Denoising diffusion models have emerged as state-of-the-art in generative tasks across image, audio, and video domains, producing high-quality, diverse, and contextually relevant data. However, their broader adoption is limited by high computational costs and large memory footprints. Post-training quantization (PTQ) offers a promising approach to mitigate these challenges by reducing model complexity through low-bandwidth parameters. Yet, direct application of PTQ to diffusion models can degrade synthesis quality due to accumulated quantization noise across multiple denoising steps, particularly in conditional tasks like text-to-audio synthesis. This work introduces PTQ4ADM, a novel framework for quantizing audio diffusion models(ADMs). Our key contributions include (1) a coverage-driven prompt augmentation method and (2) an activation-aware calibration set generation algorithm for text-conditional ADMs. These techniques ensure comprehensive coverage of audio aspects and modalities while preserving synthesis fidelity. We validate our approach on TANGO, Make-An-Audio, and AudioLDM models for text-conditional audio generation. Extensive experiments demonstrate PTQ4ADM’s capability to reduce the model size by up to 70% while achieving synthesis quality metrics comparable to full-precision models(<5% increase in FD scores). We show that specific layers in the backbone network can be quantized to 4-bit weights and 8-bit activations without significant quality loss. This work paves the way for more efficient deployment of ADMs in resource-constrained environments.

Wireless networks are undergoing an unprecedented revolution in the last decade. With the explosion of delay-sensitive applications usage on the Internet (i.e., online gaming, VoIP, and safety-critical applications), latency becomes a major issue for the development of wireless technology since it has an enormous impact on user experience. In fact, in a phenomenon known as bufferbloat, large static buffers inside the network devices results in increasing the time that packets spend in the queues and, thus, causing larger delays. Concerns have arisen about designing efficient queue management schemes to mitigate the effects of over-buffering in wireless devices. In this paper, we advocate the exploitation of machine learning techniques for dynamic buffer sizing. We propose LearnQueue, a novel reinforcement learning design that can effectively control the latency in wireless networks. LearnQueue adapts quickly and intelligently to changes in the wireless environment using a sophisticated reward structure. The latency control is performed dynamically by tuning the buffer size. Adopting a trial-and-error approach, the proposed scheme penalizes the actions resulting in longer delays or hurting the throughput. In addition, the scheme parameters are designed for an optimized operation depending on different applications requirements. Using the latest generation of WARP hardware, we investigated LearnQueue performance in various network scenarios. The testbed results prove that LearnQueue can grantee low latency while preserving throughput under various congestion situations. We also discuss the feasibility and possible limitations of large-scale deployment of the proposed scheme in wireless devices.

The controller is a critical piece of the SDN architecture, where it is considered as the mastermind of SDN networks. Thus, its failure will cause a significant portion of the network to fail. Overload is one of the common causes of failure since the controller is frequently invoked by new flows. Even through SDN controllers are often replicated, the significant recovery time can be an overkill for the availability of the entire network. In order to overcome the problem of the overloaded controller failure in SDN, this paper proposes a novel controller offload solution for failure mitigation based on a prediction module that anticipates the presence of a harmful long-term load. In fact, the long-standing load would eventually overwhelm the controller leading to a possible failure. To predict whether the load in the controller is short-term or long-term load, we used three different classification algorithms: Support Vector Machine, k-Nearest Neighbors, and Naive Bayes. Our evaluation results demonstrate that Support Vector Machine algorithm is applicable for detecting the type of load with an accuracy of 97.93% in a real-time scenario. Besides, our scheme succeeded to offload the controller by switching between the reactive and proactive mode in response to the prediction module output.

Wireless full-duplex radios can simultaneously transmit and receive using the same frequency. In theory, this can double the throughput. In fact, there is only little work addressing aspects other than throughput gains in full-duplex systems. Over-buffering in today's networks or the so-called “bufferbloat” phenomenon creates excessive end-to-end delays resulting in network performance degradation. Our analysis shows that full-duplex systems may suffer from high latency caused by bloated buffers. In this paper, we address the problem of buffer management in full-duplex networks by using Wireless Queue Management (WQM), which is an active queue management technique for wireless networks. Our solution is based on Relay Full-Duplex MAC (RFD-MAC), an asynchronous media access control protocol designed for relay full-duplexing. We compare the performance of WQM in full-duplex environment to Drop Tail mechanism over various scenarios. Our solution reduces the end-to-end delay by two orders of magnitude while achieving similar throughput in most of the cases.