing a second set of path metrics, which first storage area is coupled to a first memory input and a first memory output, and which second storage area is coupled to a second memory input and
a second memory output;
a selector arranged to connect the first memory output or the second memory output to a first selector output, and the first memory output or the second memory output to a second selector output; and
a Viterbi butterfly structure having a first set of inputs connected to the first selector output, a second set of inputs connected to the second selector output, a first set of outputs connected to the first memory input and a second set of outputs connected to the second memory input,
wherein the selector is configured to control the Viterbi butterfly structure such that new path metrics are stored in desired respective storage areas.
2. The addressing architecture according to claim 1 , wherein the selector is arranged with a control input via which the selector is configured to control the Viterbi butterfly structure such that new path metrics are stored in the desired respective storage areas.
3. The addressing architecture according to claim 1 , wherein the storage areas of the memory further comprise respective read enable inputs that are separated from each other and respective write enable inputs that are separated from each other.
4. The addressing architecture according to claim 3 , wherein a memory write address for a new path metric is set to be equal to a corresponding memory read address for an old path metric that was employed to compute said new path metric.
5. The addressing architecture according to claim 2 , wherein the selector is manipulated by applying, via said control input, parity bits based on a current memory address.
6. The addressing architecture according to claim 5 , wherein the connection of the first or second memory output to the first selector output is controlled by a first parity bit based on a current memory address of the first storage area, and the connection of the first or second memory output to the second selector output is controlled by a second parity bit based on a current memory address of the second storage area.
7. The addressing architecture according to claim 6 , wherein the current address for the first storage area is calculated by performing a right-shift of the previous address of the first storage area and adding the parity bit of the previous address of the first storage area as the most significant bit, and the current address for the second storage area is calculated by performing a right-shift of the previous address of the second storage area and adding an inverted parity bit of the previous address of the second storage area as the most significant bit.
8. The addressing architecture according to claim 5 , wherein the parity bit is set to 1 if the total number of 1's in the corresponding current memory address is odd, and to 0 if the total number of 1's in the corresponding current memory address is even.
9. The addressing architecture according to claim 1 , wherein said two storage areas are used regardless of the number of butterflies used in the Viterbi butterfly structure.
10. The addressing architecture according to claim 1 , wherein said memory arranged with a first storage area for storing a first set of path metrics and a second storage area for storing a second set of path metrics comprises of two separate double-port memories storing a respective set of path metrics.
11. A mobile device comprising an addressing architecture according to claim 1 .