Maximizing Broadcast Throughput Under Ultra-Low-Power Constraints

Abstract

Wireless object tracking applications are gaining popularity and will soon utilize emerging ultra-low-power deviceto-device communication. However, severe energy constraints require much more careful accounting of energy usage than what prior art provides. In particular, the available energy, the differing power consumption levels for listening, receiving, and transmitting, as well as the limited control bandwidth must all be considered. Therefore, we formulate the problem of maximizing the throughput among a set of heterogeneous broadcasting nodes with differing power consumption levels, each subject to a strict ultra-lowpower budget. We obtain the oracle throughput (i.e., maximum throughput achieved by an oracle) and use Lagrangian methods to design EconCast -a simple asynchronous distributed protocol in which nodes transition between sleep, listen, and transmit states, and dynamically change the transition rates. We also show that EconCast approaches the oracle throughput. The performance is evaluated numerically and via extensive simulations and it is shown that EconCast outperforms prior art by 6x -17x under realistic assumptions. Finally, we implement EconCast using the TI eZ430-RF2500-SEH energy harvesting nodes and experimentally show that in realistic environments it obtains 57% -77% of the achievable throughput.

Wake-up Radio-enabled Intermittently-powered Devices for Mesh Networking: A Power Analysis

“…5 shows that, when the overhead of each wakeup is 0, the maximum G is achieved at R opt = P EH /(2P Li E Tx ), when energy is split equally for transmitting and receiving. This result matches other authors simulation that does not consider the cost of waking up [11]. Above R opt , the reduction in listening energy and time outweighs the extra transmissions.…”

“…• We demonstrate, using a model derived from commercial radio transmitters and typical WuRxs, the best configurations for different network densities, to achieve the highest goodput. Our mathematical model shows that, without the impact of wake-ups, allocating energy equally for transmission and for listening maximizes the goodput (confirming simulation results of EconCast [11]). However, when the wake-up impact is also considered, the proportion of energy used for transmitting should be reduced.…”

“…1, which is an area unexplored by existing research. Previously, EconCast [11] considers intermittently connected mesh nodes and uses a Markov model to determine the broadcast throughput. However, it does not consider using WuRxs, nor the additional cost of each wake-up, with additional receiving and processing consumption, rather it only considers radios that have constant power consumption during the entire receive time (including listening).…”

Neighbor Discovery Latency in BLE-Like Protocols

“…In the Example from Figure 6, all hitting partitions (1)(2)(3)(4)(5)(6) in Iteration 1 will incur a partial latency of d p,0 = d s · p · 6 k=1 k · σ 0 , with p = 1 Ts , and the worst-case latency among all hitting parts is 6 · σ 0 . An advertising packet that is sent within a missing partition (a-g) will "overtake" the scan window S 1 after the corresponding number of advertising intervals.…”

“…Therefore, they make use of one of the following approaches. Birthday Protocols: These protocols require every node to switch between a low-power-state, a sendingstate and a receiving-state at random points in time with given transition probabilities [4], [5], [6]. Whereas good average latencies can be achieved, the main drawback is the lack of guarantees on upper latency bounds.…”