Method and arrangement of increasing impairment co-variance matrix estimation accuracy

• US 8,897,339 B2
• Filed: 04/20/2010
• Issued: 11/25/2014
• Est. Priority Date: 04/20/2010
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Wang, Hai
• Primary Examiner(s): Garcia, Santiago
• Application Number:
  US13/642,116
• Publication Number:
  US 20130039389A1
• Time in Patent Office: 1,680 Days
• Field of Search: 375/148
• US Class Current: 375/148
• CPC Class Codes: H04B 1/712 Weighting of fingers for combining, e.g. amplitude control or phase rotation using an inner loop
• CPC Class Codes: H04B 2201/709727 GRAKE type RAKE receivers

Abstract

A method and arrangement of increasing impairment co-variance matrix R_u estimation accuracy in downlink in a user equipment in a communication network system. De-spread is performed on HS-DSCH symbols to form a matrix X of de-spread symbols. The matrix X and channel estimates h_c from CPiCH and modulation scheme information are used to form a matrix S of recovered symbols in hard value. The matrix X and the channel estimates h_c, the modulation scheme information and the matrix S output are used to increase the estimation accuracy of the matrix S. The previous step is repeated until the output symbols are the same as the input symbols or the number of iterations reaches a pre-defined maximum value. The matrix X and the matrix S with increased estimation accuracy are used to form an impairment co-variance matrix R_u estimate.

Claims

1. A method of increasing impairment co-variance matrix R_u estimation accuracy in downlink in a user equipment in a communication network system comprising a plurality of communication network nodes communicating with one or more user equipments, where each network node is serving at least one cell in which one or more user equipments are located, the method comprising the steps of:

performing de-spread on high speed-downlink channel symbols to form a matrix X of de-spread symbols;

using the matrix X of de-spread symbols and channel estimates h_c from common pilot channel and modulation scheme information to form a matrix S of recovered symbols in hard value;

using the matrix X of de-spread symbols and the channel estimates h_c from common pilot channel, the modulation scheme information and the matrix S of recovered symbols in hard value output to increase the estimation accuracy of the matrix S of recovered symbols in hard value;

repeating the use of the matrix X of de-spread symbols and the channel estimates h_c from common pilot channel, the modulation scheme information and the matrix S of recovered symbols in hard value output until the output symbols are the same as the input symbols or the number of iterations reaches a pre-defined maximum value;

using the matrix X of de-spread symbols and the matrix S of recovered symbols in hard value with increased estimation accuracy to form an impairment co-variance matrix R_u estimate; and

using the impairment co-variance matrix R_u estimate and the channel estimates h_c from common pilot channel to determine generalized RAKE combining weight w and to estimate common pilot channel signal to interference plus noise ratio.

2. The method according to claim 1, wherein the matrix X of de-spread symbols has a dimension of L-by-N, where L is the number of combining fingers and N is the number of de-spread symbols on each combining finger during one time slot.

3. The method according to claim 2, wherein the number of de-spread symbols on each combining finger during one time slot N=160×

K, where 160 is the number of symbols per code channel in one time slot and K is the number of codes employed for estimation.

4. The method according to claim 3, wherein the number of codes employed for estimation K=N_code, such that all allocated codes are employed for estimation.

5. The method according to claim 1, wherein symbols corresponding to a few or all allocated Walsh codes are used to form the matrix X of de-spread symbols.

6. The method according to claim 1, wherein the step of using the matrix X of de-spread symbols and channel estimates h_c from common pilot channel and the modulation scheme information to form a matrix S of recovered symbols in hard value comprises deploying RAKE or generalized RAKE.

7. The method according to claim 1, wherein the step of using the matrix X of de-spread symbols and the channel estimates hc from common pilot channel, the modulation scheme information and the matrix S of recovered symbols in hard value output to increase the estimation accuracy of the matrix S of recovered symbols in hard value comprises deploying generalized RAKE.

8. The method according to claim 1, wherein the step of using the impairment co-variance matrix R_u estimate and the channel estimates h_u from common pilot channel to determine generalized RAKE combining weight w and to estimate common pilot channel signal to interference plus noise ratio comprises determining the generalized RAKE combining weight w as w=R_c⁻

1h_c and the common pilot channel signal to interference plus noise ratio estimate SINR_cpich as SINR_cpich=16h_c^HR_u<sup>−

1</sup>h_c.

9. An arrangement of increasing impairment co-variance matrix R_u estimation accuracy in downlink in a user equipment in a communication network system comprising a plurality of communication network nodes communicating with one or more user equipments, where each network node is serving at least one cell in which one or more user equipments are located, the arrangement comprising:

plural delay fingers configured to perform de-spread on high speed-downlink shared channel symbols to obtain de-spread symbols and a generator configured to form a matrix X of the obtained de-spread symbols;

a processor for using the matrix X of de-spread symbols and channel estimates h_c from common pilot channel and modulation scheme information to form a matrix S of recovered symbols in hard value;

said processor further being arranged for using the matrix X of de-spread symbols and the channel estimates h_c from common pilot channel, the modulation scheme information and the matrix S of recovered symbols in hard value to increase the estimation accuracy of the matrix S of recovered symbols in hard value;

said processor further being arranged for repeating using the matrix X of de-spread symbols and the channel estimates h_c from common pilot channel, the modulation scheme information and the matrix S of recovered symbols in hard value to increase the estimation accuracy of the matrix S of recovered symbols in hard value until the output symbols are the same as the input symbols or the number of iterations reaches a pre-defined maximum value;

a generator configured to use the matrix X of de-spread symbols and the matrix S of recovered symbols in hard value with increased estimation accuracy to form an impairment co-variance matrix R_u estimate; and

said processor further being arranged for using the impairment co-variance matrix R_u estimate and the channel estimates h_c from common pilot channel to determine generalized RAKE combining weight w and to estimate common pilot channel signal to interference plus noise ratio.

10. The arrangement according to claim 9, arranged to handle a matrix X of de-spread symbols having a dimension of L-by-N, where L is the number of combining fingers and N is the number of de-spread symbols on each combining finger during one time slot.

11. The arrangement according to claim 9, arranged to handle a number of de-spread symbols on each combining finger during one time slot N=160−

K, where 160 is the number of symbols per code channel in one time slot and K is the number of codes employed for estimation.

12. The arrangement according to claim 9, arranged such that the number of codes employed for estimation K=N_code, such that all allocated codes are employed for estimation.

13. The arrangement according to claim 9, arranged to use symbols corresponding to a few or all allocated Walsh codes to form the matrix X of de-spread symbols.

14. The arrangement according to claim 9, arranged to deploy RAKE or generalized RAKE in using the matrix X of de-spread symbols and channel estimates h_c from common pilot channel and the modulation scheme information to form a matrix S of recovered symbols in hard value.

15. The arrangement according to claim 9, arranged to deploy generalized RAKE in using the matrix X of de-spread symbols and the channel estimates hc from common pilot channel, the modulation scheme information and the matrix S of recovered symbols in hard value to increase the estimation accuracy of the matrix S of recovered symbols in hard value.

16. The arrangement according to claim 9, arranged to determine the generalized RAKE combining weight w as w=R_u⁻

1h_c and the common pilot channel signal to interference plus noise ratio estimate SINR_cpich as SINR_cpich=16h_c^HR_u<sup>−

1</sup>h_c in using the impairment co-variance matrix R_u estimate and the channel estimates h_c from common pilot channel to determine generalized RAKE combining weight w and to estimate common pilot channel signal to interference plus noise ratio.

17. A user equipment for increasing impairment co-variance matrix R_u estimation accuracy in downlink in a communication network system comprising communication network nodes each serving at least one cell in which the user equipment is located, the user equipment comprising:

plural delay fingers configured to perform de-spread on high speed-downlink shared channel symbols to obtain de-spread symbols and a generator configured to form a matrix X of the obtained de-spread symbols;

a processor for using the matrix X of de-spread symbols and channel estimates h_c from common pilot channel and modulation scheme information to form a matrix S of recovered symbols in hard value;

said processor further being arranged for using the matrix X of de-spread symbols and the channel estimates h_c from common pilot channel, the modulation scheme information and the matrix S of recovered symbols in hard value to increase the estimation accuracy of the matrix S of recovered symbols in hard value;

said processor further being arranged for repeating using the matrix X of de-spread symbols and the channel estimates h_c from common pilot channel, the modulation scheme information and the matrix S of recovered symbols in hard value to increase the estimation accuracy of the matrix S of recovered symbols in hard value until the output symbols are the same as the input symbols or the number of iterations reaches a pre-defined maximum value;

a generator configured to use the matrix X of de-spread symbols and the matrix S of recovered symbols in hard value with increased estimation accuracy to form an impairment co-variance matrix R_u estimate; and

said processor further being arranged for using the impairment co-variance matrix R_u estimate and the channel estimates h_c from common pilot channel to determine generalized RAKE combining weight w and to estimate common pilot channel signal to interference plus noise ratio.