In this paper, we propose novel low-energy static and dynamic scheduling algorithms for the heterogeneous Body Area Network (BAN) systems, where task graphs have deadlines (timing constraints) and precedence relations...
详细信息
ISBN:
(纸本)9781424415243
In this paper, we propose novel low-energy static and dynamic scheduling algorithms for the heterogeneous Body Area Network (BAN) systems, where task graphs have deadlines (timing constraints) and precedence relationships to satisfy. Our proposed algorithms, with low computational complexities, use the novel "path information track-and-update" scheme to distribute slack over tasks such that the overall energy consumption is minimized, and an observer mechanism to guarantee the application constraints. Our dynamic scheduling algorithm utilizes the results from the static scheduling algorithm and attempts to aggressively reduce the energy consumption. Simulations for the task graph for a typical BAN application show that our scheduling algorithms achieve better energy savings with less than 5% of the computational time, compared with the recent heterogenous multiprocessor scheduling algorithms.
Nowadays, real-time applications involving voice over IP (VoIP) and video conferencing are gaining an increasing popularity. MPLS is considered as the most used packet switching technology that ensures QoS for real-ti...
详细信息
ISBN:
(纸本)9781538642665
Nowadays, real-time applications involving voice over IP (VoIP) and video conferencing are gaining an increasing popularity. MPLS is considered as the most used packet switching technology that ensures QoS for real-time applications with efficient use of network resources. VPN is used to connect different sites which belong to the same corporation to work as a single site. WiMAX is an emerging broadband wireless technology that is providing last-mile solutions with efficient QoS methods. Real time applications need special QoS requirements: delay, throughput, and jitter. Powerful scheduling algorithms are essential for satisfying these requirements. In this paper, QoS performance evaluation is presented based on scheduling algorithms over MPLS/VPN/WiMAX networks. We used the delay, throughput and jitter as performance metrics to discuss the effect of each scheduling algorithm in satisfying the requirements of VoIP and video conferencing.
On-demand wireless data broadcast is an efficient way to disseminate data to a large number of mobile users. In many applications, such as stock quotes and flight schedules, users may have to download multiple data it...
详细信息
ISBN:
(纸本)9781467399531
On-demand wireless data broadcast is an efficient way to disseminate data to a large number of mobile users. In many applications, such as stock quotes and flight schedules, users may have to download multiple data items per request. However the multi-item request scheduling has not yet been thoroughly investigated for on-demand wireless data broadcasts. In this paper, we step-up on investigating this problem from viewpoint of theory and simulation. We develop a two-stage scheduling scheme to arrange the requested data items with the objective of minimizing the average access latency. The first stage is to select the data items to be broadcast in the next time period and the second stage is to schedule the broadcasting order for the data items selected in the first stage. We develop algorithms for the two stages respectively and analyze them both theoretically and practically. We also compare the proposed algorithms with other well known scheduling methods through simulation. The theoretical findings and simulation results reveal that significantly better access latency can be obtained by using our scheduling scheme rather than its competitors.
The paper presents some models and algorithms for the nonlinear optimization problem of multi-objects movement scheduling to synchronize their movement as well as properties of presented algorithms. Similarities and d...
详细信息
ISBN:
(纸本)9783642049200
The paper presents some models and algorithms for the nonlinear optimization problem of multi-objects movement scheduling to synchronize their movement as well as properties of presented algorithms. Similarities and differences between defined problems and the classical tasks scheduling problem on parallel processors are shown. Two algorithms for synchronous movement scheduling are proposed and their properties are considered. One of the algorithm is based on dynamic programming approach and the second one uses some approximation techniques. Theoretical and experimental analysis of complexity and effectiveness of the algorithms as well as their practical usefulness are discussed.
Many research efforts have been done in the domain of static scheduling algorithms based on DAG. However, most of these literatures assume that all processors are fully connected and receive communication data concurr...
详细信息
ISBN:
(纸本)0769526365
Many research efforts have been done in the domain of static scheduling algorithms based on DAG. However, most of these literatures assume that all processors are fully connected and receive communication data concurrently, while ignoring the contentions and delays on network links in real applications, which leads to low efficiency. This paper focuses on the issue of edge scheduling for dependent task set in parallel and distributed environment. Combined with conventionally efficient heuristics, two contention-aware scheduling algorithms are proposed in the paper: OIRSA (Optimal Insertion Hybrid scheduling Algorithm) and BBSA (Bandwidth Based scheduling Algorithm). Both the proposed algorithms start from the inherent characteristic of the edge scheduling problem, and select route paths with relatively low network workload to transfer communication data by modified routing algorithm. OISHA optimizes the start time of communication data transferred on links in form of theorems. BBSA exploits bandwidth of network links fully to transfer communication data as soon as possible. Therefore, the makespan yielded by our algorithms can be reduced efficiently. Moreover, the proposed algorithms adapt to not only homogeneous systems but also heterogeneous systems. The experiment results indicate that the proposed algorithms obviously outperform other algorithms so far in terms of makespan.
The use of orthogonal frequency division multiple access (OFDMA) in Long Term Evolution (LTE) and LTE-Advanced systems facilitates the potential for scheduling cell users on orthogonal time-frequency resource blocks s...
详细信息
ISBN:
(纸本)9781479944491
The use of orthogonal frequency division multiple access (OFDMA) in Long Term Evolution (LTE) and LTE-Advanced systems facilitates the potential for scheduling cell users on orthogonal time-frequency resource blocks selectively. This paper identifies several frequency selective scheduling (FSS) algorithms and studies their performance and optimality under certain identified constraints in practice. The performance is studied under the limiting factors of cell load, user mobility, the number of users per cell, data traffic characteristics, and the LTE standards constraint of using a single modulation and coding scheme (MCS) across assigned resource blocks. To address the single MCS restriction, a dynamic Proportional Fair (PF) scheduling algorithm is developed to achieve optimal allocation under this constraint. The gain either in signal-to-interference-plus-noise ratio (SINR) or cell throughput achieved from these algorithms is statistically quantified using detailed LTE system level simulations.
作者：
Lee, JuyulETRI
Wireless Telecommun Dept Taejon 305700 South Korea
Fairness consideration in the wireless resource scheduling is important for enhancing the quality of user experience, despite the consequent throughput loss. This paper analyzes the associated penalty in obtaining a c...
详细信息
ISBN:
(纸本)9781467325790
Fairness consideration in the wireless resource scheduling is important for enhancing the quality of user experience, despite the consequent throughput loss. This paper analyzes the associated penalty in obtaining a certain fairness criteria for multiple antenna broadcast channels. We define a fairness penalty as the achievable sum-rate difference in the regime of high SNR. We show that the penalty for the max-product fair scheduling criterion is negligible and derive the penalty for the max-min fair scheduling criterion. We analyze ergodic behaviors for the Rayleigh-fading channel both when the average SNRs are homogeneous (identical) and when the SNRs are heterogeneous (different). For the homogeneous case, it is shown that the penalty diminishes as the number of transmit antennas increases. For the heterogeneous case, there is an inevitable bias term which is associated with the users' SNR distribution.
A very efficient computing environment is provided by cloud computing where the customers or several tenants are in need of multiple resources to be provided as a service over the internet. The utilization of resource...
详细信息
ISBN:
(纸本)9781509027170
A very efficient computing environment is provided by cloud computing where the customers or several tenants are in need of multiple resources to be provided as a service over the internet. The utilization of resources is to be scheduled efficiently so that it helps in reducing the time for task completion. This is task scheduling which is most essential and important part in cloud computing environment. In task scheduling allocation of certain tasks to particular resources at a particular time instance is done. There are different techniques that are proposed to solve the problems of task scheduling. This paper discusses about the study of various resource scheduling algorithms in a cloud computing environment.
We consider a centralized wireless architecture in which multiple cells are connected to a centralized unit via a shared x-haul. When computing a schedule for this architecture we must lake into account capacity const...
详细信息
ISBN:
(纸本)9781538649824
We consider a centralized wireless architecture in which multiple cells are connected to a centralized unit via a shared x-haul. When computing a schedule for this architecture we must lake into account capacity constraints on both the air interface access link and the x-haul. We evaluate three scheduling approaches that differ based on the degree of coupling between the air interface/x-haul and on whether queuing is allowed at the access link. It is known that we can optimize the geometric mean of throughput via a standard backpressure approach in which the air interlace and x-haul scheduling is coupled by a queue at the access link. We demonstrate that with a joint scheduling approach we can obtain similar performance with zero queuing at the air interface (and hence zero delay).
A semi analytical model is presented to assess the impact of traffic conditions on the performance of scheduling algorithms within a queuing framework. Three different scheduling algorithms are considered and their pe...
详细信息
ISBN:
(纸本)9781424411894
A semi analytical model is presented to assess the impact of traffic conditions on the performance of scheduling algorithms within a queuing framework. Three different scheduling algorithms are considered and their performance for two different traffic conditions is compared. These two traffic conditions correspond to the cases of light traffic for few users and heavier traffic for a larger number of users, respectively. Results indicate that in the case of heavier traffic, the SNR and angular separation based scheduling algorithm has a better performance than some of the other standard scheduling algorithms while for light traffic the difference in the performance is not significant.
暂无评论