With much advancement in the field of nanotechnology, bioengineering and synthetic biology over the past decade, microscales and nanoscales devices are becoming a reality. Yet the problem of engineering a reliable communication system between tiny devices is still an open problem. At the same time, despite the prevalence of radio communication, there are still areas where traditional electromagnetic waves find it difficult or expensive to reach. Points of interest in industry, cities, and medical applications often lie in embedded and entrenched areas, accessible only by ventricles at scales too small for conventional radio waves and microwaves, or they are located in such a way that directional high frequency systems are ineffective. Inspired by nature, one solution to these problems is molecular communication (MC), where chemical signals are used to transfer information. Although biologists have studied MC for decades, it has only been researched for roughly 10 year from a communication engineering lens. Significant number of papers have been published to date, but owing to the need for interdisciplinary work, much of the results are preliminary. In this paper, the recent advancements in the field of MC engineering are highlighted. First, the biological, chemical, and physical processes used by an MC system are discussed. This includes different components of the MC transmitter and receiver, as well as the propagation and transport mechanisms. Then, a comprehensive survey of some of the recent works on MC through a communication engineering lens is provided. The paper ends with a technology readiness analysis of MC and future research directions.Comment: Accepted for publication in IEEE Communications Surveys & Tutorial

In this work, we describe the first modular, and programmable platform capable of transmitting a text message using chemical signalling – a method also known as molecular communication. This form of communication is attractive for applications where conventional wireless systems perform poorly, from nanotechnology to urban health monitoring. Using examples, we demonstrate the use of our platform as a testbed for molecular communication, and illustrate the features of these communication systems using experiments. By providing a simple and inexpensive means of performing experiments, our system fills an important gap in the molecular communication literature, where much current work is done in simulation with simplified system models. A key finding in this paper is that these systems are often nonlinear in practice, whereas current simulations and analysis often assume that the system is linear. However, as we show in this work, despite the nonlinearity, reliable communication is still possible. Furthermore, this work motivates future studies on more realistic modelling, analysis, and design of theoretical models and algorithms for these systems.

We consider the problem of joint source and channel coding of structured data such as natural language over a noisy channel. The typical approach to this problem in both theory and practice involves performing source coding to first compress the text and then channel coding to add robustness for the transmission across the channel. This approach is optimal in terms of minimizing end-to-end distortion with arbitrarily large block lengths of both the source and channel codes when transmission is over discrete memoryless channels. However, the optimality of this approach is no longer ensured for documents of finite length and limitations on the length of the encoding. We will show in this scenario that we can achieve lower word error rates by developing a deep learning based encoder and decoder. While the approach of separate source and channel coding would minimize bit error rates, our approach preserves semantic information of sentences by first embedding sentences in a semantic space where sentences closer in meaning are located closer together, and then performing joint source and channel coding on these embeddings.

In diffusion-based molecular communication, information transport is governed by diffusion through a fluid medium. The achievable data rates for these channels are very low compared to the radio-based communication system, since diffusion can be a slow process. To improve the data rate, a novel multiple-input multiple-output (MIMO) design for molecular communication is proposed that utilizes multiple molecular emitters at the transmitter and multiple molecular detectors at the receiver (in RF communication these all correspond to antennas). Using particle-based simulators, the channel's impulse response is obtained and mathematically modeled. These models are then used to determine inter-link interference (ILI) and inter-symbol interference (ISI). It is assumed that when the receiver has incomplete information regarding the system and the channel state, low complexity symbol detection methods are preferred since the receiver is small and simple. Thus four detection algorithms are proposed---adaptive thresholding, practical zero forcing with channel models excluding/including the ILI and ISI, and Genie-aided zero forcing. The proposed algorithms are evaluated extensively using numerical and analytical evaluations

We consider detection based on deep learning, and show it is possible to train detectors that perform well without any knowledge of the underlying channel models. Moreover, when the channel model is known, we demonstrate that it is possible to train detectors that do not require channel state information (CSI). In particular, a technique we call a sliding bidirectional recurrent neural network (SBRNN) is proposed for detection where, after training, the detector estimates the data in realtime as the signal stream arrives at the receiver. We evaluate this algorithm, as well as other neural network (NN) architectures, using the Poisson channel model, which is applicable to both optical and molecular communication systems. In addition, we also evaluate the performance of this detection method applied to data sent over a molecular communication platform, where the channel model is difficult to model analytically. We show that SBRNN is computationally efficient, and can perform detection under various channel conditions without knowing the underlying channel model. We also demonstrate that the bit error rate (BER) performance of the proposed SBRNN detector is better than that of a Viterbi detector with imperfect CSI as well as that of other NN detectors that have been previously proposed. Finally, we show that the SBRNN can perform well in rapidly changing channels, where the coherence time is on the order of a single symbol duration.