Abstract
In the Fog Radio Access Networks (F-RANs), the local storage and computing capability of Fog Access Points (FAPs) provide new communication resources to address the latency and computing constraints for delay-sensitive applications. To achieve the ultra-low latency, a novel joint user association and channel allocation scheme is proposed in this paper, where the FAPs are clustered from a user-centric perspective. The delay performance is improved regarding both the control signaling procedure and the data transmission procedure. Specifically, the multiple access interference (MAI) between users is analyzed, where the closed-form expression for the effective rate of a typical user with multiple FAP connections and arbitrary interfering users is obtained. With the consideration of MAI, the proposed distributed joint user association and channel allocation algorithm provides a guaranteed delay violation probability. Moreover, the distributed algorithm can be conducted on individual FAPs, whose calculation is simplified by look-up tables. Simulation results show that the proposed algorithm is capable of providing statistical delay performance guarantee including both average delay and delay bound violation probability, which demonstrates its superiority in supporting delay-sensitive applications in F-RANs.
Highlights
The advancement of smart devices, such as smartphones, high-end wearables and connected vehicles, are boosting the demand for massive wireless connections of ultra-low latency for real-time services [1] [2]
Taking advantage of these features, we propose a distributed joint user association and channel allocation algorithm for the Fog Radio Access Networks (F-RANs) in the following part
When adding a new interfering user, (16) shows that the new effective rate function will only add a maximum of K new summation items, while the previously calculated multi-variate H functions can be re-used with an adjusted wlkl
Summary
Proposed in [5], where high power nodes provide seamless coverage and execute the functions of the control plane, while RRHs are providing high-speed data transmissions in the user plane. The decisions on the new users’ association and channel allocation will influence the delay performance of existing users This requires an effective resource coordination between distributed FAPs, where the statistical delay provision for all users in the network should be addressed. Each FAP is capable to make regional decisions based on the pre-cached regional service status and the instantaneous local user information This feature enables the FAPs to scale up for larger networks, and avoids the latency due to the cooperation between FAPs during the user association and channel allocation. Hpm,q,n [·] [19] denotes the uni-variate or multi-variate H function depending on the variables as detailed in Appendix A
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
