Asymptotic detection performance of type-based multiple access in sensor networks

Abstract

The problem of communicating sensor readings over a multinrcess channel for detecting a target is considered. A natural way of communication in target detection is to let sensors simultaneously transmit one of two predetermined frequency tones indicating whether the target is detected or not. Recently, this scheme has been generalized to consider non-hinary sensor observations by letting sensors simultaneously transmit orthogonal wavefarms depending on the value of their observations-Type-Based Multiple Access (TEZMA). TBMA was shown to be asymptotically oprimal in terms of detection-error probability under the idealistic assumptions that the sensor channel gains are identical, and the sensor data are conditionally independent and identically distributed (i.i.d.). In this paper, TBMA is analyzed in a more general framework by considering non-i.i.d. data and non-identical channel gains. An asymptotically optimal detector is proposed and its error-exponents for detection probabilities are characterized using tools from Large deviatiuns theoiy. Numerical simulations are used to demonstrate that the error exponents provide reasonably accurate estimates ofthe performance of TBMA.

Type-Based Random Access for Distributed Detection Over Multiaccess Fading Channels

“…The problem of distributed detection on multiaccess channels are more recent [6], [7], [9], [15]. The transmission scheme used is the so-called TBMA proposed independently by Mergen and Tong [6], [9] and by Liu and Sayeed [7].…”

Energy Efficient Distributed Detection via Multi-Hop Transmission in Sensor Networks

“…Since binary or quadratic quantization is sufficient for most detection scenarios when the number of sensors is large, transmitting histograms instead of the exact observations conserves energy and bandwidth. Histogram fusion, also known as "typebased fusion" for distributed detection has been investigated in [13] and [14], where it is assumed that all sensors transmit the histogram (or type) of their own observations simultaneously through a multi-access channel, and the resultant error exponent is shown to be the same as centralized detection, irrespective of the transmission power. Although theoretically appealing, the implementation of such a scheme can be difficult, and lack of MAC synchronization or channel fading can significantly degrade the detection performance.…”

Decentralized Maximum-Likelihood Estimation for Sensor Networks Composed of Nonlinearly Coupled Dynamical Systems

“…Interestingly, this symmetry property is not at all artificial, as it reflects the data-centric nature of the network and it holds true in a variety of applications. Some works exploit the data-centric property to devise innovative schemes, like, for example, the type-based multiple access (TBMA) system [6]. TBMA is perfectly scalable, but it requires a high coherence of the channels from the sensors to the sink node.…”