# US 8,787,860 B2 · Image cancellation in receivers using dual adaptive filters > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,787,860 B2 - Document ID: US8787860B2 - Status: Granted - Kind: B2 - Application number: 13/549,376 - Filed: 2012-07-13 - Published: 2013-01-24 - Granted: 2014-07-22 - Earliest claimed: 2011-07-21 - Assignees: Marvell World Trade Ltd. - Applicants: Shaoan Dai; Runsheng He - Inventors: Shaoan Dai; Runsheng He - Web page: https://patentgrep.com/patent/US8787860B2 ## CPC - H04B 1/28: the receiver comprising at least one semiconductor device having three or more electrodes - H04B 1/0007: {wherein the AD/DA conversion occurs at radiofrequency or intermediate frequency stage} - H04B: TRANSMISSION ## Abstract The present disclosure includes apparatus, systems, and techniques relating to receiver image cancellation. A described technique includes receiving a downconverted signal in a digital domain, the downconverted signal including an in-phase signal and a quadrature signal; generating a first signal of a signal channel based on the downconverted signal; generating a second signal of an image channel based on the downconverted signal; filtering the second signal using first weights to produce a pilot training signal; filtering the second signal using second weights to produce an image cancellation signal; generating an output signal by subtracting the image cancellation signal from the first signal to resolve the desired signal; updating the first weights based on the first weights, the second signal, the pilot training signal, and a pilot signal; and updating the second weights based on the second weights, the output signal, and the pilot training signal. ## Claims ### Claim 1 A method comprising: receiving a downconverted signal in a digital domain, the downconverted signal comprising an in-phase signal and a quadrature signal; generating a first signal of a signal channel based on the downconverted signal; generating a second signal of an image channel based on the downconverted signal; filtering the second signal using first weights to produce a pilot training signal; filtering the second signal using second weights to produce an image cancellation signal; generating an output signal by subtracting the image cancellation signal from the first signal to resolve a desired signal; updating the first weights based on the first weights, the second signal, the pilot training signal, and a pilot signal; and updating the second weights based on the second weights, the output signal, and the pilot training signal. ### Claim 2 (depends on 1) The method of claim 1, comprising: receiving a communications signal comprising the desired signal; generating the pilot signal; and combining the communications signal and a version of the pilot signal to produce a combined signal. ### Claim 3 (depends on 2) The method of claim 2, wherein the downconverted signal is generated by causing an analog hardware section to perform a quadrature downconversion on the combined signal. ### Claim 4 (depends on 3) The method of claim 3, wherein updating the first weights comprises using a least-mean-squares technique to iteratively adapt the first weights such that the pilot training signal replicates a version of the pilot signal that is responsive to an IQ imbalance of the analog hardware section. ### Claim 5 (depends on 4) The method of claim 4, wherein updating the second weights comprises using a least-mean-squares technique to iteratively adapt the second weights such that the image cancellation signal is responsive to the pilot training signal and signal leakage between the signal channel and the image channel as caused by the IQ imbalance. ### Claim 6 (depends on 2) The method of claim 2, wherein generating the pilot signal comprises using a random signal to produce the pilot signal. ### Claim 7 (depends on 1) The method of claim 1, wherein filtering the second signal using the first weights comprises providing a version of the first weights to a multi-tap finite impulse response filter. ### Claim 8 (depends on 1) The method of claim 1, wherein filtering the second signal using the second weights comprises providing a version of the second weights to a multi-tap finite impulse response filter. ### Claim 9 An apparatus, comprising: circuitry configured to receive a downconverted signal in a digital domain, the downconverted signal comprising an in-phase signal and a quadrature signal; and circuitry configured to perform operations in the digital domain, the operations comprising generating a first signal of a signal channel based on the downconverted signal, generating a second signal of an image channel based on the downconverted signal, filtering the second signal using first weights to produce a pilot training signal, filtering the second signal using second weights to produce an image cancellation signal, generating an output signal by subtracting the image cancellation signal from the first signal to resolve a desired signal, updating the first weights based on the first weights, the second signal, the pilot training signal, and a pilot signal, and updating the second weights based on the second weights, the output signal, and the pilot training signal. ### Claim 10 (depends on 9) The apparatus of claim 9, wherein the downconverted signal is responsive to a quadrature downconversion of a combined signal in an analog hardware section, wherein the combined signal is based on a communications signal and a version of the pilot signal, and wherein the operations comprise generating the pilot signal. ### Claim 11 (depends on 10) The apparatus of claim 10, wherein updating the first weights comprises using a least-mean-squares technique to iteratively adapt the first weights such that the pilot training signal replicates a version of the pilot signal that is responsive to an IQ imbalance of the analog hardware section. ### Claim 12 (depends on 11) The apparatus of claim 11, wherein updating the second weights comprises using a least-mean-squares technique to iteratively adapt the second weights such that the image cancellation signal is responsive to the pilot training signal and signal leakage between the signal channel and the image channel as caused by the IQ imbalance. ### Claim 13 (depends on 9) The apparatus of claim 9, wherein filtering the second signal using the first weights comprises providing a version of the first weights to a multi-tap finite impulse response filter. ### Claim 14 (depends on 9) The apparatus of claim 9, wherein filtering the second signal using the second weights comprises providing a version of the second weights to a multi-tap finite impulse response filter. ### Claim 15 A system, comprising: a radio frequency interface configured to receive a communications signal comprising a desired signal; a combiner configured to add the communications signal and a version of a pilot signal to produce a combined signal; an analog hardware section configured to perform a quadrature downconversion on the combined signal to produce a downconverted signal in a digital domain, the downconverted signal comprising an in-phase signal and a quadrature signal; a memory structure configured to store data comprising first weights and second weights; and a processor, communicatively coupled with the combiner, the analog hardware section, and the memory structure, the processor configured to perform operations comprising generating the pilot signal, generating a first signal of a signal channel based on the downconverted signal, generating a second signal of an image channel based on the downconverted signal, filtering the second signal using the first weights to produce a pilot training signal, filtering the second signal using the second weights to produce an image cancellation signal, generating an output signal by subtracting the image cancellation signal from the first signal to resolve the desired signal, updating the first weights based on the first weights, the second signal, the pilot training signal, and the pilot signal, and updating the second weights based on the second weights, the output signal, and the pilot training signal. ### Claim 16 (depends on 15) The system of claim 15, wherein updating the first weights comprises using a least-mean-squares technique to iteratively adapt the first weights such that the pilot training signal replicates a version of the pilot signal that is responsive to an IQ imbalance of the analog hardware section. ### Claim 17 (depends on 16) The system of claim 16, wherein updating the second weights comprises using a least-mean-squares technique to iteratively adapt the second weights such that the image cancellation signal is responsive to the pilot training signal and signal leakage between the signal channel and the image channel as caused by the IQ imbalance. ### Claim 18 (depends on 15) The system of claim 15, wherein filtering the second signal using the first weights comprises providing a version of the first weights to a multi-tap finite impulse response filter. ### Claim 19 (depends on 15) The system of claim 15, wherein filtering the second signal using the second weights comprises providing a version of the second weights to a multi-tap finite impulse response filter. ### Claim 20 (depends on 15) The system of claim 15, wherein generating the pilot signal comprises using a random signal to produce the pilot signal. ### Claim 21 (depends on 15) The system of claim 15, wherein the communications signal comprises one or more digital television signals. ### Claim 22 (depends on 15) The system of claim 15, further comprising: an upconverter to generate the version of the pilot signal by upconverting an analog version of the pilot signal into a frequency band of the desired signal.