Interference-aware fair rate control in wireless sensor networks

Abstract

In a wireless sensor network of N nodes transmitting data to a single base station, possibly over multiple hops, what distributed mechanisms should be implemented in order to dynamically allocate fair and efficient transmission rates to each node? Our interferenceaware fair rate control (IFRC) detects incipient congestion at a node by monitoring the average queue length, communicates congestion state to exactly the set of potential interferers using a novel low-overhead congestion sharing mechanism, and converges to a fair and efficient rate using an AIMD control law. We evaluate IFRC extensively on a 40-node wireless sensor network testbed. IFRC achieves a fair and efficient rate allocation that is within 20-40% of the optimal fair rate allocation on some network topologies. Its rate adaptation mechanism is highly effective: we did not observe a single instance of queue overflow in our many experiments. Finally, IFRC can be extended easily to support situations where only a subset of the nodes transmit, where the network has multiple base stations, or where nodes are assigned different transmission weights.

Congestion Control in Wireless Sensor Networks Based on the Bird Flocking Behavior

“…rate adaptation [12], [13], [20] multi-path routing [21], [22], [23]), topology control (e.g. clustering formation [24]), and network resource management (e.g.…”

“…Some conventional CC approaches [12], [13], [21], [20] are based on rate control that alleviates congestion by throttling the injection of traffic in the network. However, rate control attempts to decrease the reporting rate of nodes during (transient or persistent) congestion phenomena and may result in the deterioration of the offered quality of service, perhaps when needed the most.…”

Pushback: A hidden Markov model based scheme for energy efficient data transmission in sensor networks

“…Fusion [15] uses the concept of hop-by-hop flow control and prioritized MAC along with token buckets to control the packet rate at each node. IFRC [16] and RAIN [17] use variations of the AIMD (Additive Increase Multiplicative Decrease) and the back-pressure mechanisms. Various schemes have been proposed, which select the backoff counter [30] and the backoff window [31], [32] based on the channel state.…”

Locally scheduled packet bursting for data collection in wireless sensor networks

“…Various approaches Email addresses: rsliu@cse.ohio-state.edu (Ren-Shiou Liu), fank@cse.ohio-state.edu (Kai-Wei Fan), prasun@cse.ohio-state.edu (Prasun Sinha). have been proposed to mitigate this problem, such as improved MAC layer designs [1,2] and back-pressure techniques at the link layer [3][4][5][6]. In [1], a hybrid TDMA/CSMA approach is proposed to address congestion near the sink.…”

“…In addition, the computation of the TDMA schedules is expensive in dynamic environments where the traffic sources change with time. Back-pressure based mechanisms for congestion control [3][4][5][6] operate over the MAC layer to maintain the queue size at acceptable levels to avoid queue drops. As these mechanisms are not integrated into the MAC layer where congestion is first observed, their impact on performance improvement is limited.…”