Deep learning-based methods have achieved significant success in the field of image fusion, where the design of network architectures plays a crucial role in the fusion task. However, most deep learning fusion architectures still operate as black boxes and lack awareness of frequency domain information. To enhance the interpretability of fusion tasks and effectively utilize the frequency domain information from source images while ensuring the generation of high-quality fused images, we propose a novel deep model-driven network for infrared and visible image fusion guided by diffusion priors. The proposed algorithm generates a distribution of multi-channel input data through a diffusion process. Features produced by the denoiser serve as knowledge priors, guiding a custom model that leverages frequency domain knowledge to reconstruct high-quality fused images. Specifically, unlike some traditional fusion networks, our model can retain multi-channel data at the input stage rather than just single-channel spatial information. It uses the multi-channel features generated by the diffusion model as priors to guide the fusion process. Moreover, we innovatively design a frequency-domain-based objective function to guide the fusion process, constructing a frequency-domain learning module to simulate an interpretable deep model-driven network. Additionally, a task-driven loss function is developed to ensure the quality of the fused images. Extensive experimental evaluations across seven diverse datasets (e.g., MSRS, M3FD, RoadSence, TNO, Havard) and multiple scenarios demonstrate that the proposed algorithm significantly outperforms 9 state-of-the-art methods. Specifically, it delivers superior fusion results on eight metrics (e.g., EN, SF, VIF) with notable improvements in interpretability and robustness, as validated through comprehensive experiments on these seven benchmark datasets.

Constructive interference (CI) enables concurrent transmissions to interfere non-destructively, so as to enhance network concurrency. In this paper, we propose deliberate synchronized constructive interference (Disco), which ensures concurrent transmissions of an identical packet to synchronize more precisely than traditional CI. Disco envisions concurrent transmissions to positively interfere at the receiver, and potentially allows orders of magnitude reductions in energy consumption and improvements in link quality. We also theoretically introduce a sufficient condition to construct Disco with IEEE 802.15.4 radio for the first time. Moreover, we propose Triggercast, a distributed middleware service, and show it is feasible to generate Disco on real sensor network platforms like TMote Sky. To synchronize transmissions of multiple senders at the chip level, Triggercast effectively compensates propagation and radio processing delays, and has 95th percentile synchronization errors of at most 250 ns. Triggercast also intelligently decides which co-senders to participate in simultaneous transmissions, and aligns their transmission time to maximize the overall link Packet Reception Ratio (PRR), under the condition of maximal system robustness. Extensive experiments in real testbeds demonstrate that Triggercast significantly improves PRR from 5 to 70 percent with seven concurrent senders. We also demonstrate that Triggercast provides 1.3χ PRR performance gains in average, when it is integrated with existing data forwarding protocols.

Constructive Interference (CI) proposed in the existing work (e.g., A-MAC [1], Glossy [2]) may degrade the packet reception performance in terms of Packet Reception Ratio (PRR) and Received Signal Strength Indication (RSSI). The packet reception performance of a set of nodes transmitting simultaneously might be no better than that of any single node transmitting individually. In this paper, we redefine CI and propose TriggerCast, a practical wireless architecture which ensures concurrent transmissions of an identical packet to interfere constructively rather than to interfere non-destructively. CI potentially allows orders of magnitude reductions in energy consumption and improvements in link quality. Moreover, we for the first time present a theoretical sufficient condition for generating CI with IEEE 802.15.4 radio: concurrent transmissions with an identical packet should be synchronized at chip level. Meanwhile, co-senders participating in concurrent transmissions should be carefully selected, and the starting instants for the concurrent transmissions should be aligned. Based on the sufficient condition, we propose practical techniques to effectively compensate propagation and radio processing delays. TriggerCast has 95thpercentile synchronization errors of at most 250ns. Extensive experiments in practical testbeds reveal that TriggerCast significantly improves PRR (from 5% to 70% with 7 concurrent senders, from 50% to 98.3% with 6 senders) and RSSI (about 6dB with 5 senders).