Federated learning (FL) is a distributed machine learning technique enabling multiple clients to jointly train a global model while preserving the privacy of their non-IID (non-independent and identically) data. However, traditional FL approaches require clients to use the same model structure as the global model, which is not suitable for scenarios where clients need to train heterogeneous local models with different architectures, known as Heterogeneous Federated Learning (HFL). Current HFL approaches often exchange all features of local data through the interaction of the client model and the server model. The local data from different clients may present similar generic features, and sharing them over clients may hinder the effective learning of personalized features which truly leads to non-IID distributions. To facilitate the effective information exchange between the server and client while maintaining efficient communication and computation, we propose a novel Heterogeneous Federated learning method (HFedPFS) based on Personalized data Feature Sharing. We designed two significant patterns for this algorithm: (1) A Feature Perception Network (FPN) separates generic and personalized features at each client. (2) A global homogeneous adapter processes the personalized features, enabling effective bidirectional feature exchange. HFedPFS outperforms six state-of-the-art HFL methods on two datasets, improving accuracy by up to 3.54 and 4.17 percentage points in homogeneous and heterogeneous scenarios, respectively while reducing training time by 29.7%.

This paper aims to better understand the performance differences between FPGAs and GPUs. We intentionally begin with a widely used GPU-friendly benchmark suite, Rodinia, and port 15 of the kernels onto FPGAs using HLS C. Then we propose an analytical model to compare their performance. We find that for 6 out of the 15 ported kernels, today's FPGAs can provide comparable performance or even achieve better performance than the GPU, while consuming an average of 28% of the GPU power. Besides lower clock frequency, FPGAs usually achieve a higher number of operations per cycle in each customized deep pipeline, but lower effective parallel factor due to the far lower off-chip memory bandwidth. With 4x more memory bandwidth, 8 out of the 15 FPGA kernels are projected to achieve at least half of the GPU kernel performance.

The notorious power wall has significantly limited the scaling for general-purpose processors. To address this issue, various accelerators, such as GPUs and FPGAs, emerged to achieve better performance and energy-efficiency. Between these two programmable accelerators, a natural question arises: which applications are better suited for FPGAs, which for GPUs, and why?

The mobile smart devices are becoming more and more powerful, and they have been pervasively applied in crowdsensing as effective tools to solve large-scale sensing tasks in urban areas. However, some crowdsensing tasks may bring high network traffic costs to participants using 3G/4G network. In this paper, a novel data diffusing and transmission method is proposed in Crowdsensing based on city Public Traffic System (PTS). This method makes full use of the advantages that the bus has predictable trajectory, wide coverage area, fast moving speed and long contact duration among passengers, so as to realize the rapid transmission of large-scale sensed data. Firstly, we design a data diffusing and transmission model in PTS, and emphatically discuss the Multi-data diffusion and transmission in budget constraints. Secondly, we propose a new algorithm called BUI-BA (Backhaul Utility Increment Based Backhaul Algorithm), which is able to transmit multi- data at the same time. And the algorithm is explained in detail with an example. Finally, the performance of BUI-BA is evaluated with comparisons to Greedy and effSense from aspects like overall transmission utility, fairness, success rate of transmission and transmission redundancy. The result has proved that BUI-BA has a better overall performance and can achieve a tradeoff between overall transmission utility and transmission redundancy, and save more network traffic costs and some other resources for mobile nodes.