Numerous embodiments of analog neural memory arrays are disclosed. In certain embodiments, each memory cell in the array has an approximately constant source impedance when that cell is being operated. In certain embodiments, power consumption is substantially constant from bit line to bit line within the array when cells are being read. In certain embodiments, weight mapping is performed adaptively for optimal performance in power and noise.

Numerous embodiments of analog neural memory arrays are disclosed. In one embodiment, an analog neural memory system comprises an array of non-volatile memory cells, wherein the cells are arranged in rows and columns, the columns arranged in physically adjacent pairs of columns, wherein within each adjacent pair one column in the adjacent pair comprises cells storing W+ values and one column in the adjacent pair comprises cells storing W− values, wherein adjacent cells in the adjacent pair store a differential weight, W, according to the formula W=(W+)−(W−). In another embodiment, an analog neural memory system comprises a first array of non-volatile memory cells storing W+ values and a second array storing W− values.

A pipelined ADC that does not wait for the residue of a signal to settle to be delivered to the next stage of the pipeline, and thus passes signals to subsequent stages at faster than conventional speeds is described. A pipelined ADC is used that processes signals representing the boundaries of the search space. Thus, each stage does not necessarily receive the signal as pre-processed by the prior stage, but rather the search space boundaries as pre-processed by the prior stage. Reducing the “search space” of the ADC is equivalent to creating the residues in each step of a pipeline as in the prior art. An ADC operating in this fashion operates without error even if the residual search space boundary outputs from one state are presented to the next stage before the outputs have settled, and can run faster for a given power and bandwidth.

A pipeline analog-to-digital converter (ADC) includes a hybrid multiplying digital-to-analog converter (MDAC) that includes multiple digital-to-analog converters (DACs), at least one conversion circuit, and at least one amplifier such that a number of conversion circuits and a number of amplifiers is less than a number of DACs. Each DAC is configured to receive an analog input signal in non-overlapping durations of a clock signal and generate a corresponding analog output signal. At least one of the conversion circuits is coupled with at least two DACs, and each conversion circuit is configured to perform conversion operation on a corresponding analog output signal to generate digital signals. At least one of the amplifiers is coupled with at least two DACs, and each amplifier is configured to perform amplification operation on a corresponding analog output signal.

A sub-ranging analog-to-digital converter (ADC) includes a coarse ADC and a fine ADC that receives a set of coarse signals from the coarse ADC. The fine ADC includes multiple digital-to-analog converters (DACs) and multiple converters such that a number of converters is less than a number of DACs. The DACs and the converters function in a partial time-interleaved manner where each DAC receives an analog input signal in different non-overlapping durations of a clock signal and generates a corresponding analog output signal. At least one of the converters is coupled with at least two DACs, and each converter is configured to receive the corresponding analog output signals and perform conversion operation to generate digital signals in non-overlapping durations of the clock signal, respectively. The durations for performing conversion operation of at least two of the converters overlap partially.

An analog-to-digital converter (ADC) includes a first ADC stage with a first sub-ADC stage configured to sample the analog input voltage in response to a first phase clock signal and output a first digital value corresponding to an analog input voltage in response to a second phase clock signal. A current steering DAC stage is configured to convert the analog input voltage and the first digital value to respective first and second current signals, determine a residue current signal representing a difference between the first and the second current signal, and convert the residue current signal to an analog residual voltage signal. A second ADC stage is coupled to the first ADC stage to receive the analog residual voltage signal, and convert the analog residue voltage signal to a second digital value. An alignment and digital error correction stage is configured to combine the first and the second digital values.

An analog-to-digital converter (ADC) circuit includes a signal input terminal, a sample-and-hold circuit, and a successive approximation register (SAR) ADC. The sample-and-hold circuit includes an input terminal coupled to the signal input terminal. The SAR ADC includes a comparator, a first capacitive digital-to-analog converter (CDAC), and a second CDAC. The first CDAC includes a first input terminal coupled to the signal input terminal, a second input terminal coupled to an output terminal of the sample-and-hold circuit, and an output terminal coupled to a first input terminal of the comparator. The second CDAC includes a first input terminal coupled to the signal input terminal, an output terminal coupled to a second input terminal of the comparator.

An analog-to-digital converter (“ADC”) includes an input terminal configured to receive an analog input voltage signal. A first ADC stage is coupled to the input terminal and is configured to output a first digital value corresponding to the analog input voltage signal and a first analog residue signal corresponding to a difference between the first digital value and the analog input signal. An inverter based residue amplifier is configured to receive the first analog residue signal, amplify the first analog residue signal, and output an amplified residue signal. The amplified residue signal is converted to a second digital value, and the first and second digital values are combined to create a digital output signal corresponding to the analog input voltage signal.

A voltage-to-time-to-digital converter (VTDC) and conversion method are provided using a coarse analog-to-digital converter (ADC). A voltage-to-time converter (VTC) receives an analog input voltage-differential signal with a first time duration and supplies an analog first time-differential signal. An ADC receives the input voltage-differential signal and supplies a first digital code representing m bit values. A time-to-digital converter (TDC) receives a second time-differential signal with a second time duration derived from the first time duration. The TDC supplies an output digital code representing p bit values, where p>m. In one aspect the first digital code programs an initial set of TDC residue generators. In another aspect, a dither circuit controls the second time duration in response to a pseudo random signal combined with the first digital code.

The invention provides a signal processing system, for transferring analog signals from a probe to a remote processing unit. The system comprises a first ASIC at a probe, which is adapted to receive an analog probe signal. The first ASIC comprises an asynchronous sigma-delta modulator, wherein the asynchronous sigma-delta modulator is adapted to: receive the analog probe signal; and output a binary bit-stream. The system further comprises a second ASIC at the remote processing unit, adapted to receive the binary bit-stream. The asynchronous may further include a time gain function circuit, the first ASIC may further comprise a multiplexer, the second ASIC may further comprise a time-to-digital converter. The time to digital converter may be a pipelined time-to-digital converter.