Digital predistortion (DPD) is a popular technique to enhance signal quality in wideband radio frequency (RF) power amplifiers (PAs). With increasing bandwidth and data rates, DPD faces significant energy consumption challenges during deployment, contrasting with its efficiency goals. State-of-the-art DPD models rely on recurrent neural networks (RNNs), whose computational complexity hinders system efficiency. This letter introduces DeltaDPD, exploring the dynamic temporal sparsity of input signals and neuronal hidden states in RNNs for energy-efficient DPD, reducing arithmetic operations and memory accesses while preserving satisfactory linearization performance. Applying a TM3.1a 200 MHz-BW 256-QAM OFDM signal to a 3.5-GHz GaN Doherty RF PA, DeltaDPD achieves −50.03 dBc in adjacent channel power ratio (ACPR), −37.22dB in normalized mean square error (NMSE) and −38.52 dB in error vector magnitude (EVM) with 52% temporal sparsity, leading to a 1.8\times reduction in estimated inference power.

The development of safe and high-performance Li-metal anodes is crucial to meet the demanded increase in energy density of batteries. However, severe reactivity of Li metal with typical electrolytes and dendrite formation leads to a poor cycle life and safety concerns. Therefore, it is essential to develop electrolytes that passivate the reactivity toward Li metal and suppress dendrite formation. Carbonate electrolytes display severe reactivity toward Li metal; however, they are preferred above the more volatile ether-based electrolytes. Here, a carbonate electrolyte gel polymer approach is combined with LiNO₃ as an additive to stabilize Li-metal plating. This electrolyte design strategy is systematically monitored by operando neutron depth profiling (NDP) to follow the evolution of the plated Li-metal density and the inactive lithium in the solid electrolyte interface (SEI) during cycling. Individually, the application of the LiNO₃ electrolyte additive and the gel polymer approach are shown to be effective. Moreover, when used in conjunction, the effects are complementary in increasing the plated Li density, reducing inactive Li species, and reducing the overpotentials. The LiNO₃ additive leads to more compact plating; however, it results in a significant buildup of inactive Li species in a double-layer SEI structure, which challenges the cell performance over longer cycling. In contrast, the gel polymer strongly suppresses the buildup of inactive Li species by immobilizing the carbonate electrolyte species; however, the plating is less dense and occurs with a significant overpotential. Combining the LiNO₃ additive with the gel polymer approach results in a thin and homogeneous SEI with a high conductivity through the presence of Li₃N and a limited buildup of inactive Li species over cycling. Through this approach, even high plating capacities, reaching 7 mAh/cm², can be maintained at a high efficiency. The rational design strategy, empowered by monitoring the Li-density evolution, demonstrates the possibilities of achieving stable operation of Li metal in carbonate-based electrolytes.