Energy efficient content-based image retrieval for mobile systems

Hong, Yu-Ju; Kumar, Karthik; Lu, Yung-Hsiang

doi:10.1109/iscas.2009.5118095

Cited by 31 publications

(28 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method has been proven effective as seen in previous studies [5][6]. In addition, there already are various studies [7][8][9][10] on estimating and calculating energy consumption, and the suggested algorithms in each study show optimal performance for a specifically targeted mobile program. However, the downside is that this performance cannot be expected with mobile application programs other than the ones they target.…”

Section: Introductionmentioning

confidence: 80%

“…Hong et al [10] proposed a formula to calculate energy consumption according to the number of images based on image extraction time, database fetch time, and image search time in content-based image retrieval (CBIR). Therefore, the advantage of their method is that energy consumption can be calculated dynamically during runtime regardless of changes in the number of images.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

History-based dynamic estimation of energy consumption for mobile applications

Cheng

Lee

2014

16th International Conference on Advanced Communication Technology

View full text Add to dashboard Cite

Abstract-Mobile cloud computing is one of the primary research areas that aims to overcome the problem of limited mobile device energy. For this purpose, computational tasks of mobile applications that consume a high amount of energy are offloaded to the cloud. In order to perform effective computation offloading, the energy consumed by application programs in mobile devices and the cloud needs to be estimated first. Previous studies have proven that energy consumption estimators showed optimal performance with targeted mobile application programs. However, optimal performance has yet to be achieved with other mobile application programs. Thus, this paper proposes a dynamic energy consumption estimator in which a mobile application program selects an appropriate energy consumption estimator for its own benefit. In addition, performance and effectiveness of the proposed dynamic energy consumption estimator are validated through experiments.

show abstract

Section: Introductionmentioning

confidence: 80%

Section: Related Workmentioning

confidence: 99%

History-based dynamic estimation of energy consumption for mobile applications

Cheng

Lee

2014

16th International Conference on Advanced Communication Technology

View full text Add to dashboard Cite

show abstract

“…Cloud-based approach aims on leveraging the information and computation resources on cloud server to reduce workload on mobile devices so as to reduce energy consumption on mobile devices [9], [10], [11]. However, the goal of our work is not only reducing resource consumption on mobile device but also providing timely response on mobile devices.…”

Section: A Related Workmentioning

confidence: 99%

Elastic Computation Middleware for Interactive Wearable Devices in Cyber-Physical Systems

Shih

Wang

Yang

et al. 2015

2015 IEEE 3rd International Conference on Cyber-Physical Systems, Networks, and Applications

View full text Add to dashboard Cite

The contribution of this work is to enable time-sensitive, computation intensive applications on wearable interactive devices. To achieve this goal, this work developed an elastic computation middle-ware to federate computation resources on wearable devices, mobile devices and the connected devices that can be connected via local-and wide-area networks. With the middle-ware, the mobile applications receive mandatory/-critical results before individual deadlines and optional/noncritical results when computation resources are available on either local or connected devices. Compared with multi-tier and resource-aware mobile computation frameworks, the elastic computation middle-ware does not only offload computation workloads to connected devices but also make use of the computation and storage resources on connected devices to enhance the computation results. In addition, the mobility of the devices are taken into account to avoid deny of service attack and fruitless workloads, whose requesters are disconnected at the time of completion. Our evaluation shows that the costperformance ratio of the middle-ware is the best among all the compared algorithms. To be specific, the cost-performance of the developed algorithm can be at least two times better than that of compared algorithms.

show abstract

“…A timeout mechanism is used for the offloading decision in [26] to reduce the energy consumption on battery-powered systems. Hong et al [7] propose an offloading strategy to save energy for mobile systems. The strategy is used for performing content-based image retrieval.…”

Section: Related Work For Computation Offloadingmentioning

confidence: 99%

Computation Offloading for Frame-Based Real-Time Tasks with Resource Reservation Servers

Toma

Chen

2013

2013 25th Euromicro Conference on Real-Time Systems

View full text Add to dashboard Cite

Computation offloading concept has been recently adopted to improve the performance of embedded systems by moving some computation-intensive tasks (partially or wholly) to a powerful remote server. In this paper, we consider a computation offloading problem for frame-based real-time tasks, in which all the tasks have the same arrival time and the same relative deadline/period, by adopting the total bandwidth server (TBS) as resource reservations in the server side (remote execution unit). We prove that the problem is N P -complete and propose two algorithms in this paper. The first algorithm is a greedy algorithm with low complexity and provides a quick heuristic approach to decide which tasks to be offloaded and how the tasks are scheduled. The maximum finishing time of the solution derived from the greedy algorithm is at most twice of the finishing time (makespan, maximal on the client and on the server) of any schedule. The second algorithm is a dynamic programming approach, which builds a three-dimensional table and requires pseudo-polynomial time complexity, to make an optimal decision for computation offloading. The algorithms are evaluated with a case study of a surveillance system and synthesized benchmarks.

show abstract

Energy efficient content-based image retrieval for mobile systems

Cited by 31 publications

References 8 publications

History-based dynamic estimation of energy consumption for mobile applications

History-based dynamic estimation of energy consumption for mobile applications

Elastic Computation Middleware for Interactive Wearable Devices in Cyber-Physical Systems

Computation Offloading for Frame-Based Real-Time Tasks with Resource Reservation Servers

Contact Info

Product

Resources

About