Tools for Design, Implementation and Verification of Emerging Information Technologies

With the increasing complexity of blockchain network business requirements, there is a growing demand for interconnection mechanisms among different blockchains, leading to the emergence of cross-chain technology. This article discusses three mainstream cross-chain technologies: the notary mechanism, hash time lock, and side chain/relay mode, and highlights their limitations. In order to address issues such as low efficiency in relay chain reorganization and instability in cross-chain systems within the relay mode involving node turnover, this paper proposes a node trust-based cross-chain relay scheme. The scheme includes the construction of a trust model for blockchain nodes, the design of a weighted random election algorithm for relay nodes, and the development of a complete cross-chain transaction process. Simulation experiments are conducted to demonstrate the performance of the proposed scheme.

As the cryptocurrency universe continues to expand at an unprecedented pace, efficient and routine transactions have become a critical necessity. Yet, the current transaction processing capabilities of many blockchains fail to meet the burgeoning demands of retail payments. While off-chain scaling solutions present promising alternatives to augment blockchain throughput and uphold compatibility with established blockchains, they often impose impractical burdens on users or experience significant payment latency. This study introduces a hardware-driven off-chain payment system designed to accommodate contingent payments. Merchants can accept payments instantly without waiting for transaction confirmation. Our assessment reveals that this system can achieve a peak throughput of approximately 10000 payments per second, representing an 84-fold improvement over traditional Ethereum transactions. This hardware-driven solution holds significant promise for instant and collateral-efficient transactions, effectively unlocking the untapped potential of instant payments with high throughput.

The field of music composition has seen significant advancements with the introduction of artificial intelligence (AI) techniques. However, traditional cloud-based approaches suffer from limitations such as latency and network dependency. This survey paper explores the emerging concept of edge intelligence and its application in music composition. Edge intelligence leverages local computational resources to enable real-time and on-device music generation, enhancing the creative process and expanding accessibility. By examining various aspects of music composition, including melody creation, harmonization, rhythm generation, arrangement and orchestration, and lyric writing, this paper showcases the potential benefits of incorporating edge intelligence. It also discusses the challenges and limitations associated with this paradigm, such as limited computational resources and privacy concerns. Through a review of existing AI-based music composition tools and platforms, examples of edge intelligence in action are highlighted. The survey paper concludes by emphasizing the transformative potential of edge intelligence in revolutionizing the field of music composition and identifies future research opportunities to further advance this promising domain.

In the realm of music composition, sentiment plays a pivotal role in connecting compositions with their audience, evoking emotions and memories. With the rapid evolution of artificial intelligence (AI), there exists a burgeoning interest in utilizing AI for sentiment analysis in various domains, including textual data, social media, and film. This paper delves into the novel application of AI-driven sentiment analysis specifically tailored for music composition. Leveraging diverse music datasets across multiple genres and eras, we introduce an innovative methodology that breaks down music into foundational features such as melody, rhythm, timbre, and harmony. Through the application of advanced AI techniques, including neural networks and Long Short-Term Memory (LSTM) models, we aim to accurately map these features to a wide spectrum of sentiments. Our results showcase not only the potential accuracy and precision of our chosen models but also the richness of music compositions they can produce, underscoring the viability of AI in enhancing the emotional depth of musical works. The implications of this research stretch from aiding composers in creating more resonant pieces to the potential therapeutic applications of AI-composed music, tailored to specific emotional needs.

To minimize the risk of Hydro-Power Plant failure, it’s crucial to detect and precisely repair the damaged components. In this paper, we propose a knowledge graph-based fault diagnosis method for Hydro-Power Plants. Then, the improved BiLSTM-CRF algorithm is developed to recognize entities for fault diagnosis, and the BERT relationship extraction algorithm is designed to construct a fault diagnosis knowledge graph for the Hydro-Power Plant. The real experimental test results validate the proposed methodology.

As a distributed machine learning framework, federated learning addresses the challenges of data isolation and privacy concerns, ensuring that user data remains private during the model training process. However, the privacy-preserving nature of federated learning also makes it has vulnerability to security attacks, particularly in the form of backdoor attacks. These attacks aim to compromise the integrity of the model by embedding a malicious behavior that can be triggered under specific conditions. In our study, aiming to counteract backdoor threats in federated learning, we introduce a new protective mechanism termed zero-knowledge with robust learning (ZKRL). The ZKRL scheme introduces the robust learning rate and non-interactive zero-knowledge proof techniques to filter out malicious model updates and preserve the privacy of the global model parameters of the federated learning process. The extensive experiments conduct on real-world data demonstrate its effectiveness in improving the accuracy on the verification set by 2\(\%\) and significantly reducing the accuracy of backdoor attacks compared to existing state-of-the-art defense schemes. In summary, the proposed ZKRL defense scheme provides a robust solution for protecting federated learning models against backdoor attacks, ensuring the integrity of the trained models while preserving user privacy.

In the realm of distributed machine learning, although federated learning has received considerable attention, it still confronts grave challenges such as user privacy leakage and poisoning attacks. Regrettably, the demands for privacy preservation and protection against poisoning attacks are conflicting. Measures for privacy protection generally assure the indistinguishability of local parameter updates, which conversely complicates the strategy of defending against poisoning attacks by making it harder to identify malicious users. To address these issues, we propose a privacy-preserving and anti-poisoning attack federated learning (PPAPAFL) scheme. This scheme employs the CKKS homomorphic encryption technique for gradient packaging encryption, thus ensuring data privacy. Concurrently, our designed robust aggregation algorithm can effectively resist poisoning attacks, guaranteeing the model’s integrity and accuracy, and is capable of supporting heterogeneous data in a friendly manner. A plethora of comparative experimental results demonstrate that our scheme can significantly improve the model’s accuracy and robustness, drastically reduce the attack success rate, and effectively protect data privacy. In comparison with advanced schemes such as Trum and PEFL, our scheme achieves a 10–50% improvement in model accuracy and reduces the attack success rate to less than 3%.

In recent years, with the development of artificial intelligence, many facial manipulation methods based on deep neural networks have been developed, known as DeepFakes. Unfortunately, DeepFakes are always maliciously used, and if the spread of DeepFakes cannot be controlled in a timely manner, it will pose a certain threat to both society and individuals. Researchers have studied the detection of DeepFakes, but this type of detection belongs to post-evidence collection and still has a certain degree of negative impact. Therefore, we propose a retentive and proactive defense method to protect DeepFakes before malicious operations. The main idea is to train a perturbation generator end-to-end, and introduce the perturbation generated by the perturbation generator into the image to make it adversarial and immune to DeepFakes. White-box experiments on a typical DeepFake manipulation method (facial attribute editing) demonstrate the effectiveness of our proposed method, and a comparison with an adversarial attack PGD proves the superiority of our method in terms of similarity and inference efficiency.

With the remarkable successes of machine learning, it is becoming increasingly popular and widespread. Machine learning as a Service (MLaaS) provided by cloud services is widely utilized to address the challenge of users unable to bear the burden of training machine learning models. However, the privacy issues involved present a significant challenge. Homomorphic encryption, known for its capability to perform efficient operations on ciphertexts, is widely employed in Privacy computing domain. In order to address the security vulnerabilities and excessive communication and computation costs of interactive privacy-preserving neural networks, and in light of the significant time consumption of linear layers and the challenges SIMD HE faces in computing arbitrary nonlinear functions precisely, we propose a non-interactive framework for privacy-preserving neural networks that accelerates linear computations and ensures accurate computation of any non-linear functions. Specifically, we utilize CKKS encryption to enable private neural network inference under floating-point arithmetic. Leveraging the characteristics of both wordwise HE and bitwise HE, we design a non-interactive and fast matrix multiplication scheme, achieving up to 500× acceleration across different matrix dimensions. By transforming various types of homomorphic encryption ciphertexts and employing lookup tables, we realize accurate computation of arbitrary non-linear operations without requiring interaction. Experimental results demonstrate that our framework achieves the same level of accuracy as pre-trained neural network models on plaintext without incurring any additional accuracy loss.