Content-Aware Proactive Caching for Backhaul Offloading in Cellular Network

Doan, Khai Nguyen; Nguyen, Thang Van; Quek, Tony Q. S.; Shin, Hyundong

doi:10.1109/twc.2018.2806971

Cited by 68 publications

(59 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we present our simulation results to show the performance of the proposed probabilistic content request model denoted by "PGP". To compare our results, we use as benchmark methods the ones suggested in [18] and [14]. Specifically, the authors in [18] used several regression methods in order to predict the popularity of contents based on their similarity with the seen contents.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…To compare our results, we use as benchmark methods the ones suggested in [18] and [14]. Specifically, the authors in [18] used several regression methods in order to predict the popularity of contents based on their similarity with the seen contents. In the simulations, we use the support vector regression (SVR) with the radial basis kernel which they showed that it provides the best prediction performance for unseen contents.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…where log p (D|θ) is independent of the variational parameter ϕ and therefore the minimization problem (17) is equivalent to minimizing L (ϕ, θ). Intuitively, minimizing (18) incentivizes distributions that place high mass on configurations of the approximate distribution that explain the observations well (the first term) and also rewards distributions that are entropic, meaning that they maximize uncertainty by spreading their mass on many configurations (the second term). Now, given the simplified posterior distribution (16), the objective is to find the optimal value of θ.…”

Section: A Low Complexity Inference Methodsmentioning

confidence: 99%

“…On the other hand, content features (e.g topic categories) can easily and cheaply be obtained from the content server without jeopardizing users' privacy. In [18], the authors used the content features to predict the popularity of unseen contents. More recently, the content features are used to learn the userlevel preferences [19].…”

Section: A Related Workmentioning

confidence: 99%

“…In other words, it acts as a regularizer and as a result it provides a better estimation accuracy. Therefore, our model is much more flexible for popularity prediction than the method proposed in [18] where features are used only to predict the popularity of unseen content. Second, similar to [18] but in a more efficient way, our model can predict the popularity of unseen contents where statistical information is not available in advance.…”

Section: B Contributionsmentioning

confidence: 99%

See 4 more Smart Citations

A Bayesian Poisson–Gaussian Process Model for Popularity Learning in Edge-Caching Networks

et al. 2019

View full text Add to dashboard Cite

Edge-caching is recognized as an efficient technique for future cellular networks to improve network capacity and user-perceived quality of experience. To enhance the performance of caching systems, designing an accurate content request prediction algorithm plays an important role. In this paper, we develop a flexible model, a Poisson regressor based on a Gaussian process, for the content request distribution. The first important advantage of the proposed model is that it encourages the already existing or seen contents with similar features to be correlated in the feature space and therefore it acts as a regularizer for the estimation. Second, it allows to predict the popularities of newly-added or unseen contents whose statistical data is not available in advance. In order to learn the model parameters, which yield the Poisson arrival rates or alternatively the content popularities, we invoke the Bayesian approach which is robust against over-fitting. However, the resulting posterior distribution is analytically intractable to compute. To tackle this, we apply a Markov Chain Monte Carlo (MCMC) method to approximate this distribution which is also asymptotically exact. Nevertheless, the MCMC is computationally demanding especially when the number of contents is large. Thus, we employ the Variational Bayes (VB) method as an alternative low complexity solution. More specifically, the VB method addresses the approximation of the posterior distribution through an optimization problem. Subsequently, we present a fast block-coordinate descent algorithm to solve this optimization problem. Finally, extensive simulation results both on synthetic and real-world datasets are provided to show the accuracy of our prediction algorithm and the cache hit ratio (CHR) gain compared to existing methods from the literature.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Section: Simulation Resultsmentioning

confidence: 99%

Section: A Low Complexity Inference Methodsmentioning

confidence: 99%

Section: A Related Workmentioning

confidence: 99%

Section: B Contributionsmentioning

confidence: 99%

See 3 more Smart Citations

A Bayesian Poisson–Gaussian Process Model for Popularity Learning in Edge-Caching Networks

et al. 2019

View full text Add to dashboard Cite

show abstract

Backhauling for Small‐Cell Networks

Song

Shan

2019

Wiley 5G Ref

View full text Add to dashboard Cite

Densely deploying small cells is an attractive solution to the explosion of network traffic. However, it is well known that backhaul is a major challenge limiting the improvement of network performance. In this article, we first introduce the characteristics of different wired or wireless backhaul technologies, including fiber, xDSL, sub‐6 GHz, microwave, and millimeter wave (mmWave). As the main performance factors, throughput, latency, energy efficiency, and cost models are analyzed to provide useful guidance for the selection and deployment of backhaul links. Next, we survey existing works on joint optimization of access and backhaul links. Base station (BS) association schemes and caching techniques are also studied for their importance to further improve backhaul performance. Finally, research challenges on backhauling for small‐cell networks are summarized to inspire future research directions.

show abstract

Wireless Edge Caching with Short-Term Content Popularity Prediction for Cost Minimization

Tran

2022

30th Biennial Symposium on Communications 2021

View full text Add to dashboard Cite

Content-Aware Proactive Caching for Backhaul Offloading in Cellular Network

Cited by 68 publications

References 34 publications

A Bayesian Poisson–Gaussian Process Model for Popularity Learning in Edge-Caching Networks

A Bayesian Poisson–Gaussian Process Model for Popularity Learning in Edge-Caching Networks

Backhauling for Small‐Cell Networks

Wireless Edge Caching with Short-Term Content Popularity Prediction for Cost Minimization

Contact Info

Product

Resources

About