Approximate bit-loading for data transmission over frequency-selective channels

A data receiving subsystem, comprising:

• a plurality of parallel data receiving circuits, each having an input coupled to a input node for receiving a common input signal, each data receiving circuit including:
• a multiplier for multiplying the common input signal by a respective vector to produce a respective sub-channel signal, wherein the respective vector corresponds to a sequence having N elements and the respective vector used by the multiplier in each of the plurality of data receiving circuits is substantially orthogonal to respective vectors used by the multipliers in each of the other data receiving circuits, and wherein the respective vectors are substantially orthogonal during a finite time interval corresponding to each sequence of N elements;
• a timing recovery mechanism to determine synchronization information from the respective sub-channel signal produced by the multiplier;
• an integrator for integrating the respective sub-channel signal in accordance with the synchronization information determined by the timing recovery mechanism to produce a respective integrated sub-channel signal, wherein the respective integrated sub-channel signal has a data rate substantially less than a data rate of the common input signal;
• and
• a converter configured to convert the respective integrated sub-channel signal into a digital signal.

A method of transmitting data, comprising:

• receiving a plurality of subsets of a data stream, wherein each of the respective subsets of the data stream has a respective data rate substantially less than a data rate of the data stream;
• converting each of the respective subsets of the data stream into a respective unmodulated analog signals;
• multiplying each of the respective analog signals by a respective vector to produce a respective sub-channel signal, wherein the respective vector corresponds to a sequence having N elements and the respective vector is substantially orthogonal to respective vectors used for multiplying each of the other respective analog signals, and wherein the respective vectors are substantially orthogonal during a finite time interval corresponding to each sequence of the N elements;
• and
• combining the respective sub-channel signals prior to transmission;
• and further comprising allocating portions of the data stream to form the respective subsets of the data stream such that spectral content of resulting respective sub-channel signals substantially corresponds to a transfer function of a transmission channel.

A method of transmitting data, comprising:

• receiving a plurality of subsets of a data stream, wherein each of the respective subsets of the data stream has a respective data rate substantially less than a data rate of the data stream;
• coding respective subsets of the data stream using a bit-to-symbol code such that respective subsets of the data stream convey different amounts of information per unit time than other respective subsets of the data stream, thereby producing a spectrum corresponding to the resulting respective sub-channel signals that substantially corresponds to a transfer function of a communications channel;
• converting each of the respective subsets of the data stream into a respective unmodulated analog signals;
• multiplying each of the respective analog signals by a respective vector to produce a respective sub-channel signal, wherein the respective vector corresponds to a sequence having N elements and the respective vector is substantially orthogonal to respective vectors used for multiplying each of the other respective analog signals, and wherein the respective vectors are substantially orthogonal during a finite time interval corresponding to each sequence of the N elements;
• and
• combining the respective sub-channel signals prior to transmission.

A method of transmitting data, comprising:

• receiving a plurality of subsets of a data stream, wherein each of the respective subsets of the data stream has a respective data rate substantially less than a data rate of the data stream;
• converting each of the respective subsets of the data stream into a respective unmodulated analog signals;
• multiplying each of the respective analog signals by a respective vector to produce a respective sub-channel signal, wherein the respective vector corresponds to a sequence having N elements and the respective vector is substantially orthogonal to respective vectors used for multiplying each of the other respective analog signals, and wherein the respective vectors are substantially orthogonal during a finite time interval corresponding to each sequence of the N elements;
• and
• combining the respective sub-channel signals prior to transmission;
• and further comprising demultiplexing the data stream to form the respective subsets of the data stream such that spectral content of resulting respective sub-channel signals substantially corresponds to a transfer function of a transmission channel.

A method of receiving data, comprising:

• receiving a common input signal;
• multiplying the common input signal by a plurality of respective vectors to produce a plurality of respective sub-channel signals, wherein each of the respective vectors corresponds to a sequence having N elements and each of the respective vectors is substantially orthogonal to the other respective vectors, and wherein the respective vectors are substantially orthogonal during a finite time interval corresponding to each sequence of N elements;
• recovering synchronization information from each of the sub-channel signals;
• integrating each of the respective sub-channel signals in accordance with the recovered synchronization information recovered from the respective sub-channel signal to produce respective integrated sub-channel signals, wherein each of the respective integrated sub-channel signals has a data rate substantially less than a data rate of the common input signal;
• and
• converting each of the respective integrated sub-channel signals into a digital signal.

A data receiving subsystem, comprising:

• a mechanism configured to receive a common input signal, to multiply the common input signal by a plurality of respective vectors to produce a plurality of respective sub-channel signals, to recover synchronization information from each of the respective sub-channel signals, to integrate each of the respective sub-channel signals in accordance with the synchronization information recovered from the respective sub-channel signal to produce respective integrated sub-channel signals, and to convert the respective integrated sub-channel signals into digital signals,
• wherein an inverse of the data rate of the common input signal is an integer multiple of a bit time of each of the respective vectors, each of the respective vectors corresponds to one of a plurality of sequences of binary numbers with a number of elements N that is an even integer, each element of the respective vectors corresponds to one of two voltage levels, each of the respective vectors is substantially orthogonal to other respective vectors during a finite time interval of N times a bit time for each of the respective vectors, each of the respective sub-channel signals is integrated for substantially the finite time interval, each of the respective sub-channel signals corresponds to at least one frequency band that at least partially overlaps a frequency band for at least one other of the respective sub-channel signals and each of the respective integrated sub-channel signals has a data rate substantially less than a data rate of the common input signal.