Switching off base stations is a popular approach to improving the energy efficiency a network. The challenge is how to determine which and how many base stations to switch off given a network configuration while ensuring that the network coverage is not compromised. Different algorithms have been formulated to solve this challenge ranging from load-aware algorithms to random selection. Results show that load-aware algorithms have better performance. Most algorithms assume universal frequency reuse by all base stations even in the case of heterogeneous networks. In this paper we assume an indoor, femtocell network where subcarrier allocation is based on an existing subcarrier allocation technique which ensures maximum reuse. A load-aware base station switch-off algorithm is developed that uses the allocation matrix when all base stations are active and an estimate of required subcarriers by each base station to determine which base stations to switch off. It will be shown that even without any power control, the proposed technique can result in significant energy savings for lower femtocell densities.

No abstract

Summary IEEE 802.11ah is a WiFi standard developed to address requirements associated with the heterogeneous nature of the Internet of Things. It uses the sub‐1 GHz bands to enable long‐range transmissions for low‐power devices. However, long‐range coverage exacerbates the hidden node problem. To mitigate this problem, sectorization and grouping have been proposed by the standard. In this paper, hierarchical clustering and a modified Welsh‐Powell algorithm are combined to create hidden node‐free groups in a sectorized wireless local area network. An analytical model is developed for a deterministic backoff technique that uses association identifiers to set the backoff in order to eliminate collisions. The performance of the deterministic backoff technique is compared with that of existing schemes. Our results show that this approach results in higher throughput and lower latency than other deterministic backoff techniques.

No abstract

No abstract