Hash center-based deep hashing methods improve upon pairwise or triplet-based approaches by assigning fixed hash centers to each class as learning targets, thereby avoiding the inefficiency of local similarity optimization. However, random center initialization often disregards inter-class semantic relationships. While existing two-stage methods mitigate this by first refining hash centers with semantics and then training the hash function, they introduce additional complexity, computational overhead, and suboptimal performance due to stage-wise discrepancies. To address these limitations, we propose Center-Reassigned Hashing (CRH), an end-to-end framework that dynamically reassigns hash centers from a preset codebook while jointly optimizing the hash function. Unlike previous methods, CRH adapts hash centers to the data distribution without explicit center optimization phases, enabling seamless integration of semantic relationships into the learning process. Furthermore, a multi-head mechanism enhances the representational capacity of hash centers, capturing richer semantic structures. Extensive experiments on three benchmarks demonstrate that CRH learns semantically meaningful hash centers and outperforms state-of-the-art deep hashing methods in retrieval tasks.

Data compression is an effective way to improve the seismic data transmission efficiency. The features of seismic exploration are long sampling time and large data quantity, so the compression algorithm should achieve high‐fidelity, high‐compression ratio (CR) and low‐compression time. Since the existing compression algorithms cannot meet the requirements of site‐collected seismic data compression, the segmented matching pursuit (SMP) compression algorithm based on new atom dictionary is proposed. This method is based on the principle of MP. A novel‐segmented compression structure is adopted. The modified Morlet wave atom dictionary is designed to replace the previous dictionaries. The results of comparative experiments show that the proposed algorithm makes improvements in CR and compression time with the same fidelity requirement.

One of the solutions to improve mobile Delay Tolerant Network (DTN) performance is to place additional stationary nodes, called throwboxes, to create a greater number of contact opportunities. In this paper, we study a key optimization problem in a time-evolving throwbox-assisted DTN: throwbox selection, to answer the questions such as “how many active throwboxes do I need?” and “which throwboxes should be activated?” We formally define two throwbox optimization problems: min-throwbox problem and k-throwbox problem for time-evolving DTNs modeled by weighted space-time graphs. We show that min-throwbox problem is NP-hard and propose a set of greedy algorithms which can efficiently provide quality solutions for both challenging problems.

Recent advances in ambient energy-harvesting wireless sensor networks (WSNs) technologies have made it possible to power the network by energy generated from the environment and thereby increase its lifetime. Various energy sources including light, vibration and heat can be harvested by sensor nodes. However, time-varying energy harvesting also bring new design challenging for WSNs. In this paper, we study a fault-tolerant topology design problem for an energy-harvesting heterogeneous WSN, where multiple supernodes with rich resources are used to improve the performance. We first model the network as a directed and weighted space-time graph in which both spacial and temporal information are preserved. We then define the fault-tolerant topology problem which aims to build a sparser time-varying structure from the original space-time graph while maintaining k-connectivity for the fault-tolerant purpose. Six different algorithms are proposed to solve the problem. Simulation results demonstrate that our proposed methods can save up to around 80% costs.

Delay tolerant networks (DTNs) recently have drawn much attention from researchers due to their wide applications in various challenging environments. Previous DTN research mainly concentrates on information propagation and packet delivery. However, with possible participation of a large number of mobile devices, how to maintain efficient and dynamic topology becomes crucial. In this paper, we study the topology design problem in a predictable DTN where the time-evolving topology is known a priori or can be predicted. We model such a time-evolving network as a weighted directed space-time graph which includes both spacial and temporal information. Links inside the space-time graph are unreliable due to either the dynamic nature of wireless communications or the rough prediction of underlying human/device mobility. The purpose of our reliable topology design problem is to build a sparse structure from the original space-time graph such that (1) for any pair of devices, there is a space-time path connecting them with a reliability higher than the required threshold; (2) the total cost of the structure is minimized. Such an optimization problem is NP-hard, thus we propose several heuristics which can significantly reduce the total cost of the topology while maintain the “reliable” connectivity overtime. In this paper, we consider both unicast and broadcast reliability of a topology. Finally, extensive simulations are conducted on random DTNs, a synthetic space DTN, and a real-world DTN tracing data. Results demonstrate the efficiency of the proposed methods.

In mobile cloud computing systems, cloud computing has a significant impact on wireless networks. Cloud computing and wireless networks have traditionally been addressed separately in the literature. In this paper, we jointly study the operations of cloud computing and wireless networks in mobile computing environments, where the objective is to improve the end-to-end performances of cloud mobile media delivered through mobile cloud computing systems. Unlike most existing studies on wireless networks, where only the spectrum efficiency is considered, we consider not only the spectrum efficiency in wireless networks but also the pricing information in the cloud, based on which power allocation and interference management in the wireless networks are performed. We formulate the problems encountered in the operations of mobile cloud computing environments, including determining the price to charge for media services, resource allocation, and interference management, as a Stackelberg game model. Moreover, we extend this game model with multiple players through network virtualization technology, and adopt the replicator dynamics method to solve the evolutionary game between the different groups of small cells. Furthermore, a backward induction method is used to analyze the proposed Stackelberg game. Simulation results are presented to show the effectiveness of the proposed techniques.

Recent advances in Delay Tolerant Networks (DTNs) have overcome limitations in connectivity by relying on intermittent contacts between mobile nodes to deliver packets. However, lack of rich contact opportunities still causes poor delivery ratio and long delay of DTN routing. One of the solutions to improve mobile DTN performance is to place additional stationary nodes, called throwboxes, to create a greater number of contact opportunities. In this paper, we study a key optimization problem in a time-evolving throwbox-assisted DTN: k-throwbox placement problem, to answer "where should I put my k throwboxes to optimize the performance?". We model a time-evolving DTN as a weighted space-time graph which includes both spacial and temporal information. We prove that k-throwbox placement problem is NP-hard and propose a set of greedy algorithms which can efficiently provide quality solutions. One of the proposed algorithms can guarantee an (1 - 1/e) approximation for the k-throwbox placement problem. Simulation results based on random time-evolving DTNs and real life DTN traces demonstrate the efficiency of the proposed methods.

In mobile cloud computing systems, cloud computing can have significant impacts on wireless networks. Cloud computing and wireless networks have traditionally been addressed separately in the literature. In this paper, we propose to jointly study dynamic cloud and wireless networks operations so as to improve end-to-end performance in the mobile cloud computing environment. Unlike most existing studies on wireless networks, where only the spectrum efficiency is considered, we consider not only the spectrum efficiency in wireless networks but also the pricing information in the cloud, based on which power allocation and interference management in wireless networks are performed. We formulate the problems of cloud media service price decision, resource allocation and the interference management in the mobile cloud computing environment as a three-level Stackelberg game. A backward induction method is used to analyze the proposed Stackelberg game. Simulation results are presented to show the effectiveness of the proposed scheme.

We consider the problem of online dynamic channel accessing in multi-hop cognitive radio networks. Previous works on online dynamic channel accessing mainly focus on single-hop networks that assume complete conflicts among all secondary users. In the multi-hop multi-channel network settings studied here, there is more general competition among different communication pairs. A simple application of models for single-hop case to multi-hop case with N nodes and M channels leads to exponential time/space complexity O (MN), and poor theoretical guarantee on throughput performance. We thus novelly formulate the problem as a linearly combinatorial multi-armed bandits (MAB) problem that involves a maximum weighted independent set (MWIS) problem with unknown weights. To efficiently address the problem, we propose a distributed channel access algorithm that can achieve 1/ρ of the optimum averaged throughput where each node has communication complexity O (r2+D) and space complexity O (m) in the learning process, and time complexity O (D mρr) in strategy decision process for an arbitrary wireless network. Here ρ = 1 + ε is the approximation ratio to MWIS for a local r-hop network with m <; N nodes, and D is the number of mini-rounds inside each round of strategy decision.