Data-aided SIR estimation
A signal-to-interference estimate is generated using unknown data symbols in place of or in addition to pilot symbols. Data received over a data channel (traffic channel or control channel) are collected. The data symbols are then used to compute an observation metric based on deviations of the data symbols from a predetermined set of possible data symbols, wherein one of the data symbols and symbol constellation is normalized. A data channel signal-to-interference ratio is then computed based on the observation metric.
1. A method of generating a signal quality estimate, said method comprising:
receiving unknown data symbols on a data channel;
computing an observation metric based on deviations of said data symbols from a predetermined set of possible data symbols in a known symbol constellation, wherein one of the data symbols and the symbol constellation are normalized; and
correcting the observation metric by applying an inverse mapping function to obtain a corrected observation metric;
estimating a data channel signal-to-interference ratio based on the corrected observation metric.
2. The method of claim 1 wherein computing an observation metric based on deviations of said data symbols from a predetermined set of possible data symbols in a known symbol constellation comprises:
demodulating the received symbols to generate detected symbols;
computing normalized data symbols from said received data symbols; and
computing a noise power estimate as an average energy of the difference between said detected symbols and said normalized data symbols.
3. The method of claim 2 wherein demodulating the received symbols to generate detected symbols comprises determining, for each received symbol, a corresponding symbol in a predetermined symbol constellation that is closest to said received normalized data symbol.
4. The method of claim 2 wherein computing normalized data symbols from said received data symbols comprises:
computing a decision boundary scaling factor based on said received data symbols;
computing a normalization factor based on said decision boundary scaling factor; and
applying said normalization factor to said received data symbols.
5. The method of claim 2 wherein computing normalized data symbols from said received data symbols comprises:
determining a pilot SIR from received pilot symbols;
computing a normalization factor based on said pilot SIR and a data-to-pilot power ratio; and
applying said normalization factor to said received data symbols.
6. The method of claim 1 wherein computing an observation metric based on deviations of said data symbols from a predetermined set of possible data symbols comprises:
computing normalized data symbols from said received data symbols;
extracting soft values for one or more least significant bits of said normalized data symbols; and
computing a noise power estimate as an average soft value deviation from a p-norm measure.
7. The method of claim 1 wherein computing an observation metric based on deviations of said data symbols from a predetermined set of possible data symbols comprises:
computing normalized data symbols from said received data symbols; and
computing a count of said normalized data symbols falling outside a symbol constellation envelope.
8. The method of claim 1 further comprising transforming said data channel signal-to-interference ratio to an estimated pilot channel signal-to-interference ratio.
9. The method of claim 8 wherein transforming said data channel signal-to-interference ratio to an estimated pilot channel signal-to-interference ratio comprises determining a data-to-pilot power ratio; and scaling said data channel signal-to-interference ratio according to a said data-to-pilot power ratio.
10. A communication device comprising:
a receiver for receiving unknown data symbols over a data channel from a base station in a wireless communication network; and
processing circuits for processing said data symbols, said processing circuit configured to:
compute an observation metric based on deviations of said data symbols from a predetermined set of possible data symbols in a known symbol constellation, wherein one of the data symbols and the symbol constellation are normalized; and
correct the observation metric by applying an inverse mapping function to obtain a corrected observation metric;
estimate a data channel signal-to-interference ratio based on the corrected observation metric.
11. The communication device of claim 10 wherein said processing circuit is configured to compute said observation metric by:
computing normalized data symbols from said received data symbols;
demodulating the received symbols to generate detected symbols; and
computing a noise power estimate as an average energy of the difference between said detected symbols and said normalized data symbols.
12. The communication device of claim 11 wherein said processing circuit is configured to determine, for each received symbol, a corresponding symbol in a predetermined symbol constellation that is closest to said received normalized data symbol to demodulate said received data symbols.
13. The communication device of claim 11 wherein said processing circuit is configured to compute normalized data symbols from said received data symbols by:
computing a decision boundary scaling factor based on said received data symbols;
computing a normalization factor based on said decision boundary scaling factor; and
applying said normalization factor to said received data symbols.
14. The communication device of claim 11 wherein said processing circuit is configured to compute normalized data symbols from said received data symbols by:
determining a pilot SIR from received pilot symbols;
computing a normalization factor based on said pilot SIR and a data-to-pilot power ratio; and
applying said normalization factor to said received data symbols.
15. The communication device of claim 10 wherein said processing circuit is configured to compute an observation metric based on deviations of said data symbols from a predetermined set of possible data symbols by:
computing normalized data symbols from said received data symbols;
extracting soft values for one or more least significant bits of said normalized data symbols; and
computing a noise power estimate as an average soft value deviation from a p-norm measure.
16. The communication device of claim 10 wherein said processing circuit is configured to compute an observation metric based on deviations of said data symbols from a predetermined set of possible data symbols by:
computing normalized data symbols from said received data symbols; and
computing a count of said normalized data symbols falling outside a symbol constellation envelope.
17. The communication device of claim 10 wherein said processing circuit is further configured to transform said data channel signal-to-interference ratio to an estimated pilot channel signal-to-interference ratio.
18. The communication device of claim 17 wherein said processing circuit is configured to transform said data channel signal-to-interference ratio to an estimated pilot channel signal-to-interference ratio by determining a data-to-pilot power ratio; and scaling said data channel signal-to-interference ratio according to a said data-to-pilot power ratio.