On cooperative wireless relaying: A joint routing and scheduling flow-based framework

Abstract

Abstract-We investigate the impact of cooperative relaying used to create virtual multipoint-to-point links (as opposed to conventional multihop relaying) on the throughput optimal configuration of a wireless network. We achieve this by formulating a cross-layer framework for a joint routing and scheduling problem with cooperative relaying. We consider a general case, where cooperation is allowed between any pair of nodes in a given network. We optimally solve this formulation for max-min throughput in mesh-like networks of medium size and quantify gains for key performance metrics. We establish that, contrary to popular belief, cooperative relaying provides performance gains in a mid-size network surprisingly rarely. Moreover, if gains can be obtained, these gains are typically only marginal. We quantify these gains and provide engineering insights based on numerical results for 200 random realizations of a network with 16 nodes.

Achieving Optimal Throughput in Cooperative Wireless Multihop Networks With Rate Adaptation and Continuous Power Control

“…Our framework is generic in the sense that it can be used to obtain a throughput-optimal configuration of the network that employs any type of cooperative combining technique or system architecture such as heterogeneous (e.g., [12]), ad-hoc, etc. In this paper, we extend our previous work [13] by allowing each node in the network to use rate adaptation and CPC.…”

“…The use of cooperative relaying with power control was also studied in cognitive networks, [21], but only for a three-terminal network. Our early work [13] was among the first to study cooperative relaying jointly with multi-path routing, scheduling, and spatial reuse under a physical SINR-based interference model. It was assumed that each node used the same transmit power and same unique rate.…”

“…By formulating the flow-based cross-layer framework as a linear program and optimally solving it, we quantified the end-to-end performance gains that can be achieved in wireless mesh networks with cooperative relaying. It was established in [13] that in single-rate and single-power networks, cooperative relaying does not always provide a gain in throughput or connectivity and if an improvement can be achieved then the gain is only marginal.…”

“…Link scheduling constraints are given in (11) and (13). Specifically, constraints in (11) restrict the total amount of flow that can be scheduled over a link to its link rate capacity r( ) s∈I k α s 1 { ∈s} .…”

“…Typically in practical wireless networks, nodes need to support flows with unequal rates. For this reason, we formulate problem [M] with the flow weighted constraints (12), where w f denotes the weight of flow f with respect to R. Constraints (13) state that the ISets must be scheduled over a unit period of time.…”