Related papers: Efficient Hybrid Amplitude-Phase Quantization for …
We develop a coordinated hybrid automatic repeat request (HARQ) approach. With the proposed scheme, if a user message is correctly decoded in the first HARQ rounds, its spectrum is allocated to other users, to improve the network outage…
Federated Learning (FL) is a decentralized model training approach that preserves data privacy but struggles with low efficiency. Quantization, a powerful training optimization technique, has been widely explored for integration into FL.…
Cooperative hybrid-ARQ (HARQ) protocols, which can exploit the spatial and temporal diversities, have been widely studied. The efficiency of cooperative HARQ protocols is higher than that of cooperative protocols, because retransmissions…
Mixed-precision quantization (MPQ) is crucial for deploying deep neural networks on resource-constrained devices, but finding the optimal bit-width for each layer represents a complex combinatorial optimization problem. Current…
Many wireless systems divide the baseband processing between two locations, interconnected by a fronthaul. This paper examines the impact of fronthaul quantization on multiple-input multiple-output (MIMO) systems. Starting from a…
Large language models (LLMs) show great performance in various tasks, but face deployment challenges from limited memory capacity and bandwidth. Low-bit weight quantization can save memory and accelerate inference. Although floating-point…
Quantization techniques commonly reduce the inference costs of neural networks by restricting the precision of weights and activations. Recent studies show that also reducing the precision of the accumulator can further improve hardware…
We consider a multi-pair two-way amplify-and-forward massive multi-input multi-output (MIMO) hybrid relay with MIMO user-pairs. A hybrid relay has lesser number of radio frequency (RF) chains than the antennas, which significantly reduces…
We present accumulator-aware quantization (A2Q), a novel weight quantization method designed to train quantized neural networks (QNNs) to avoid overflow when using low-precision accumulators during inference. A2Q introduces a unique…
Fronthaul quantization causes a significant distortion in cell-free massive MIMO networks. Due to the limited capacity of fronthaul links, information exchange among access points (APs) must be quantized significantly. Furthermore, the…
In this paper, we consider a three node relay network comprising a source, a relay, and a destination. The source transmits the message to the destination using hybrid automatic repeat request (HARQ) with incremental redundancy (IR). The…
In this paper, a multiple-relay network in considered, in which $K$ single-antenna relays assist a single-antenna transmitter to communicate with a single-antenna receiver in a half-duplex mode. A new Amplify and Forward (AF) scheme is…
Inspired by several delay-bounded mission-critical applications, this paper investigates chase-combining-based hybrid automatic repeat request (CC-HARQ) protocols to achieve high reliability in delay-constrained applications. A salient…
Hybrid quantum systems in which acoustic resonators couple to superconducting qubits are promising quantum information platforms. High quality factors and small mode volumes make acoustic modes ideal quantum memories, while the qubit-phonon…
Neural networks with sub-microsecond inference latency are required by many critical applications. Targeting such applications deployed on FPGAs, we present High Granularity Quantization (HGQ), a quantization-aware training framework that…
Quantization is an effective method for reducing memory footprint and inference time of Neural Networks, e.g., for efficient inference in the cloud, especially at the edge. However, ultra low precision quantization could lead to significant…
Scalar Quantize & Forward (QF) schemes are studied for the Two-Way Relay Channel. Different QF approaches are compared in terms of rates as well as relay and decoder complexity. A coding scheme not requiring Slepian-Wolf coding at the relay…
The widespread adoption of Federated Learning (FL), a privacy-preserving distributed learning methodology, has been impeded by the challenge of high communication overheads, typically arising from the transmission of large-scale models.…
Massive multiple-input multiple-output (MIMO) avails of simple transceiver design which can tackle many drawbacks of relay systems in terms of complicated signal processing, latency, and noise amplification. However, the cost and circuit…
Hybrid models that combine convolutional and transformer blocks offer strong performance in computer vision (CV) tasks but are resource-intensive for edge deployment. Although post-training quantization (PTQ) can help reduce resource…