Multiplex signal error correction method and device

A method of reducing a phase error caused by a plurality of error sources in a signal which is present in a digital frequency representation in the form of a sequence of a plurality of digital partial signals which are associated with a number of subcarriers (k) of a carrier, comprising, for each partial signal, the steps of:

• equalization of the partial signal (Y(i,k)),
• estimation of the phase error of the equalized partial signal (X(i,k)), and
• correction of the estimated phase error of the equalized partial signal,
• characterized in that:
• the equalization step includes a step of at least partially eliminating an accumulation of a phase error of the partial signal, caused by a sampling frequency error, over the sequence of the partial signals, such that the accumulation is negligible,
• the estimation step includes a step of detecting a plurality of predetermined pilot signals and a step of determining a phase correction factor on the basis of the detected pilot signals, wherein at least one multiplication operation is carried out solely by means of shift and adding operations,
• the correction step includes a step of complex multiplication of the equalized partial signal with the phase correction factor, and
• that the estimation step includes a step of calculating the following parameters on the basis of the pilot signals:
• cos(φ0)=(¼)(e{P−21}+e{P−7}+e{P+7}+e{P+21}),
• sin(φ0)=(¼)(m{P−21}+m{P−7}+m{P+7}+m{P+21}),
• 2πp0 sin(φ0)≈(2e{P−21}+3e{P−7}−3e{P+7}−2e{P+21})/128,
• −2πp0 cos(φ0)≈(2m{P−21}+3m{P−7}−3m{P+7}−2m{P+21})/128,
• wherein P−21, P−7, P+7, P+21 denote pilot signals and e and ℑm denote the operation of determining the real and imaginary part respectively.

A phase correction unit including a first signal input and a computing unit for determining a complex phase correction factor by means of arithmetic operations which include multiplication, wherein there is provided a second signal input to which the computing unit is connected, and at least one multiplication operation is implemented in the form of at least one shift operator in conjunction with at least one adder

• characterized in that the computing unit is adapted to determine the following parameters with the signals (P−21, R−7, P+7, P+21) received by way of the second signal input:
• cos(φ0)=(¼)(e{P−21}+e{P−7}+e{P+7}+e{P+21}),
• sin(φ0)=(¼)(ℑm{P−21}+ℑm{P−7}+ℑm{P+7}+ℑm{P+21}),
• 2πp0 sin(φ0)≈(2e{P−21}+3e{P−7}−3e{P+7}−2e{P+21})/128,
• −2πp0 cos(φ0)≈(2ℑm{P−21}+3ℑm{P−7}−3ℑm{P+7}−2ℑm{P+21})/128,
• wherein P−21, P−7, P+7, P+21 denote pilot signals and e and ℑm denote the operation of determining the real and imaginary part respectively.

A device for reducing a phase error caused by a plurality of error sources in a signal which is present in a digital frequency representation in the form of a sequence of a plurality of digital partial signals which are associated with a number of subcarriers (k) of a carrier, comprising

• an equalization unit configured for equalizing each partner signal,
• an estimation unit configured to estimate the phase error of the equalized partial signal, and
• a phase correction unit configured to correct the estimated phase error of the equalized partial signal, wherein:
• the equalization unit is configured to at least partially eliminate an accumulation of a phase error of the partial signal, caused by a sampling frequency error, over the sequence of the partial signals, such that the accumulation is negligible,
• the equalization unit is configured to divide the partial signal by a complex second signal which has the phase of the partial signal preceding in the sequence of the partial signals, wherein the second signal represents a frequency response function of the associated subcarrier for the preceding partial signal in the sequence of partial signals
• the estimation unit is configured to perform, as part of the estimation of the phase error of the equalized partial signal, a detection of a plurality of predetermined pilot signals and determination of a phase correction factor on the basis of the detected pilot signals, wherein at least one multiplication operation is carried out solely by means of shift and adding operations, and
• the estimation unit is configured to determine, for each partial signal with the exception of the pilot partial signals, the frequency response function on the basis of a preceding partial signal in the sequence of partial signals, wherein the determining of the frequency response function includes a division of the preceding partial signal in the sequence by a third signal, wherein the third signal represents the equalized, phase-corrected and decoded and then recoded preceding partial signal.