The entorhinal cortex serves as a major gateway connecting the hippocampus and neocortex, playing a pivotal role in episodic memory formation. Neurons in the entorhinal cortex exhibit two notable features associated with temporal information processing: a population-level ability to encode long temporal signals and a single-cell characteristic known as graded-persistent activity, where some neurons maintain activity for extended periods even without external inputs. However, the relationship between these single-cell characteristics and population dynamics has remained unclear, largely due to the absence of a framework to describe the dynamics of neural populations with highly heterogeneous time scales. To address this gap, we extend the dynamical mean field theory, a powerful framework for analyzing large-scale population dynamics, to study the dynamics of heterogeneous neural populations. By proposing an analytically tractable model of graded-persistent activity, we demonstrate that the introduction of graded-persistent neurons shifts the chaos-order phase transition point and expands the network's dynamical region, a preferable region for temporal information computation. Furthermore, we validate our framework by applying it to a system with heterogeneous adaptation, demonstrating that such heterogeneity can reduce the dynamical regime, contrary to previous simplified approximations. These findings establish a theoretical foundation for understanding the functional advantages of diversity in biological systems and offer insights applicable to a wide range of heterogeneous networks beyond neural populations.

For the realization of an Internet of Things (IoT) with high densities of nodes it is necessary that wireless communication protocols are developed that offer (1) low energy consumption, (2) simplicity of encoding and decoding, (3) an asynchronous mode of communication, and (4) an effective but simple method to deal with interference between transmissions. This paper presents the implementation, experimentation, and analysis of a protocol on the MAC sublayer that is based on the encoding of information by the silent intervals between pulses. This encoding allows for few conflicts between messages that are broadcast on the same band overlapped in time. This is demonstrated experimentally, while no adverse effect is detected on transmissions from broadcasts of a different group of nodes using the same 315 MHz band but with a different code length as coding parameter. Theoretical analysis confirms that the probability of conflicts between messages is low, even if the number of nodes increases to the order of ten thousand. This protocol facilitates the implementation of IoT nodes that are restricted in terms of hardware and energy resources.

Impulse-radio wireless sensor network is a recently proposed concept of wireless sensor network aiming to sense an extended area with high spatial resolution. Extremely simple architecture of the network enables one to install huge number of sensors to the target area. Users of the network are able to precisely estimate location of event-detecting sensor nodes by measuring time of arrivals of impulse signals at anchor nodes. The localization protocol, however, assumes that sensor nodes keep their radio receiver on all the time, which is the main obstacle to reduce energy consumption of the network. To solve the problem, here, we propose a sleep/wake-up protocol of sensor nodes for the network. We show that the protocol can reduces activity of radio receiver of each node by more than 90 % with keeping precision of localization remain intact. We numerically confirm validity of the proposed sleep/wake-up protocol for various conditions including target area with various shape.

The Internet of Nano-Things will employ wireless sensor nodes that are not only extremely small in size and large in number but also severely restricted in their power supply. While a large part of the design of such networks will rely on the engineering of physical aspects, such as developing technology for efficient energy harvesting, sensing, and communication, a significant role will also be played by the algorithms underlying the operation of nodes to ensure that optimal use is made of the limited supply of energy. This paper discusses concepts that will facilitate the design of such algorithms, such as spike-based encoding of signals and data aggregation through gossiping. We propose a method to extract information from a sensor network using nonpacket spike-based communication with local interactions between nodes. Rather than gathering sensor values from each individual node and sending them to a central destination node, measurements from the nodes are processed through the interactions between nodes, while diffusing in the network, to obtain an integrated value characteristic for all nodes' measurements.

The connectivity structure of neural networks has significant implications for neural information processing, and much experimental effort has been made to clarify the structure of neural networks in the brain, i.e., both graph structure and weight structure of synaptic connections. A traditional view of neural information processing suggests that neurons compute in a highly parallel and distributed manner, in which the cooperation of many weak synaptic inputs is necessary to activate a single neuron. Recent experiments, however, have shown that not all synapses are weak in cortical circuits, but some synapses are extremely strong (several tens of times larger than the average weight). In fact, the weights of excitatory synapses between cortical excitatory neurons often obey a lognormal distribution with a long tail of strong synapses. Here, we review some of our important and recent works on computation with sparsely distributed synaptic weights and discuss the possible implications of this synaptic principle for neural computation by spiking neurons. We demonstrate that internal noise emerges from long-tailed distributions of synaptic weights to produce stochastic resonance effect in the reverberating synaptic pathways constituted by strong synapses. We show a spike-timing-dependent plasticity rule and other mechanisms that produce such weight distributions. A possible hardware realization of lognormally connected networks is also shown.