In asymmetric memory systems, the pull-up and pull-down data from the channel can have different amounts of inter-symbol interference (ISI), so they cannot be fully corrected by an equalizer which uses a single tap weight. We introduce a receiver-side single-ended 1-tap asymmetric decision-feedback equalizer (DFE) with a weight selection multiplexer, which allows the application of a different weight to each direction of data value transition. Implemented in a 55nm CMOS technology, our DFE compensated for asymmetric ISI in a 10.4Gb/s signal from a channel with an insertion loss of -8.3dB, leading to a 50% wider horizontal margin at a bit-error-rate of 10-12 than symmetric correction. The energy efficiency is 0.16pJ/bit.
20 July 2023 in IEEE Transactions on Circuits and Systems I: Regular Papers
In multi-lane interfaces using single-ended signaling such as memory interfaces, far-end crosstalk (FEXT) noise of the aggressor signal severely degrades signal integrity of the victim signal. A circuit for crosstalk cancellation (XTC) can reduce FEXT noise. However, the channel spacing makes the flight time difference between the forward and FEXT signals. As a result, the residual FEXT can remain. To further minimize the residual FEXT, this study proposes an XTC method to adjust the amplitude and timing independently. The timing is adjusted according to the passive element’s value of the differentiator, and the amplitude is varied by using the high-frequency boosting circuit. Moreover, a continuous-time linear equalizer (CTLE) and a 1-tap decision feedback equalizer (DFE) are applied to reduce inter-symbol interference. A prototype chip of 2-lane receiver was fabricated in a 55-nm CMOS process to verify this scheme. Using the proposed XTC, CTLE, and 1-tap DFE, timing margins of 0.21 UI and 0.36 UI were achieved in 6-mil and 15-mil channel spacings at 6.4 Gb/s, both at a bit error rate of 10 $^{-12}$ .
26 March 2018 in IEEE Transactions on Circuits and Systems II: Express Briefs
A 4266Mb/s/pin LPDDR4 interface with an asynchronous feedback continuous-time linear equalizer (AF-CTLE) and an adaptive 3-step eye detection algorithm for memory controller is presented. The AF-CTLE removes the glitch of DQS without training by applying an offset larger than the noise, and improves read margin by operating as a decision feedback equalizer in DQ path. The adaptive 3-step eye detection algorithm reduces power consumption and black-out time in initialization sequence and retraining in comparison to the 2-D full scanning. A prototype chip was implemented in a 65-nm CMOS process with fine-pitch ball grid array package and tested with commodity LPDDR4 memory. The write margin was 0.36 UI and 148 mV; and the read margin was enhanced from 0.30 UI and 76 mV without AF-CTLE to 0.47 UI and 80 mV to with AF-CTLE. The power efficiency during burst write and read were 5.68 pJ/bit and 1.83 pJ/bit, respectively.
The signal integrity of dual-rank LPDDR5 interface is challenged by reflections from the stub of the inactive rank within the redistribution layer (RDL). These reflections arrive at the active rank within the unit-interval of the data-rate and distort not only the post-cursor but also the pre-cursor of the signal that the decision feedback equalizer (DFE) cannot remove. We introduce a differentiating receiver that restores data through edge information to eliminate the reflective distortions, including the pre-cursor, and a transition-detecting DFE to address remaining inter-symbol interference. A prototype receiver, fabricated in a 65 nm CMOS process, eliminates pre-cursor distortion with a reflection flight time of 50 ps and 70 ps, at 8 Gb/s and 6.4 Gb/s, resulting in a wide horizontal margin at a bit-error-rate of 10 -12 . The energy efficiency is 0.80 pJ/bit at 8 Gb/s.
12 November 2021 in IEEE Transactions on Circuits and Systems I: Regular Papers
The dual-rank configuration is one of the parallelization methods to increase the memory capacity for mobile applications. However, the stub mainly composed of the redistribution layer causes resonance reflections and its reflection interval reaches the bit period, which distorts signal and limits the signal bandwidth. A single-ended duobinary transmitter that utilizes the reflection is presented for dual-rank mobile memory interfaces where reflections dominate. The reflection is included in the transfer function for duobinary modulation, which allows the transmitter to have a wide eye opening and high energy efficiency. Two-tap reverse feed-forward equalization and slew-rate control are used to support the duobinary modulation in combination with the reflection. A prototype chip fabricated in a 65 nm CMOS process has an area of 0.0378 mm 2 and consumes 1.24 pJ/bit. It is verified at data rates of 8, 9 and 10 Gb/s/pin where the flight time of the 9 mm stub is 55.6 ps.