Method and apparatus of joint adaptive channel encoding, adaptive system filtering, and maximum likelihood sequence estimation process by means of an unknown data training

Claim 2

In an adaptive receiver for use in an automatic and reliable detection of data signals coming through a transmission channel whose channel characteristics is dynamic and unknown, whereupon said adaptive receiver includes a receiver sampler and an adaptive system filtering device in series which concatenate said transmission channel to form an overall adaptive system connected to a decision device of said adaptive receiver for producing detected data symbols, and which enable an adaptive control over the unknown characteristics of said overall adaptive system under the guidance of an error estimation criterion underlying the present invention, a method of effecting an adaptive adjustment of the characteristics of said adaptive system filtering device with an unknown data training, comprising the steps of: (a) choosing an ideal system reference for the characteristics of said overall adaptive system by a user of said adaptive receiver, (b) making an adaptive estimate of the characteristics of said overall adaptive system with said unknown data training, (c) forming an error measurement on the basis of said error estimation criterion when the results in steps (a) and (b) are available, and (d) using a signal resulting from said error measurement in step (c) as a control signal which is fedback to said adaptive system filtering device such that its characteristics will be adjusted and hence the characteristics of said overall adaptive system will be under said adaptive control according to said ideal system reference referred to in step (a).

Claim 3

Method according to claim 1, characterized in that (a) a basic part of said error estimation criterion underlying the present invention is defined as an error function based on the difference between said adaptive estimate of the characteristics of said overall adaptive system pointed out in claim 1 step (b) and said ideal system reference pointed out in claim 1 step (a), (b) said adaptive estimate of the characteristics of said overall adaptive system has been formulated in terms of sampled input and output signals of said adaptive system filtering device and detected data symbols from said decision device, all of which are unknown yet measurable quantities, and (c) said error estimation criterion has been extended to an extended error estimation criterion by adding to said basic part defined on a system-characteristics basis in (a) a second term which is defined with respect to both symbol-by-symbol basis formed by a comparison between a function of detected data symbol and an estimated symbol, and system-characteristics basis, gaining advantages of: (c 1 ) being able to weight said basic part of the extended error estimation criterion by a numerical factor and said second term of said extended error estimation criterion by another numerical factor to render the flexibility of said extended error estimation criterion in application, (c 2 ) maintaining validity of said adaptive control over the unknown characteristics of said overall adaptive system stated in claim 1 when said extended error estimation criterion is used instead of said error estimation criterion originally used in claim 1, defined in (a) of claim 2, (c 3 ) improving convergence properties of said adaptive adjustment of the characteristics of said adaptive system filtering device and said adaptive control over the characteristics of said overall adaptive system, and (c 4 ) being capable of presenting the result of an adaptive procedure by a conventional MSE criterion with a known data training the same as what said adaptive procedure converges to by said extended error estimation criterion with an unknown data training.

Claim 4

Method according to claim 2, characterized in that when an adaptive sampling device is used instead of said receiver sampler and augmented with a division device, a differentiator, an adaptive timing adjustment device, an adder, and a duplicate of said adaptive system filtering device to form an adaptive timing recovery loop, there is a joint working of the following two adaptive processes (a) said adaptive adjustment of the characteristics of said adaptive system filtering device, and (b) an adaptive adjustment of timing in said adaptive sampling device to support each other by the same extended error estimation criterion with an unknown data training, regardless of whether a sampling rate is set to a value that is equal to or higher than input signal rate to said overall adaptive system.

Claim 5

Method according to claim 1, characterized in that an adaptive process is designed for updating the characteristics of said adaptive system filtering device by said error estimation criterion to such an extent that the adaptive algorithms, which are underlying said adaptive process, can be set into a storage area of said adaptive system filtering device and executed by the logic circuits built in said adaptive system filtering device when said control signal is applied to said adaptive system filtering device as stated in claim 1 step (d).

Claim 6

Method according to claim 1, characterized in that there is a system identification model in the sense that when unknown characteristics of said transmission channel and a system are to be identified, they can be derived from the knowledge of the characteristics of both said adaptive system filtering device and said overall adaptive system which is made available by operating said system identification model by: (a) sending a data sequence, which is unknown at the receiving end of said transmission channel and said system, as input to said transmission channel and said system to excite said overall adaptive system, namely, said system identification model, and producing an output signal at the output end of said system identification model, (b) adjusting adaptively the characteristics of said adaptive system filtering device according to said output signal from step (a) of claim 5 and evaluating the adjustment result according to said extended error estimation criterion, and concurrently (c) controlling adaptively the characteristics of said overall adaptive system towards said ideal system reference chosen by a user intending to do said system identification.

Claim 7

For a digital communication system using a quadrature amplitude modulation (QAM) technique which is a concatenation of a transmission channel of unknown characteristics and a receiving apparatus wherein an adaptive convolutional channel encoding process and an adaptive compensation process for the working of said receiving apparatus are concurrently optimized with an unknown data training in their operation under the guidance of the extended error estimation criterion underlying the present invention, an apparatus for realizing the joint optimization of the two processes just stated, characterized in that it includes: means for carrying out said adaptive convolutional channel encoding process through an adaptive channel encoder, which is formed by the concatenation of said transmission channel, and an adaptive system filtering device within said receiving apparatus, wherein the unknown characteristics of said digital communication system performed as said adaptive channel encoder is under adaptive control by the extended error estimation criterion with said unknown data training and wherefore a user of said apparatus has to specify a system reference involved in the extended error estimation criterion, and means for carrying out said adaptive compensation process through an adaptive sampling device within an adaptive timing recovery loop for the working of said receiving apparatus by the extended error estimation criterion whose system reference has been specified by said user of said apparatus.

Claim 8

For a digital communication system using a quadrature amplitude modulation (QAM) technique which is a concatenation of a transmission channel of unknown characteristics and a receiving apparatus wherein an adaptive convolutional channel encoding process and an adaptive compensation process for the working of said receiving apparatus are concurrently optimized with an unknown data training in their operation under the guidance of the extended error estimation criterion underlying the present invention, an apparatus for realizing the joint optimization of the two processes just stated, characterized in that it includes: means for carrying out said adaptive convolutional channel encoding process through an adaptive channel encoder, which is formed by the concatenation of said transmission channel, and an adaptive system filtering device within said receiving apparatus, wherein the unknown characteristics of said digital communication system performed as said adaptive channel encoder is under adaptive control by the extended error estimation criterion with said unknown data training and wherefore a user of said apparatus has to specify a system reference involved in the extended error estimation criterion, and means for carrying out said adaptive compensation process through an adaptive sampling device within an adaptive timing recovery loop for the working of said receiving apparatus by the extended error estimation criterion whose system reference has been specified by said user of said apparatus.

Claim 9

Apparatus according to claim 6, characterized in that said receiving apparatus includes an adaptive sampling device and an adaptive system filtering device in series which concatenate said transmission channel to form said digital communication system connected to a decision device of said receiving apparatus such that when said user's specification of said system reference required for the extended error estimation criterion is an ideal system reference of a partial response signaling format, said digital communication system forms an adaptive convolutional channel encoder by the extended error estimation criterion in the sense that input signals to said digital communication system are transformed by convolution of the characteristics of said transmission channel and said adaptive system filtering device and are thus encoded in a manner determined by said user's specification of said ideal system reference of said partial response signaling format in the extended error estimation criterion.

Claim 10

Apparatus according to claim 7, characterized in that the extended error estimation criterion used for said adaptive system filtering device to control the characteristics of said adaptive convolutional channel encoder provides a means to match the distribution of spectral nulls if any in the characteristics of said transmission channel with spectral nulls of the characteristics of said adaptive convolutional channel encoder whose locations are controllable.

Claim 11

Apparatus according to claim 6, characterized in that said adaptive sampling device in claim 7 operated within an adaptive timing recovery loop, which is formed by the concatenation of said adaptive sampling device, a division device, a differentiator, an adaptive timing adjustment device, an adder, said adaptive system filtering device, a duplicate of said adaptive system filtering device, and said decision device, plays the role of an adaptive system compensator to carry out said adaptive compensation process for the working of said receiving apparatus by using the same extended error estimation criterion in claim 7.

Claim 12

For an automatic, flexible and reliable digital communication system using a QAM technique which is formed by the concatenation of a transmitter encoder, a transmission channel of unknown characteristics and a receiving apparatus wherein an adaptive convolutional channel encoding process, an adaptive compensation process for the working of said receiving apparatus and a maximum likelihood sequence estimation (MLSE) process are concurrently optimized with an unknown data training in their operation under the guidance of the extended error estimation criterion underlying the present invention, an apparatus for realizing the joint optimization of the three processes just stated, characterized in that it includes: means for carrying out said adaptive convolutional channel encoding process through an adaptive convolutional channel encoder wherein the characteristics of said adaptive convolutional channel encoder, which is formed by the concatenation of said transmission channel, an adaptive sampling device and an adaptive system filtering device within said receiving apparatus, is under adaptive control by the extended error estimation criterion with said unknown data training and wherefore a user of said receiving apparatus has to specify an ideal system reference involved in the extended error estimation criterion, means for carrying out said adaptive compensation process which is jointly implemented by said adaptive sampling device and said adaptive system filtering device within said receiving apparatus to support said adaptive convolutional channel encoding process by the extended error estimation criterion whose ideal system reference has been specified by said user of said receiving apparatus, and means for carrying out said MLSE process which is implemented by an MLSE receiver-decoder to remove the possibility of an error propagation in a partial response signaling system and to support said adaptive compensation process.

Claim 13

For an automatic, flexible and reliable digital communication system using a QAM technique which is formed by the concatenation of a transmitter encoder, a transmission channel of unknown characteristics and a receiving apparatus wherein an adaptive convolutional channel encoding process, an adaptive compensation process for the working of said receiving apparatus and a maximum likelihood sequence estimation (MLSE) process are concurrently optimized with an unknown data training in their operation under the guidance of the extended error estimation criterion underlying the present invention, an apparatus for realizing the joint optimization of the three processes just stated, characterized in that it includes: means for carrying out said adaptive convolutional channel encoding process through an adaptive convolutional channel encoder wherein the characteristics of said adaptive convolutional channel encoder, which is formed by the concatenation of said transmission channel, an adaptive sampling device and an adaptive system filtering device within said receiving apparatus, is under adaptive control by the extended error estimation criterion with said unknown data training and wherefore a user of said receiving apparatus has to specify an ideal system reference involved in the extended error estimation criterion, means for carrying out said adaptive compensation process which is jointly implemented by said adaptive sampling device and said adaptive system filtering device within said receiving apparatus to support said adaptive convolutional channel encoding process by the extended error estimation criterion whose ideal system reference has been specified by said user of said receiving apparatus, and means for carrying out said MLSE process which is implemented by an MLSE receiver-decoder to remove the possibility of an error propagation in a partial response signaling system and to support said adaptive compensation process.

Claim 14

Apparatus according to claim 10, characterized in that said receiving apparatus includes an adaptive sampling device and an adaptive system filtering device in series with an MLSE receiver-decoder as a soft decision device, wherein an adaptive timing recovery loop for said adaptive compensation process is formed by the concatenation of said adaptive sampling device, a division device, a differentiator, an adaptive timing adjustment device, an adder, said adaptive system filtering device and its duplicate, while said convolutional channel encoding process is carried out in the combination of said transmission channel, said adaptive sampling device and said adaptive system filtering device.

Claim 15

Apparatus according to claim 10, characterized in that in order to remove the possibility of said error propagation at the output of said receiving apparatus for a partial response signaling transmission system, said transmitter encoder which comprises a signal mapping device and a precoding device is installed at the input of said transmission channel, while said MLSE receiver-decoder which comprises an MLSE processor, a decoding device and an inverse mapping device is used as said soft decision device of said receiving apparatus for said MLSE process.