Channel estimation in CDMA communications systems using both lower power pilot channel and higher power date channel
View Patent ↗An iterative arrangement of channel estimator ( 12 ), a Kalman filter unit ( 14 ), and a soft demodulator ( 16 ) is provided for producing an estimate of complex gain of a CDMA communications channel carrying data and pilot channels, and demodulated data of the data channel. An initial channel estimate is produced by the estimator ( 12 ) using only the low-power pilot channel. This later is improved by the iterative process using the data channel. The complex gains is represented by a sum of sinusoidal signals with different frequencies and randomly variable amplitude and phase.
1. A method of estimating complex gain of a communications channel in a CDMA system in which a received signal communicated via the communications channel comprises at least one relatively higher power data channel and a relatively lower power pilot channel, comprising the steps of:
producing an estimated complex gain of the communications channel initially from only a pilot channel component of the received signal; and, in successive iterations:
estimating demodulated data of the data channel from the received signal and the estimated complex gain of the communications channel; and
improving the estimated complex gain of the communications channel in dependence upon the received signal and the estimated demodulated data of the data channel;
wherein the estimated complex gain of the communications channel is represented by a sum of a plurality of sinusoidal signals with different frequencies and with randomly variable amplitude and phase.
2. A method as claimed in claim 1 and comprising the step of transforming a vector of complex data samples representing the received signal to scalar complex numeric sequences, wherein the estimated complex gain of the communications channel is produced using said sequences.
3. A method as claimed in claim 2 wherein the step of transforming comprises multiplying the vector of complex data samples representing the received signal by a transposed impulse response matrix and by a vector that is conjugate to a complex PN code vector to produce a resulting vector, multiplying the resulting vector by a Walsh function of the data channel to produce a first scalar complex numeric sequence, and adding components of said resulting vector to produce a second scalar complex numeric sequence.
4. A method as claimed in claim 3 wherein the demodulated data of the data channel is estimated using the estimated complex gain of the communications channel and said first sequence.
5. A method as claimed in claim 1 wherein the step of producing and improving the estimated complex gain of the communications channel comprise Kalman filtering estimates of the complex gain of the communications channel.
6. A method as claimed in claim 2 wherein the step of producing and improving the estimated complex gain of the communications channel comprise Kalman filtering estimates of the complex gain of the communications channel.
7. A method as claimed in claim 3 wherein the step of producing and improving the estimated complex gain of the communications channel comprise Kalman filtering estimates of the complex gain of the communications channel.
8. A method as claimed in claim 4 wherein the step of producing and improving the estimated complex gain of the communications channel comprise Kalman filtering estimates of the complex gain of the communications channel.
9. A method as claimed in claim 5 and comprising the step of averaging estimates of the complex gain of the communications channel over a plurality of PN code chips prior to said Kalman filtering.
10. A method as claimed in claim 6 and comprising the step of averaging estimates of the complex gain of the communications channel over a plurality of PN code chips prior to said Kalman filtering.
11. A method as claimed in claim 7 and comprising the step of averaging estimates of the complex gain of the communications channel over a plurality of PN code chips prior to said Kalman filtering.
12. A method as claimed in claim 8 and comprising the step of averaging estimates of the complex gain of the communications channel over a plurality of PN code chips prior to said Kalman filtering.
13. Apparatus for carrying out the method of claim 1 , comprising a data transform unit for producing two scalar complex numeric sequences from a vector of complex data samples representing the received signal, and an iterative arrangement comprising:
a demodulator responsive to one of the two scalar complex numeric sequences and to an estimated complex gain of the communications channel to produce an estimated demodulated symbol of the data channel;
a complex gain estimator responsive to the two scalar complex numeric sequences and, except in a first iteration, to the estimated demodulated data symbol produced by the demodulator, for producing an estimated complex gain of the communications channel; and
a Kalman filter for filtering the estimated complex gain produced by the complex gain estimator to produce the estimated complex gain for the demodulator.
14. Apparatus as claimed in claim 13 and including an averaging unit for averaging complex gain estimates produced by the complex gain estimator over a plurality of PN code chips and for supplying an averaged complex gain estimate to the Kalman filter at a sampling rate less than a PM code chip rate.